├── data ├── earl_agenda.xlsx └── tweets │ ├── output_got_earlconf.csv │ └── output_got_earl2017.csv ├── EARL_tweets_files └── figure-html │ ├── unnamed-chunk-2-1.png │ ├── unnamed-chunk-3-1.png │ ├── unnamed-chunk-4-1.png │ ├── unnamed-chunk-6-1.png │ ├── unnamed-chunk-11-1.png │ ├── unnamed-chunk-13-1.png │ └── unnamed-chunk-14-1.png ├── EARL_tweets.Rmd └── README.md /data/earl_agenda.xlsx: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MangoTheCat/blog_earl_tweets/master/data/earl_agenda.xlsx -------------------------------------------------------------------------------- /EARL_tweets_files/figure-html/unnamed-chunk-2-1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MangoTheCat/blog_earl_tweets/master/EARL_tweets_files/figure-html/unnamed-chunk-2-1.png -------------------------------------------------------------------------------- /EARL_tweets_files/figure-html/unnamed-chunk-3-1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MangoTheCat/blog_earl_tweets/master/EARL_tweets_files/figure-html/unnamed-chunk-3-1.png -------------------------------------------------------------------------------- /EARL_tweets_files/figure-html/unnamed-chunk-4-1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MangoTheCat/blog_earl_tweets/master/EARL_tweets_files/figure-html/unnamed-chunk-4-1.png -------------------------------------------------------------------------------- /EARL_tweets_files/figure-html/unnamed-chunk-6-1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MangoTheCat/blog_earl_tweets/master/EARL_tweets_files/figure-html/unnamed-chunk-6-1.png -------------------------------------------------------------------------------- /EARL_tweets_files/figure-html/unnamed-chunk-11-1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MangoTheCat/blog_earl_tweets/master/EARL_tweets_files/figure-html/unnamed-chunk-11-1.png -------------------------------------------------------------------------------- /EARL_tweets_files/figure-html/unnamed-chunk-13-1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MangoTheCat/blog_earl_tweets/master/EARL_tweets_files/figure-html/unnamed-chunk-13-1.png -------------------------------------------------------------------------------- /EARL_tweets_files/figure-html/unnamed-chunk-14-1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/MangoTheCat/blog_earl_tweets/master/EARL_tweets_files/figure-html/unnamed-chunk-14-1.png -------------------------------------------------------------------------------- /EARL_tweets.Rmd: -------------------------------------------------------------------------------- 1 | --- 2 | title: "Analysing EARL Tweets from EARL Tweeps" 3 | author: "Adnan Fiaz" 4 | output: 5 | html_document: 6 | keep_md: true 7 | --- 8 | 9 | ```{r setup, include=FALSE} 10 | knitr::opts_chunk$set(warning = FALSE, message=FALSE) 11 | library(tidyverse) 12 | library(purrr) 13 | library(scales) 14 | library(lubridate) 15 | library(stringdist) 16 | ``` 17 | 18 | With two out of three EARL conferences part of R history we're really excited about the next EARL conference in [Boston](https://earlconf.com/boston/) (only 1 week away!). This calls for an(other) EARL conference analysis, this time with Twitter data. Twitter is an amazingly rich data source and a great starting point for any data analysis (I feel there should be a awesome-twitter-blogposts list somewhere). 19 | 20 | I was planning on using the wonderful [rtweet](http://rtweet.info/) package by Michael Kearney (as advertised by [Bob Rudis](https://rud.is/b/2017/10/22/a-call-to-tweets-blog-posts/)) but unfortunately the Twitter API doesn't provide a full history of tweets. Instead I had to revert to a Python package (*gasp*) called [GetOldTweets](https://github.com/Jefferson-Henrique/GetOldTweets-python). I strongly recommend using the official Twitter API first before going down this path. 21 | 22 | ### The Data 23 | 24 | ```{r} 25 | # I have used the Exporter script with the hashtags #EARLConf2017, #EARLConf and #EARL2017 26 | tweets_df <- purrr::map_df(list.files('data/tweets', full.names = TRUE), 27 | ~ readr::read_delim(.x, delim=";", quote="")) %>% 28 | # filter out company accounts 29 | filter(username!="earlconf", username!="MangoTheCat") %>% 30 | mutate(shorttext = stringr::str_sub(text, end=50)) 31 | 32 | tweets_df %>% 33 | select(username, date, shorttext) %>% 34 | head() %>% 35 | knitr::kable() 36 | ``` 37 | 38 | First things first, let's get a timeline up: 39 | 40 | ```{r, echo=FALSE} 41 | tweets_df %>% 42 | mutate(day = lubridate::date(date)) %>% 43 | count(day) %>% 44 | ggplot(aes(x=day, y=n)) + 45 | geom_line() + 46 | labs(x='Date', y='Number of tweets', title='Number of EARL-related tweets by day') + 47 | scale_x_date(date_breaks="2 months", labels=date_format('%b-%y')) + 48 | theme_classic() 49 | ``` 50 | 51 | The hashtags I used to search tweets were generic so the results include tweets from last year's conferences. Let's zoom in on this year's conferences: EARL San Francisco (5-7 June) and EARL London (12-14 September). They clearly explain the large peaks in the above graph. 52 | 53 | ```{r, echo=FALSE} 54 | tweets_df %>% 55 | filter(date > "2017-05-01") %>% 56 | mutate(day = lubridate::date(date)) %>% 57 | count(day) %>% 58 | ggplot(aes(x=day, y=n)) + 59 | geom_line(size=1.1) + 60 | annotate("rect",xmin=lubridate::date("2017-06-05"), 61 | xmax=lubridate::date("2017-06-08"), 62 | ymin=0, ymax=Inf, alpha=.6, fill="orange") + 63 | annotate("rect",xmin=lubridate::date("2017-09-12"), 64 | xmax=lubridate::date("2017-09-15"), 65 | ymin=0, ymax=Inf, alpha=.5, fill="blue") + 66 | labs(x='Date', y='Number of tweets', title='Number of EARL-related tweets by day') + 67 | scale_x_date(date_breaks="1 months", labels=date_format('%b-%y')) + 68 | theme_classic() 69 | 70 | ``` 71 | 72 | I've tried to highlight the period when the conferences were on but I don't quite like the result. Let's see if it works better with a bar chart. 73 | 74 | ```{r} 75 | earlconf_sf_dates <- lubridate::interval("2017-06-05", "2017-06-08") 76 | earlconf_lon_dates <- lubridate::interval("2017-09-12", "2017-09-15") 77 | tweets_df %>% 78 | filter(date > "2017-05-01") %>% 79 | mutate(day = lubridate::date(date)) %>% 80 | count(day) %>% 81 | mutate(conference = case_when(day %within% earlconf_sf_dates ~ "SF", 82 | day %within% earlconf_lon_dates ~ "LON", 83 | TRUE ~ "NONE")) %>% 84 | ggplot(aes(x=day, y=n)) + 85 | geom_bar(stat="identity", aes(fill=conference)) + 86 | scale_fill_manual(guide=FALSE, values=c("#F8766D","black","#619CFF")) + 87 | labs(x='Date', y='Number of tweets', title='Number of EARL-related tweets by day') + 88 | scale_x_date(date_breaks="1 months", labels=date_format('%b-%y')) + 89 | theme_classic() 90 | 91 | ``` 92 | 93 | Now that's a lot better. The tweet counts in black surrounding the conferences look like small buildings which make the conference tweet counts look like giant skyscrapers (I was a failed art critic in a previous life). 94 | 95 | ### Activity during conferences 96 | 97 | I've been to my fair share of conferences/presentations and I've always wondered how people tweet so fast during a talk. It could be just my ancient phone or I may lack the necessary skills. Either way it would be interesting to analyse the tweets at the talk level. First I will need to link the tweets to a specific talks. I've translated the published agenda into a nicer format by hand and read it in below. 98 | 99 | ```{r} 100 | earl_agenda <- map_df(c("EARL_SF", "EARL_LON"), 101 | ~ readxl::read_xlsx('data/earl_agenda.xlsx', sheet = .x) ) 102 | earl_agenda %>% 103 | select(StartTime, EndTime, Title, Presenter) %>% 104 | head() %>% 105 | knitr::kable() 106 | ``` 107 | 108 | Before I merge the tweets with the agenda it's a good idea to zoom in on the conference tweets (who doesn't like a facetted plot). 109 | 110 | ```{r} 111 | conference_tweets <- tweets_df %>% 112 | mutate(conference = case_when(date %within% earlconf_sf_dates ~ "SF", 113 | date %within% earlconf_lon_dates ~ "LON", 114 | TRUE ~ "NONE")) %>% 115 | filter(conference != "NONE") 116 | 117 | ggplot(conference_tweets, aes(x=date)) + 118 | geom_histogram() + 119 | facet_wrap(~ conference, scales = 'free_x') 120 | ``` 121 | 122 | Nothing odd in the pattern of tweets: there are no talks on the first day so barely any tweets; the amount of tweets spikes at the beginning of the other two days and then declines as the day progresses. There is something odd about the timing of the tweets though. I didn't notice it before but when I compared the position of the bars on the x-axis the San Francisco tweets look shifted. And then my lack of travel experience hit me: time zones! The tweets were recorded in UTC time but the talks obviously weren't in the evening in San Francisco. 123 | 124 | After correcting for time zones I can finally merge the tweets with the agenda. 125 | 126 | ```{r} 127 | selection <- conference_tweets$conference=='SF' 128 | conference_tweets[selection, 'date'] <- conference_tweets[selection, 'date'] - 8*60*60 129 | ``` 130 | 131 | ```{r} 132 | # I intended to use a fuzzy join here and check if the tweet timestamp falls within the [start, end) of a talk 133 | # unfortunately I couldn't get it to work with datetime objects 134 | # so I resort to determining the cartesian product and simply filtering the relevant records 135 | tweets_and_talks <- conference_tweets %>% 136 | mutate(dummy = 1) %>% 137 | left_join(earl_agenda %>% mutate(dummy=1)) %>% 138 | filter(date >= StartTime, date < EndTime) 139 | 140 | tweets_and_talks %>% 141 | select(username, date, shorttext, Title, Presenter) %>% 142 | tail() %>% 143 | knitr::kable() 144 | ``` 145 | 146 | You ever have that feeling that you're forgetting something and then you're at the airport without your passport? From the above table it's obvious I've forgotten that talks are organised in parallel. So matching on time only will create duplicates. However, you may notice that some tweets also mention the presenter (that is considered good tweetiquette). We can use that information to further improve the matching. 147 | 148 | ```{r} 149 | talks_and_tweets <- tweets_and_talks %>% 150 | # calculate various scores based on what is said in the tweet text 151 | mutate(presenter_score = ifelse(!is.na(mentions) & !is.na(TwitterHandle), stringr::str_detect(mentions, TwitterHandle), 0), 152 | # check if the presenter's name is mentioned 153 | presenter_score2 = stringr::str_detect(text, Presenter), 154 | # check if the company name is mentioned 155 | company_score = stringr::str_detect(text, Company), 156 | # check if what is mentioned has any overlap with the title (description would've been better) 157 | overall_score = stringsim(text, Title), 158 | # sum all the scores 159 | score = overall_score + presenter_score + presenter_score2 + company_score) %>% 160 | select(-presenter_score, -presenter_score2, -company_score, -overall_score) %>% 161 | # now select the highest scoring match 162 | group_by(username, date) %>% 163 | top_n(1, score) %>% 164 | ungroup() 165 | 166 | talks_and_tweets %>% 167 | select(username, date, shorttext, Title, Presenter) %>% 168 | tail() %>% 169 | knitr::kable() 170 | ``` 171 | 172 | That looks better but I am disappointed at the number of tweets (`r nrow(talks_and_tweets)`) during talks. Maybe attendees are too busy listening to the talk instead of tweeting which is a good thing I suppose. Nevertheless I can still try to create some interesting visualisations with this data. 173 | 174 | ```{r} 175 | tweets_by_presenter <- talks_and_tweets %>% 176 | count(conference, Title, Presenter) %>% 177 | ungroup() %>% 178 | arrange(conference, n) 179 | 180 | tweets_by_presenter$Presenter <- factor(tweets_by_presenter$Presenter, levels=tweets_by_presenter$Presenter) 181 | ``` 182 | ```{r, echo=FALSE} 183 | tweets_by_presenter %>% 184 | ggplot(aes(x=Presenter, y=n)) + 185 | geom_bar(stat="identity") + 186 | coord_flip() + 187 | facet_wrap(~ conference, scales="free") + 188 | labs(title="Number of tweets by presenter", y="Number of tweets") 189 | ``` 190 | 191 | 192 | The visualisation doesn't really work for the large number of presenters although I don't really see another way to add the information about a talk. I also tried to sort the levels of the factor so they appear sorted in the plot but for some reason the _SF_ facet doesn't want to cooperate. There are a number of talks vying for the top spot in San Francisco but the differences aren't that large. I'm of course assuming my matching heuristic worked perfectly but one or two mismatches and the results could look completely different. The same applies to EARL London but here Joe Cheng clearly takes the crown. 193 | 194 | ### Follow the leader... 195 | 196 | Let's go down a somewhat more creepy road and see what talks people go to. 197 | 198 | ```{r} 199 | tweeters <- talks_and_tweets %>% 200 | group_by(username) %>% 201 | mutate(num_tweets = n()) %>% 202 | ungroup() %>% 203 | filter(num_tweets > 4) %>% 204 | mutate(day = ifelse(conference == "SF", (Session > 6)+1, (Session > 9)+1), 205 | day = ifelse(day==1, "Day 1", "Day 2")) %>% 206 | select(username, conference, StartTime, Stream, day) 207 | ``` 208 | ```{r, echo=FALSE} 209 | tweeters %>% 210 | filter(conference=="LON") %>% 211 | ggplot(aes(x=StartTime, y=Stream, colour=username, group=username)) + 212 | geom_path() + 213 | geom_point() + 214 | guides(colour=FALSE) + 215 | facet_grid(~day, scales="free_x") + 216 | theme_bw() + 217 | labs(title="Talks visited by attendees at EARL London", x="Time") + 218 | scale_y_continuous(labels=c("", "1", "", "2", "", "3", ""), breaks=seq(0.5, 3.5, 0.5), 219 | limits = c(0.5, 3.5)) + 220 | theme(panel.grid.major.y = element_blank(), panel.grid.minor.y = element_blank()) 221 | ``` 222 | ```{r, echo=FALSE} 223 | tweeters %>% 224 | filter(conference=="SF") %>% 225 | ggplot(aes(x=StartTime, y=Stream, colour=username, group=username)) + 226 | geom_path() + 227 | geom_point() + 228 | guides(colour=FALSE) + 229 | facet_grid(~day, scales="free_x") + 230 | theme_bw() + 231 | labs(title="Talks visited by attendees at EARL San Francisco", x="Time") + 232 | scale_y_continuous(labels = c("", "1", "", "2", ""), breaks=seq(0.5, 2.5, 0.5), 233 | limits=c(0.5, 2.5)) + 234 | theme(panel.grid.major.y = element_blank(), panel.grid.minor.y = element_blank()) 235 | ``` 236 | 237 | Each line is a twitter user (twitterer? tweeter? tweep?) and each observation represents a tweet during a presentation. My expectation was that by drawing a line between the observations you could see how people switch (or don't switch) between talks. That has clearly failed as the tweeting behaviour isn't consistent or numerous enough to actually see that. I'm quite glad it's not possible since tracking people isn't what Twitter is for. 238 | 239 | The code and data for this blogpost are available on [GitHub](https://github.com/MangoTheCat/blog_earl_tweets) so feel free to play around with it yourself. Do let us know if you create any awesome visualisations or if we can improve on any of the above. If you also want to tweet at conferences, EARL Boston is happening on 1-3 November and [tickets](https://ti.to/earl-conference/boston-2017) are still available. I promise we won't track you! 240 | 241 | 242 | 243 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Analysing EARL Tweets from EARL Tweeps 2 | Adnan Fiaz 3 | 4 | 5 | 6 | With two out of three EARL conferences part of R history we're really excited about the next EARL conference in [Boston](https://earlconf.com/boston/) (only 1 week away!). This calls for an(other) EARL conference analysis, this time with Twitter data. Twitter is an amazingly rich data source and a great starting point for any data analysis (I feel there should be a awesome-twitter-blogposts list somewhere). 7 | 8 | I was planning on using the wonderful [rtweet](http://rtweet.info/) package by Michael Kearney (as advertised by [Bob Rudis](https://rud.is/b/2017/10/22/a-call-to-tweets-blog-posts/)) but unfortunately the Twitter API doesn't provide a full history of tweets. Instead I had to revert to a Python package (*gasp*) called [GetOldTweets](https://github.com/Jefferson-Henrique/GetOldTweets-python). I strongly recommend using the official Twitter API first before going down this path. 9 | 10 | ### The Data 11 | 12 | 13 | ```r 14 | # I have used the Exporter script with the hashtags #EARLConf2017, #EARLConf and #EARL2017 15 | tweets_df <- purrr::map_df(list.files('data/tweets', full.names = TRUE), 16 | ~ readr::read_delim(.x, delim=";", quote="")) %>% 17 | # filter out company accounts 18 | filter(username!="earlconf", username!="MangoTheCat") %>% 19 | mutate(shorttext = stringr::str_sub(text, end=50)) 20 | 21 | tweets_df %>% 22 | select(username, date, shorttext) %>% 23 | head() %>% 24 | knitr::kable() 25 | ``` 26 | 27 | 28 | 29 | username date shorttext 30 | ------------- -------------------- --------------------------------------------------- 31 | AlanHoKT 2017-10-02 02:15:00 ". @TIBCO 's @LouBajuk spoke at #EARL2017 London o 32 | johnon2 2017-09-23 16:02:00 ". @TIBCO 's @LouBajuk spoke at #EARL2017 London o 33 | AndySugs 2017-09-21 22:19:00 "RT: LearnRinaDay: EARL London 2017 – That’s a wra 34 | LearnRinaDay 2017-09-21 22:17:00 "EARL London 2017 – That’s a wrap! https://www. r- 35 | LouBajuk 2017-09-20 23:15:00 ". @TIBCO 's @LouBajuk spoke at #EARL2017 London o 36 | pjevrard 2017-09-20 13:02:00 ". @TIBCO 's @LouBajuk spoke at #EARL2017 London o 37 | 38 | First things first, let's get a timeline up: 39 | 40 | ![](EARL_tweets_files/figure-html/unnamed-chunk-2-1.png) 41 | 42 | The hashtags I used to search tweets were generic so the results include tweets from last year's conferences. Let's zoom in on this year's conferences: EARL San Francisco (5-7 June) and EARL London (12-14 September). They clearly explain the large peaks in the above graph. 43 | 44 | ![](EARL_tweets_files/figure-html/unnamed-chunk-3-1.png) 45 | 46 | I've tried to highlight the period when the conferences were on but I don't quite like the result. Let's see if it works better with a bar chart. 47 | 48 | 49 | ```r 50 | earlconf_sf_dates <- lubridate::interval("2017-06-05", "2017-06-08") 51 | earlconf_lon_dates <- lubridate::interval("2017-09-12", "2017-09-15") 52 | tweets_df %>% 53 | filter(date > "2017-05-01") %>% 54 | mutate(day = lubridate::date(date)) %>% 55 | count(day) %>% 56 | mutate(conference = case_when(day %within% earlconf_sf_dates ~ "SF", 57 | day %within% earlconf_lon_dates ~ "LON", 58 | TRUE ~ "NONE")) %>% 59 | ggplot(aes(x=day, y=n)) + 60 | geom_bar(stat="identity", aes(fill=conference)) + 61 | scale_fill_manual(guide=FALSE, values=c("#F8766D","black","#619CFF")) + 62 | labs(x='Date', y='Number of tweets', title='Number of EARL-related tweets by day') + 63 | scale_x_date(date_breaks="1 months", labels=date_format('%b-%y')) + 64 | theme_classic() 65 | ``` 66 | 67 | ![](EARL_tweets_files/figure-html/unnamed-chunk-4-1.png) 68 | 69 | Now that's a lot better. The tweet counts in black surrounding the conferences look like small buildings which make the conference tweet counts look like giant skyscrapers (I was a failed art critic in a previous life). 70 | 71 | ### Activity during conferences 72 | 73 | I've been to my fair share of conferences/presentations and I've always wondered how people tweet so fast during a talk. It could be just my ancient phone or I may lack the necessary skills. Either way it would be interesting to analyse the tweets at the talk level. First I will need to link the tweets to a specific talks. I've translated the published agenda into a nicer format by hand and read it in below. 74 | 75 | 76 | ```r 77 | earl_agenda <- map_df(c("EARL_SF", "EARL_LON"), 78 | ~ readxl::read_xlsx('data/earl_agenda.xlsx', sheet = .x) ) 79 | earl_agenda %>% 80 | select(StartTime, EndTime, Title, Presenter) %>% 81 | head() %>% 82 | knitr::kable() 83 | ``` 84 | 85 | 86 | 87 | StartTime EndTime Title Presenter 88 | -------------------- -------------------- -------------------------------------------------------------------------------------------------- -------------------- 89 | 2017-06-06 11:00:00 2017-06-06 11:30:00 R’s role in Data Science Joe Cheng 90 | 2017-06-06 11:30:00 2017-06-06 12:00:00 ‘Full Stack’ Data Science with R: production data science and engineering with open source tools Gabriela de Queiroz 91 | 2017-06-06 12:00:00 2017-06-06 12:30:00 R Operating Model Mark Sellors 92 | 2017-06-06 11:00:00 2017-06-06 11:30:00 Large-scale reproducible simulation pipelines in R using Docker Mike Gahan 93 | 2017-06-06 11:30:00 2017-06-06 12:00:00 Using data to identify risky prescribing habits in physicians Aaron Hamming 94 | 2017-06-06 12:00:00 2017-06-06 12:30:00 How we built a Shiny App for 700 users Filip Stachura 95 | 96 | Before I merge the tweets with the agenda it's a good idea to zoom in on the conference tweets (who doesn't like a facetted plot). 97 | 98 | 99 | ```r 100 | conference_tweets <- tweets_df %>% 101 | mutate(conference = case_when(date %within% earlconf_sf_dates ~ "SF", 102 | date %within% earlconf_lon_dates ~ "LON", 103 | TRUE ~ "NONE")) %>% 104 | filter(conference != "NONE") 105 | 106 | ggplot(conference_tweets, aes(x=date)) + 107 | geom_histogram() + 108 | facet_wrap(~ conference, scales = 'free_x') 109 | ``` 110 | 111 | ![](EARL_tweets_files/figure-html/unnamed-chunk-6-1.png) 112 | 113 | Nothing odd in the pattern of tweets: there are no talks on the first day so barely any tweets; the amount of tweets spikes at the beginning of the other two days and then declines as the day progresses. There is something odd about the timing of the tweets though. I didn't notice it before but when I compared the position of the bars on the x-axis the San Francisco tweets look shifted. And then my lack of travel experience hit me: time zones! The tweets were recorded in UTC time but the talks obviously weren't in the evening in San Francisco. 114 | 115 | After correcting for time zones I can finally merge the tweets with the agenda. 116 | 117 | 118 | ```r 119 | selection <- conference_tweets$conference=='SF' 120 | conference_tweets[selection, 'date'] <- conference_tweets[selection, 'date'] - 8*60*60 121 | ``` 122 | 123 | 124 | ```r 125 | # I intended to use a fuzzy join here and check if the tweet timestamp falls within the [start, end) of a talk 126 | # unfortunately I couldn't get it to work with datetime objects 127 | # so I resort to determining the cartesian product and simply filtering the relevant records 128 | tweets_and_talks <- conference_tweets %>% 129 | mutate(dummy = 1) %>% 130 | left_join(earl_agenda %>% mutate(dummy=1)) %>% 131 | filter(date >= StartTime, date < EndTime) 132 | 133 | tweets_and_talks %>% 134 | select(username, date, shorttext, Title, Presenter) %>% 135 | tail() %>% 136 | knitr::kable() 137 | ``` 138 | 139 | 140 | 141 | username date shorttext Title Presenter 142 | -------------- -------------------- --------------------------------------------------- ---------------------------------------------------------------- ----------- 143 | hspter 2017-06-06 11:17:00 "Nice shout out to @rOpenSci as prodigious package R’s role in Data Science Joe Cheng 144 | hspter 2017-06-06 11:17:00 "Nice shout out to @rOpenSci as prodigious package Large-scale reproducible simulation pipelines in R using Docker Mike Gahan 145 | RLadiesGlobal 2017-06-06 11:14:00 "#RLadies @b23kelly talking about #rstats at #EARL R’s role in Data Science Joe Cheng 146 | RLadiesGlobal 2017-06-06 11:14:00 "#RLadies @b23kelly talking about #rstats at #EARL Large-scale reproducible simulation pipelines in R using Docker Mike Gahan 147 | hspter 2017-06-06 11:14:00 "I'm digging the postmodern data scientist from @R R’s role in Data Science Joe Cheng 148 | hspter 2017-06-06 11:14:00 "I'm digging the postmodern data scientist from @R Large-scale reproducible simulation pipelines in R using Docker Mike Gahan 149 | 150 | You ever have that feeling that you're forgetting something and then you're at the airport without your passport? From the above table it's obvious I've forgotten that talks are organised in parallel. So matching on time only will create duplicates. However, you may notice that some tweets also mention the presenter (that is considered good tweetiquette). We can use that information to further improve the matching. 151 | 152 | 153 | ```r 154 | talks_and_tweets <- tweets_and_talks %>% 155 | # calculate various scores based on what is said in the tweet text 156 | mutate(presenter_score = ifelse(!is.na(mentions) & !is.na(TwitterHandle), stringr::str_detect(mentions, TwitterHandle), 0), 157 | # check if the presenter's name is mentioned 158 | presenter_score2 = stringr::str_detect(text, Presenter), 159 | # check if the company name is mentioned 160 | company_score = stringr::str_detect(text, Company), 161 | # check if what is mentioned has any overlap with the title (description would've been better) 162 | overall_score = stringsim(text, Title), 163 | # sum all the scores 164 | score = overall_score + presenter_score + presenter_score2 + company_score) %>% 165 | select(-presenter_score, -presenter_score2, -company_score, -overall_score) %>% 166 | # now select the highest scoring match 167 | group_by(username, date) %>% 168 | top_n(1, score) %>% 169 | ungroup() 170 | 171 | talks_and_tweets %>% 172 | select(username, date, shorttext, Title, Presenter) %>% 173 | tail() %>% 174 | knitr::kable() 175 | ``` 176 | 177 | 178 | 179 | username date shorttext Title Presenter 180 | -------------- -------------------- --------------------------------------------------- -------------------------------------------------------------------------------------------------- -------------------- 181 | Madhuraraju 2017-06-06 11:39:00 "@aj2z @gdequeiroz from @SelfScore talking about u ‘Full Stack’ Data Science with R: production data science and engineering with open source tools Gabriela de Queiroz 182 | hspter 2017-06-06 11:22:00 "#rstats is great for achieving "flow" while doing R’s role in Data Science Joe Cheng 183 | RLadiesGlobal 2017-06-06 11:20:00 "@RStudioJoe showing the #RLadies logo and a big m R’s role in Data Science Joe Cheng 184 | hspter 2017-06-06 11:17:00 "Nice shout out to @rOpenSci as prodigious package Large-scale reproducible simulation pipelines in R using Docker Mike Gahan 185 | RLadiesGlobal 2017-06-06 11:14:00 "#RLadies @b23kelly talking about #rstats at #EARL Large-scale reproducible simulation pipelines in R using Docker Mike Gahan 186 | hspter 2017-06-06 11:14:00 "I'm digging the postmodern data scientist from @R R’s role in Data Science Joe Cheng 187 | 188 | That looks better but I am disappointed at the number of tweets (263) during talks. Maybe attendees are too busy listening to the talk instead of tweeting which is a good thing I suppose. Nevertheless I can still try to create some interesting visualisations with this data. 189 | 190 | 191 | ```r 192 | tweets_by_presenter <- talks_and_tweets %>% 193 | count(conference, Title, Presenter) %>% 194 | ungroup() %>% 195 | arrange(conference, n) 196 | 197 | tweets_by_presenter$Presenter <- factor(tweets_by_presenter$Presenter, levels=tweets_by_presenter$Presenter) 198 | ``` 199 | ![](EARL_tweets_files/figure-html/unnamed-chunk-11-1.png) 200 | 201 | 202 | The visualisation doesn't really work for the large number of presenters although I don't really see another way to add the information about a talk. I also tried to sort the levels of the factor so they appear sorted in the plot but for some reason the _SF_ facet doesn't want to cooperate. There are a number of talks vying for the top spot in San Francisco but the differences aren't that large. I'm of course assuming my matching heuristic worked perfectly but one or two mismatches and the results could look completely different. The same applies to EARL London but here Joe Cheng clearly takes the crown. 203 | 204 | ### Follow the leader... 205 | 206 | Let's go down a somewhat more creepy road and see what talks people go to. 207 | 208 | 209 | ```r 210 | tweeters <- talks_and_tweets %>% 211 | group_by(username) %>% 212 | mutate(num_tweets = n()) %>% 213 | ungroup() %>% 214 | filter(num_tweets > 4) %>% 215 | mutate(day = ifelse(conference == "SF", (Session > 6)+1, (Session > 9)+1), 216 | day = ifelse(day==1, "Day 1", "Day 2")) %>% 217 | select(username, conference, StartTime, Stream, day) 218 | ``` 219 | ![](EARL_tweets_files/figure-html/unnamed-chunk-13-1.png) 220 | ![](EARL_tweets_files/figure-html/unnamed-chunk-14-1.png) 221 | 222 | Each line is a twitter user (twitterer? tweeter? tweep?) and each observation represents a tweet during a presentation. My expectation was that by drawing a line between the observations you could see how people switch (or don't switch) between talks. That has clearly failed as the tweeting behaviour isn't consistent or numerous enough to actually see that. I'm quite glad it's not possible since tracking people isn't what Twitter is for. 223 | 224 | The code and data for this blogpost are available on [GitHub](https://github.com/MangoTheCat/blog_earl_tweets) so feel free to play around with it yourself. Do let us know if you create any awesome visualisations or if we can improve on any of the above. If you also want to tweet at conferences, EARL Boston is happening on 1-3 November and [tickets](https://ti.to/earl-conference/boston-2017) are still available. I promise we won't track you! 225 | 226 | 227 | 228 | -------------------------------------------------------------------------------- /data/tweets/output_got_earlconf.csv: -------------------------------------------------------------------------------- 1 | username;date;retweets;favorites;text;geo;mentions;hashtags;id;permalink 2 | nibrivia;2017-10-16 03:46;0;0;"Go go go! (Although maybe I shouldn't encourage my competition too much...) #EARLConf https:// twitter.com/earlconf/statu s/919510170165760001 …";;;#EARLConf;"919756429447385088";https://twitter.com/nibrivia/status/919756429447385088 3 | kosinski_rblog;2017-10-16 00:23;1;7;"#EARLConf I organized Polish R Conf for 200 ppl last mnth! http:// fb.me/whyRconf Show me how you organize a conf so we can exchange ideas https:// twitter.com/earlconf/statu s/919510170165760001 …";;;#EARLConf;"919705248796696576";https://twitter.com/kosinski_rblog/status/919705248796696576 4 | dmi3k;2017-10-15 14:28;0;1;"To live-tweet? #EARLconf Flight from would be too expensive for me, though";;;#EARLconf;"919555470549049346";https://twitter.com/dmi3k/status/919555470549049346 5 | earlconf;2017-10-15 11:28;0;2;"Want to come to EARL Boston for FREE? Just tweet to tell us why you should come - include #EARLConf . Closes today! pic.twitter.com/qzhpEyUbao";;;#EARLConf;"919510170165760001";https://twitter.com/earlconf/status/919510170165760001 6 | earlconf;2017-10-14 20:27;2;1;"Why do you want to come to EARL Boston? Tweet us why and you could win a 2-day pass. Make sure to use #EARLConf to enter!";;;#EARLConf;"919283418533265408";https://twitter.com/earlconf/status/919283418533265408 7 | MangoTheCat;2017-10-14 17:15;0;0;"EARL Boston must-see sessions and 15% off!: https:// goo.gl/ir97eV #EARLConf #rstats pic.twitter.com/TvhxduJtEg";;;#EARLConf #rstats;"919235109764698113";https://twitter.com/MangoTheCat/status/919235109764698113 8 | ChristineIyer;2017-10-14 17:10;0;0;"I hope I win the pass...I wonder what the odds are? #EARLConf https:// twitter.com/christineiyer/ status/918423709458759680 …";;;#EARLConf;"919234029404278784";https://twitter.com/ChristineIyer/status/919234029404278784 9 | MustardBethan;2017-10-13 16:41;1;2;"AND because I got so excited at the prospect of it I forgot to include the #EARLconf hashtag in my reply! #numpty";;;#EARLconf #numpty;"918864162574385154";https://twitter.com/MustardBethan/status/918864162574385154 10 | earlconf;2017-10-13 14:05;0;0;"Winning a 2-day EARL Boston pass couldn’t be simpler. Just tweet us why you should come, be sure to use #EARLConf to enter pic.twitter.com/S4fVo25YcW";;;#EARLConf;"918824905759604736";https://twitter.com/earlconf/status/918824905759604736 11 | earlconf;2017-10-12 20:43;0;6;"We are delighted to have @MariannaFoos keynoting at EARL Boston on behalf of @RLadiesBoston http:// bit.ly/2d76nho #EARLConf pic.twitter.com/3dYBOpO4vK";;@MariannaFoos @RLadiesBoston;#EARLConf;"918562677273882626";https://twitter.com/earlconf/status/918562677273882626 12 | earlconf;2017-10-12 11:56;7;12;"Join @frlazzeri to hear her talk on one of the biggest trends in #MachineLearning – deep learning: https:// goo.gl/TA69Ju #EARLConf pic.twitter.com/i7S8OUw6gA";;@frlazzeri;#MachineLearning #EARLConf;"918430172097007616";https://twitter.com/earlconf/status/918430172097007616 13 | earlconf;2017-10-11 16:25;3;0;"Win a 2-day Boston EARL pass. Tweet us why you should go – best answer wins. Use #EARLConf to enter. pic.twitter.com/4Op3xnlbax";;;#EARLConf;"918135364296028161";https://twitter.com/earlconf/status/918135364296028161 14 | MangoTheCat;2017-10-11 11:15;0;0;"EARL Boston must-see sessions and 15% off!: https:// goo.gl/ir97eV #EARLConf #rstats pic.twitter.com/OyXuyRBhLC";;;#EARLConf #rstats;"918057345766313984";https://twitter.com/MangoTheCat/status/918057345766313984 15 | RLadiesBoston;2017-10-11 00:28;9;9;"Just announced a pre #EARLconf meetup for 10/31 I want to catch all the R-Ladies in town for the conference!";;;#EARLconf;"917894749780144133";https://twitter.com/RLadiesBoston/status/917894749780144133 16 | EmilyRiederer;2017-10-10 17:36;1;3;"Don't miss a chance to win free tickets to #EARLConf Boston or MANGO15PC for a 15% discount! https:// twitter.com/earlconf/statu s/917727676034703360 …";;;#EARLConf;"917790950927237122";https://twitter.com/EmilyRiederer/status/917790950927237122 17 | earlconf;2017-10-10 16:38;2;1;"EARL Boston must-see sessions and 15% off!: https:// goo.gl/ir97eV #EARLConf #rstats pic.twitter.com/oMS0wsEyt9";;;#EARLConf #rstats;"917776405735362562";https://twitter.com/earlconf/status/917776405735362562 18 | earlconf;2017-10-10 13:25;2;3;"Why do you want to come to Boston EARL? Tweet us why and you could win a 2-day pass. Make sure to use #EARLConf to enter! pic.twitter.com/yDZ1RfRpxI";;;#EARLConf;"917727676034703360";https://twitter.com/earlconf/status/917727676034703360 19 | earlconf;2017-10-09 17:18;7;2;"Want to come to Boston EARL for FREE? Just tweet to tell us why you should come - include #EARLConf . Closes 15 Oct pic.twitter.com/0XPwaPQvvG";;;#EARLConf;"917424152860127232";https://twitter.com/earlconf/status/917424152860127232 20 | AlFirrell;2017-09-14 16:32;1;1;"Oooh, @jcheng introduces Async programming for Shiny. Happy days! #rstats #EARLConf";;@jcheng;#rstats #EARLConf;"908352685056946176";https://twitter.com/AlFirrell/status/908352685056946176 21 | JennyBryan;2017-09-14 16:31;2;31;"I really love that @jcheng wears his Shiny shirt for talks. Hearing about async programming at #EARLConf #rstats pic.twitter.com/x4yosdB2oB";;@jcheng;#EARLConf #rstats;"908352457448804352";https://twitter.com/JennyBryan/status/908352457448804352 22 | dougashton;2017-09-14 14:08;1;2;""An updated dataset is not a version. It's a new data set" - @revodavid on should you version control data #earlconf";;@revodavid;#earlconf;"908316624087928832";https://twitter.com/dougashton/status/908316624087928832 23 | m_a_upson;2017-09-14 13:45;2;4;"Great tips for #reproducibility from @revodavid @earlconf #EARLConf pic.twitter.com/UPr35dWs6l";;@revodavid @earlconf;#reproducibility #EARLConf;"908310743711350784";https://twitter.com/m_a_upson/status/908310743711350784 24 | JennyBryan;2017-09-14 10:00;2;3;"Does your business benefit from #rstats ? Check out the @RConsortium ! @LouBajuk explains the importance of building the ecosystem. #EARLConf";;@RConsortium @LouBajuk;#rstats #EARLConf;"908254091104702465";https://twitter.com/JennyBryan/status/908254091104702465 25 | dougashton;2017-09-14 09:21;1;3;"I will be stealing the term "technical artifact" from @hspter for a reproducible analysis output #earlconf";;@hspter;#earlconf;"908244307815084032";https://twitter.com/dougashton/status/908244307815084032 26 | dougashton;2017-09-13 18:02;1;4;"Hope it stays dry for the boat! #earlconf pic.twitter.com/doIedWMkEX";;;#earlconf;"908013091014791170";https://twitter.com/dougashton/status/908013091014791170 27 | AlFirrell;2017-09-13 17:30;2;6;"An experiment: 2 @MangoTheCat s, 2 daughters. Will they a. happily choose one each, b. fight for the "best", or c. not care? #EARLConf pic.twitter.com/RI9fCB4U4I";;@MangoTheCat;#EARLConf;"908005026303488000";https://twitter.com/AlFirrell/status/908005026303488000 28 | hfcfrick;2017-09-13 17:10;1;3;"@mjaeugster taking the literal #datavis route at #EARLConf pic.twitter.com/8xFsJmKRrW";;@mjaeugster;#datavis #EARLConf;"907999816734724096";https://twitter.com/hfcfrick/status/907999816734724096 29 | dougashton;2017-09-13 16:16;2;3;"The R class wars seem to be lining up behind R6 (although I still love S3 #EARLConf2017 #earlconf pic.twitter.com/ZLCZpXlHQe";;;#EARLConf2017 #earlconf;"907986408018771968";https://twitter.com/dougashton/status/907986408018771968 30 | sellorm;2017-09-13 15:50;2;4;"Really nice to see some real world advice on best practice for deploying shiny apps #earlconf pic.twitter.com/Lfp9at67Jb";;;#earlconf;"907979922278363136";https://twitter.com/sellorm/status/907979922278363136 31 | dougashton;2017-09-13 15:47;2;6;"An early theme from #EARLConf is that everyone uses @gitlab to store #rstats code. pic.twitter.com/RsjlH84lzF";;@gitlab;#EARLConf #rstats;"907978963439570945";https://twitter.com/dougashton/status/907978963439570945 32 | sellorm;2017-09-13 15:40;1;3;"Grace Meyer talks about being a victim of your own success with #rstats and #rshiny - #earlconf pic.twitter.com/nadBwepfRU";;;#rstats #rshiny #earlconf;"907977385613676545";https://twitter.com/sellorm/status/907977385613676545 33 | sellorm;2017-09-13 14:53;1;1;"Fascinating talk on the perceptual differences in colour space for visualisations in #rstats from Jonathan Usmar. #earlconf";;;#rstats #earlconf;"907965371998765056";https://twitter.com/sellorm/status/907965371998765056 34 | sellorm;2017-09-13 14:20;1;4;"Really interesting talk from @neilcharles_uk on how @MediaComUK are using #rstats and shiny. #earlconf";;@neilcharles_uk @MediaComUK;#rstats #earlconf;"907957108578152448";https://twitter.com/sellorm/status/907957108578152448 35 | TraineeGeek;2017-09-13 13:43;1;3;"@EARLConf really enjoying #EARLConf day 1, informative talks &keynotes. Lots of interesting projects using R &Shiny, incl sewer performance";;@EARLConf;#EARLConf;"907947920112177152";https://twitter.com/TraineeGeek/status/907947920112177152 36 | sellorm;2017-09-13 10:24;3;13;"Mash-up of rage tweets by @JennyBryan and @tpoi at #EARLconf #rstats pic.twitter.com/R5T2OVi71v";;@JennyBryan @tpoi;#EARLconf #rstats;"907897788511506432";https://twitter.com/sellorm/status/907897788511506432 37 | AlFirrell;2017-09-13 10:23;3;12;"Amen, sister! @JennyBryan #rstats #EARLConf pic.twitter.com/dwcs87sFXj";;@JennyBryan;#rstats #EARLConf;"907897592725610496";https://twitter.com/AlFirrell/status/907897592725610496 38 | revodavid;2017-09-13 09:25;2;9;"UK Office of National Statistics is building up its Data Science practice — @_datasmith at #EARLConf pic.twitter.com/Hr8mjEODgi";;@_datasmith;#EARLConf;"907882975035052034";https://twitter.com/revodavid/status/907882975035052034 39 | AlFirrell;2017-09-13 08:53;3;7;"Coffee? Check. Pastry? Check. 2 days of fascinating #rstats geek chat? Check. Bring on the #EARLConf !";;;#rstats #EARLConf;"907874979026481152";https://twitter.com/AlFirrell/status/907874979026481152 40 | sellorm;2017-09-12 11:50;1;3;". @dougashton talks SparklyR and MLlib at #earlconf #rstats pic.twitter.com/xOBYjWta6t";;@dougashton;#earlconf #rstats;"907557059645800453";https://twitter.com/sellorm/status/907557059645800453 41 | sellorm;2017-09-12 10:29;1;5;"#EARLconf and @aimeegott_R leading the workshop about #BigData and #rstats with SparklyR pic.twitter.com/KFg5EVs7lA";;@aimeegott_R;#EARLconf #BigData #rstats;"907536570659622912";https://twitter.com/sellorm/status/907536570659622912 42 | dougashton;2017-09-12 10:27;2;3;"@aimeegott_R is starting dplyr (Spark SQL) and I'm prepping for MLlib in the -r workshop #EARLConf pic.twitter.com/OT6v8wCSAu";;@aimeegott_R;#EARLConf;"907536148171513858";https://twitter.com/dougashton/status/907536148171513858 43 | timvg80;2017-09-12 10:08;0;3;"Nice view from work today. #EARLconf pic.twitter.com/aTNNiGDipG";;;#EARLconf;"907531433333411840";https://twitter.com/timvg80/status/907531433333411840 44 | aimeegott_R;2017-09-12 10:05;2;4;"Talking big data in #rstats in the sparklyr workshop at #earlconf with @dougashton and @sellorm pic.twitter.com/4EJpi2ntBg";;@dougashton @sellorm;#rstats #earlconf;"907530608620040192";https://twitter.com/aimeegott_R/status/907530608620040192 45 | earlconf;2017-09-07 17:19;3;6;"Less than a week now! We can't wait! #rstats #earlconf https:// twitter.com/MangoTheCat/st atus/905827674718920706 …";;;#rstats #earlconf;"905827865115189248";https://twitter.com/earlconf/status/905827865115189248 46 | MangoTheCat;2017-09-06 08:06;1;1;"Bring on next week! Thanks for the great post @revodavid !! #rstats #DataScience #EARLConf https:// twitter.com/revodavid/stat us/905086173659500545 …";;@revodavid;#rstats #DataScience #EARLConf;"905326405130620929";https://twitter.com/MangoTheCat/status/905326405130620929 47 | earlconf;2017-09-05 17:22;0;1;"We're getting super excited! #rstats #EARLConf #datascience https:// twitter.com/revodavid/stat us/905086173659500545 …";;;#rstats #EARLConf #datascience;"905103810502287362";https://twitter.com/earlconf/status/905103810502287362 48 | MangoTheCat;2017-09-03 12:01;1;1;"The deadline is today! #rstats #earlconf https:// twitter.com/earlconf/statu s/903206146894426116 …";;;#rstats #earlconf;"904298422345326592";https://twitter.com/MangoTheCat/status/904298422345326592 49 | RLadiesGlobal;2017-09-02 19:45;4;6;"Two more days to apply to #earlconf Boston! The deadline is tomorrow. #RLadies https:// twitter.com/earlconf/statu s/903206146894426116 …";;;#earlconf #RLadies;"904052743853060096";https://twitter.com/RLadiesGlobal/status/904052743853060096 50 | MangoTheCat;2017-08-31 11:42;0;0;"You've got all weekend to submit your abstract! #rstats #datascience #EARLConf https:// twitter.com/earlconf/statu s/903206146894426116 …";;;#rstats #datascience #EARLConf;"903206398015787008";https://twitter.com/MangoTheCat/status/903206398015787008 51 | earlconf;2017-08-29 09:37;1;1;"The Shiny workshop has sold out! Fear not, we still have 5 other great workshops to choose from! https://www. mango-solutions.com/blog/expand-yo ur-knowledge-r-workshops-at-earl-london … #rstats #EARLConf";;;#rstats #EARLConf;"902450136151982080";https://twitter.com/earlconf/status/902450136151982080 52 | tanyacash21;2017-08-24 17:21;4;9;". @dataandme represent! http:// tcbanalytics.com/2017/08/24/mar a-averick-keynote-boston-earl-2017/ … #EARLconf @MangoTheCat @tcbanalytics";;@dataandme @MangoTheCat @tcbanalytics;#EARLconf;"900754958718521344";https://twitter.com/tanyacash21/status/900754958718521344 53 | earlconf;2017-08-11 11:04;5;11;"Don't miss @JennyBryan 's workshop at #EARLConf #London ! Book now: https:// ti.to/earl-conferenc e/london-2017/with/jrdkfkf5hpo … #rstats @rstudio pic.twitter.com/glG97EXX7c";;@JennyBryan @rstudio;#EARLConf #London #rstats;"895948993674530816";https://twitter.com/earlconf/status/895948993674530816 54 | tcbanalytics;2017-08-10 22:35;2;4;"The EARL conference is great every year, can't wait to have our own @dataandme keynote at it! #rstats #earlconf @earlconf @MangoTheCat https:// twitter.com/dataandme/stat us/895279145659322368 …";;@dataandme @earlconf @MangoTheCat;#rstats #earlconf;"895760460187881472";https://twitter.com/tcbanalytics/status/895760460187881472 55 | MangoMattGlover;2017-06-06 21:41;2;2;"Dennis Salguero talking about predictive policing in R! #earl2017 #earlconf #rstats pic.twitter.com/YLZ9ZtkYaF";;;#earl2017 #earlconf #rstats;"872191709496524800";https://twitter.com/MangoMattGlover/status/872191709496524800 56 | b23kelly;2017-05-22 16:23;1;6;"Excited for it! #earlconf #rstats https:// twitter.com/earlconf/statu s/866507413033279488 …";;;#earlconf #rstats;"866675885080023040";https://twitter.com/b23kelly/status/866675885080023040 57 | MangoMattGlover;2017-05-11 09:50;0;0;"A gift for you! 15% off San Francisco EARL tickets: https:// ti.to/earl-conferenc e/earl-sf-2017/discount/LKIN15OFF … #rstats #earlconf pic.twitter.com/zbkURS19gg";;;#rstats #earlconf;"862590851612102657";https://twitter.com/MangoMattGlover/status/862590851612102657 58 | sellorm;2017-05-05 16:45;4;2;"The #earlconf San Francisco agenda is looking really good! https:// earlconf.com/sanfrancisco/ #stats";;;#earlconf #stats;"860520750889005058";https://twitter.com/sellorm/status/860520750889005058 59 | MangoMattGlover;2017-05-02 11:58;0;0;"Coming to EARL San Francisco? Use the discount code MANGO17MG #rstats #earlconf https:// earlconf.com/sanfrancisco/";;;#rstats #earlconf;"859361328305233922";https://twitter.com/MangoMattGlover/status/859361328305233922 60 | aimeegott_R;2017-04-26 16:00;2;7;"Can't quite believe it but found a use for the new emo package from @hadleywickham #rstats #earlconf";;@hadleywickham;#rstats #earlconf;"857247899163525122";https://twitter.com/aimeegott_R/status/857247899163525122 61 | MangoTheCat;2017-01-02 19:02;5;2;"The London @earlconf returns this year! Block out 12-14 September in your calendar. #rstats #earlconf #datascience #savethedate";;@earlconf;#rstats #earlconf #datascience #savethedate;"815996832228720640";https://twitter.com/MangoTheCat/status/815996832228720640 62 | earlconf;2016-12-27 10:41;3;3;"Register your interest for EARL 2017 & be the first to know when registrations open: https:// earlconf.com/2017/ #rstats #earlconf #datascience";;;#rstats #earlconf #datascience;"813696176788799488";https://twitter.com/earlconf/status/813696176788799488 63 | earlconf;2016-12-22 09:58;2;1;"In time for Christmas, register your interest for EARL 2017! https:// earlconf.com/2017/ #rstats #datascience #earlconf";;;#rstats #datascience #earlconf;"811873525913026560";https://twitter.com/earlconf/status/811873525913026560 64 | tanyacash21;2016-11-09 19:15;0;4;". @revodavid presenting on the rising popularity of #rstats and its applications. #earlconf pic.twitter.com/bIlgOAad9O – bei Museum of Science";;@revodavid;#rstats #earlconf;"796431109802360833";https://twitter.com/tanyacash21/status/796431109802360833 65 | tanyacash21;2016-11-09 18:45;2;4;". @AedinCulhane talking about how @DanaFarber used #rstats for cancer research. #earlconf pic.twitter.com/WpJ1uszuzk";;@AedinCulhane @DanaFarber;#rstats #earlconf;"796423398213746697";https://twitter.com/tanyacash21/status/796423398213746697 66 | ajlugtigheid;2016-09-13 16:46;0;0;"#earlconf Joe Cheng talking about a feature in Shiny 0.14 (released yesterday): bookmarkable state. pic.twitter.com/It2puVd5IR";;;#earlconf;"775722217585905664";https://twitter.com/ajlugtigheid/status/775722217585905664 67 | SteffLocke;2016-07-19 14:07;0;1;"Nice - #earlconf making it back to Boston this year! Well worth putting in sessions and/or attending https:// twitter.com/earlconf/statu s/755387470875721728 …";;;#earlconf;"755388660305031168";https://twitter.com/SteffLocke/status/755388660305031168 68 | CDRC_UK;2016-06-14 12:47;0;0;"'The power of maps with R' - video presentation of @spatialanalysis at the #EARLconf available to view: http:// bit.ly/1PoSUMR";;@spatialanalysis;#EARLconf;"742684821177139200";https://twitter.com/CDRC_UK/status/742684821177139200 69 | CSJCampbell;2016-06-07 23:37;0;0;"Earlybird for #earlconf available until Friday https:// earlconf.com/london/registe r/ … excited to meet up with everyone and talk #rstats";;;#earlconf #rstats;"740311782523293698";https://twitter.com/CSJCampbell/status/740311782523293698 -------------------------------------------------------------------------------- /data/tweets/output_got_earl2017.csv: -------------------------------------------------------------------------------- 1 | username;date;retweets;favorites;text;geo;mentions;hashtags;id;permalink 2 | AlanHoKT;2017-10-02 02:15;0;0;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2fIwWPG pic.twitter.com/wLldlx1NI4";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"914660062034931712";https://twitter.com/AlanHoKT/status/914660062034931712 3 | johnon2;2017-09-23 16:02;0;0;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2xq6a2R pic.twitter.com/xw1nAgOKNq";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"911606693393457154";https://twitter.com/johnon2/status/911606693393457154 4 | AndySugs;2017-09-21 22:19;0;1;"RT: LearnRinaDay: EARL London 2017 – That’s a wrap! https://www. r-bloggers.com/earl-london-20 17-thats-a-wrap/ … #rstats #EARL2017 #programming #BigData #OpenSource #coding #D …";;;#rstats #EARL2017 #programming #BigData #OpenSource #coding #D;"910976736862654464";https://twitter.com/AndySugs/status/910976736862654464 5 | LearnRinaDay;2017-09-21 22:17;3;6;"EARL London 2017 – That’s a wrap! https://www. r-bloggers.com/earl-london-20 17-thats-a-wrap/ … #rstats #EARL2017 #programming #BigData #OpenSource #coding #DataScience #DataViz";;;#rstats #EARL2017 #programming #BigData #OpenSource #coding #DataScience #DataViz;"910976256493092864";https://twitter.com/LearnRinaDay/status/910976256493092864 6 | LouBajuk;2017-09-20 23:15;0;0;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2xi6sZD pic.twitter.com/8ypyfnvWmj";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"910628499316105216";https://twitter.com/LouBajuk/status/910628499316105216 7 | pjevrard;2017-09-20 13:02;0;0;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2wy7Xpr pic.twitter.com/h7K3KHVapg";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"910474272501506050";https://twitter.com/pjevrard/status/910474272501506050 8 | tombee74;2017-09-18 18:11;1;2;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2ykxEXz pic.twitter.com/sathU6vYr4";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"909827319392432128";https://twitter.com/tombee74/status/909827319392432128 9 | LeoLeoThePewPew;2017-09-18 10:53;0;0;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2y8DESx pic.twitter.com/nGibBD4G3L";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"909716954872606725";https://twitter.com/LeoLeoThePewPew/status/909716954872606725 10 | RJStriano;2017-09-18 02:45;0;0;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2y7Af6q pic.twitter.com/5X9riObuNY";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"909594192736698368";https://twitter.com/RJStriano/status/909594192736698368 11 | rachelhilll;2017-09-15 17:02;0;0;". @TIBCO 's @loubajuk will be discussing how the @RConsortium helps create vital infrastructure for the R community at #London #EARL2017 . pic.twitter.com/gw7aejx4Ua";;@TIBCO @loubajuk @RConsortium;#London #EARL2017;"908722826340851712";https://twitter.com/rachelhilll/status/908722826340851712 12 | erich_gerber;2017-09-15 07:46;0;1;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2fnnHRi pic.twitter.com/XS78v0nY6c";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"908582707264753664";https://twitter.com/erich_gerber/status/908582707264753664 13 | MunjalVaidya;2017-09-15 05:27;0;1;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2h5LHbI pic.twitter.com/qYPGasaU4Q";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"908547921531961351";https://twitter.com/MunjalVaidya/status/908547921531961351 14 | erich_gerber;2017-09-15 04:48;0;1;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2h6mRZw pic.twitter.com/PYEXieO7eK";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"908537865872728064";https://twitter.com/erich_gerber/status/908537865872728064 15 | andrew_weekm;2017-09-15 02:58;0;1;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2h5dIQB pic.twitter.com/EOMFVDn86W";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"908510422570909697";https://twitter.com/andrew_weekm/status/908510422570909697 16 | LeoLeoThePewPew;2017-09-15 02:37;0;1;". @TIBCO 's @loubajuk will be discussing how the @RConsortium helps create vital infrastructure for the R community at #London #EARL2017 . pic.twitter.com/5mpcxH7zOP";;@TIBCO @loubajuk @RConsortium;#London #EARL2017;"908505140243578880";https://twitter.com/LeoLeoThePewPew/status/908505140243578880 17 | wyattcadetucker;2017-09-14 19:30;0;0;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2vWxqnT pic.twitter.com/PLJX6s2JUR";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"908397588398739456";https://twitter.com/wyattcadetucker/status/908397588398739456 18 | djeda1959;2017-09-14 16:42;3;3;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2y8Qvor pic.twitter.com/uJnD76HkFe";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"908355379075047424";https://twitter.com/djeda1959/status/908355379075047424 19 | G0RAN_J0HANSS0N;2017-09-14 16:15;0;0;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2wc3zMH pic.twitter.com/4BbMvUiWRH";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"908348485782474752";https://twitter.com/G0RAN_J0HANSS0N/status/908348485782474752 20 | LouBajuk;2017-09-14 14:34;1;3;"Dan Thompson from Kluster Intelligence, on using R in sales to win more business #EARL2017 pic.twitter.com/iYZnqdmWBW";;;#EARL2017;"908323024511922177";https://twitter.com/LouBajuk/status/908323024511922177 21 | TIBCO;2017-09-14 13:42;2;4;". @TIBCO 's @LouBajuk spoke at #EARL2017 London on the great work done by the @rconsortium to benefit the R community: http:// tibco.cm/2wYijOC pic.twitter.com/BGnwvNgYak";;@TIBCO @LouBajuk @rconsortium;#EARL2017;"908309917039251456";https://twitter.com/TIBCO/status/908309917039251456 22 | LouBajuk;2017-09-14 13:38;2;4;"Great summary by Reed's Annabel St. John on the drivers behind Facebook's prophet package #EARL2017 pic.twitter.com/a8b3j9ix39";;;#EARL2017;"908309046725414913";https://twitter.com/LouBajuk/status/908309046725414913 23 | alice_data;2017-09-14 13:38;2;3;"Facts about @Facebook prophet available in #rstats and #Python #EARL2017 pic.twitter.com/bQiHWTPxhh";;@Facebook;#rstats #Python #EARL2017;"908309020758474752";https://twitter.com/alice_data/status/908309020758474752 24 | LouBajuk;2017-09-14 12:43;1;2;"Thanks to #EARL2017 for the opportunity to give an update on the great work the @RConsortium is doing. https://www. slideshare.net/mobile/loubaju kyorgan/r-consortium-update-earl-london-sept-2017 … pic.twitter.com/agwCstUfPt";;@RConsortium;#EARL2017;"908295253920305153";https://twitter.com/LouBajuk/status/908295253920305153 25 | LouBajuk;2017-09-14 12:24;1;0;". @TIBCO 's @loubajuk will be discussing how the @RConsortium helps create vital infrastructure for the R community at #London #EARL2017 . pic.twitter.com/hNDzsXH24N";;@TIBCO @loubajuk @RConsortium;#London #EARL2017;"908290248848826369";https://twitter.com/LouBajuk/status/908290248848826369 26 | LouBajuk;2017-09-14 12:14;1;0;"Why an R operating model is important for your enterprise, from @sellorm at #EARL2017 pic.twitter.com/knE4jsFoLb";;@sellorm;#EARL2017;"908287758137008128";https://twitter.com/LouBajuk/status/908287758137008128 27 | LouBajuk;2017-09-14 12:11;2;3;". @sellorm from @MangoTheCat gives a comprehensive (and visually engaging) ppt on an R Operating Model for the Enterprise #EARL2017 pic.twitter.com/8CXzNb6nN6";;@sellorm @MangoTheCat;#EARL2017;"908287145407901696";https://twitter.com/LouBajuk/status/908287145407901696 28 | LouBajuk;2017-09-14 11:28;1;1;"Always good advice: "be as close as possible to the decision makers when building a model"--to make sure it's actually useful. #EARL2017";;;#EARL2017;"908276214720327680";https://twitter.com/LouBajuk/status/908276214720327680 29 | LouBajuk;2017-09-14 11:17;1;5;"Alex Campbell at Receipt Bank on RFM analysis for Customer Segmentation--very clear & useful presentation of an analysis #EARL2017 pic.twitter.com/oYdr5yu2lg";;;#EARL2017;"908273489991409664";https://twitter.com/LouBajuk/status/908273489991409664 30 | LouBajuk;2017-09-13 16:47;3;5;". @peak_bi considered various solutions to put their R ML models into production #EARL2017 pic.twitter.com/N4EdSNGqU1";;@peak_bi;#EARL2017;"907994237177929728";https://twitter.com/LouBajuk/status/907994237177929728 31 | LouBajuk;2017-09-13 16:44;1;3;". @peak_bi tried various ML algorithms to score leads for a car buying service, settled on xgboost. Drove 25% increase in revenue #EARL2017 pic.twitter.com/VpQfJFE8Od";;@peak_bi;#EARL2017;"907993414179975170";https://twitter.com/LouBajuk/status/907993414179975170 32 | LouBajuk;2017-09-13 16:24;1;2;"Great overview of using R for cash flow modeling by @richatmango #EARL2017 pic.twitter.com/ge0tgEZMCq";;@richatmango;#EARL2017;"907988429178331141";https://twitter.com/LouBajuk/status/907988429178331141 33 | LouBajuk;2017-09-13 16:03;1;2;"Thanks to everyone who attended my talk on Making Data Science Accessible at #earl2017 today. Here are my slides. https://www. slideshare.net/secret/83Gy4Xr EqR6VCO …";;;#earl2017;"907983198931562496";https://twitter.com/LouBajuk/status/907983198931562496 34 | LouBajuk;2017-09-13 15:55;1;1;"Role of people in a successful project is too often neglected in a presentation. Great to see this in the PartnerRe presentation #EARL2017 pic.twitter.com/XW5Ip74bTv";;;#EARL2017;"907981085912567808";https://twitter.com/LouBajuk/status/907981085912567808 35 | LouBajuk;2017-09-13 15:47;1;2;"ParrnerRe's process for using R for ReInsurance pricing #EARL2017 pic.twitter.com/XIKwibRpGF";;;#EARL2017;"907978966086115328";https://twitter.com/LouBajuk/status/907978966086115328 36 | LouBajuk;2017-09-13 15:45;2;3;"Great summary of why PartnerRe uses R for Life ReInsurance pricing (and why R is used by many companies for many applications) #EARL2017 pic.twitter.com/DgJ63ItLN0";;;#EARL2017;"907978434189684736";https://twitter.com/LouBajuk/status/907978434189684736 37 | dmi3k;2017-09-13 09:03;1;4;"Exciting!! @earlconf #EARL2017 pic.twitter.com/XMjhnRgUNt";;@earlconf;#EARL2017;"907877406291230721";https://twitter.com/dmi3k/status/907877406291230721 38 | erich_gerber;2017-09-12 23:43;1;2;". @TIBCO 's @loubajuk will be discussing how the @RConsortium helps create vital infrastructure for the R community at #London #EARL2017 . pic.twitter.com/4NRJMRB4DE";;@TIBCO @loubajuk @RConsortium;#London #EARL2017;"907736405979176961";https://twitter.com/erich_gerber/status/907736405979176961 39 | G0RAN_J0HANSS0N;2017-09-12 20:15;0;1;". @TIBCO 's @loubajuk will be discussing how the @RConsortium helps create vital infrastructure for the R community at #London #EARL2017 . pic.twitter.com/BtEssXVsjK";;@TIBCO @loubajuk @RConsortium;#London #EARL2017;"907684107236237312";https://twitter.com/G0RAN_J0HANSS0N/status/907684107236237312 40 | andrew_weekm;2017-09-12 18:22;0;0;". @TIBCO 's @loubajuk will be discussing how the @RConsortium helps create vital infrastructure for the R community at #London #EARL2017 . pic.twitter.com/LrmpQrQeby";;@TIBCO @loubajuk @RConsortium;#London #EARL2017;"907655771504091136";https://twitter.com/andrew_weekm/status/907655771504091136 41 | TIBCO;2017-09-12 18:01;0;0;". @TIBCO 's @loubajuk will be discussing how the @RConsortium helps create vital infrastructure for the R community at #London #EARL2017 . pic.twitter.com/iNPbosM9Dh";;@TIBCO @loubajuk @RConsortium;#London #EARL2017;"907650457610407936";https://twitter.com/TIBCO/status/907650457610407936 42 | djeda1959;2017-09-12 15:33;2;2;". @TIBCO 's @loubajuk will be discussing how the @RConsortium helps create vital infrastructure for the R community at #London #EARL2017 . pic.twitter.com/IJJdTA1FxR";;@TIBCO @loubajuk @RConsortium;#London #EARL2017;"907613226052890624";https://twitter.com/djeda1959/status/907613226052890624 43 | dmi3k;2017-09-12 14:48;2;4;"Shiny wisdom @earlconf with @aimeegott_R #earl2017 pic.twitter.com/cXRlv0lbmR";;@earlconf @aimeegott_R;#earl2017;"907601903411965953";https://twitter.com/dmi3k/status/907601903411965953 44 | dmi3k;2017-09-12 12:49;3;11;"Awesome workshop on writing functions by @JennyBryan at @earlconf #earl2017 pic.twitter.com/D5LlZQpi48";;@JennyBryan @earlconf;#earl2017;"907571935680364544";https://twitter.com/dmi3k/status/907571935680364544 45 | amandadailey_;2017-09-07 04:21;0;0;"Earl where ya at #stillwaiting #earl2017";;;#stillwaiting #earl2017;"905632210429984768";https://twitter.com/amandadailey_/status/905632210429984768 46 | LearnRinaDay;2017-09-06 21:53;3;7;"Preview of EARL London 2017: https://www. r-bloggers.com/mapping-the-la rgest-cities-in-asia-using-r/ … #rstats #EARL2017 #coding #programming #conference #BigData #OpenSource #DataScience #R";;;#rstats #EARL2017 #coding #programming #conference #BigData #OpenSource #DataScience #R;"905534442726461444";https://twitter.com/LearnRinaDay/status/905534442726461444 47 | amandadailey_;2017-09-06 18:28;0;0;"Love this time of year #earl2017";;;#earl2017;"905482852636254208";https://twitter.com/amandadailey_/status/905482852636254208 48 | JDorbeaux;2017-08-02 10:27;0;0;"#earl2017 #earlforpresident";;;#earl2017 #earlforpresident;"892678238387838976";https://twitter.com/JDorbeaux/status/892678238387838976 49 | davidejansen;2017-06-29 15:36;0;2;"How R Powers Data Science at Microsoft http:// buff.ly/2u0rc6a #rstats #EARL2017 #datascience pic.twitter.com/Ynjx4PB5OM";;;#rstats #EARL2017 #datascience;"880434698559909889";https://twitter.com/davidejansen/status/880434698559909889 50 | DD_Thoughtflow;2017-06-16 14:14;1;0;"Thoughtful discussion of #EARL2017 talk by @groundwalkergmb (thread). #Thoughtflow can indeed deliver, but still a ways to go https:// twitter.com/MikeKSmith/sta tus/875664529975324673 …";;@groundwalkergmb;#EARL2017 #Thoughtflow;"875703077231263744";https://twitter.com/DD_Thoughtflow/status/875703077231263744 51 | AriLamstein;2017-06-08 03:36;2;7;"San Francisco #EARL2017 was great! Wonderful group of speakers and attendees. I hope that there's another one next year! #rstats #ThankYou";;;#EARL2017 #rstats #ThankYou;"872643454131609600";https://twitter.com/AriLamstein/status/872643454131609600 52 | LouBajuk;2017-06-07 23:26;2;1;"Slides from my talk today on the @RConsortium , and why @TIBCO is a part of it, from #EARL2017 https://www. slideshare.net/loubajukyorgan /r-consortium-update-for-earl-june-2017 …";;@RConsortium @TIBCO;#EARL2017;"872580524589539328";https://twitter.com/LouBajuk/status/872580524589539328 53 | LouBajuk;2017-06-07 22:21;1;2;"Great talk from @AriLamstein on http:// MonetizingRPackages.com #EARL2017";;@AriLamstein;#EARL2017;"872564205760651265";https://twitter.com/LouBajuk/status/872564205760651265 54 | LouBajuk;2017-06-07 22:02;3;5;"Great talk from @AriLamstein on monetizing #rstats packages #EARL2017 pic.twitter.com/S2pg3eTyYt";;@AriLamstein;#rstats #EARL2017;"872559529971531777";https://twitter.com/LouBajuk/status/872559529971531777 55 | bashlee_R;2017-06-07 21:56;1;4;"Great talk from @richatmango Also loved the Ancient Greek lesson ! #EARL2017 pic.twitter.com/Bafn1x6cq4";;@richatmango;#EARL2017;"872558018885320705";https://twitter.com/bashlee_R/status/872558018885320705 56 | LouBajuk;2017-06-07 20:24;1;2;"Advice for transitioning from Excel to #rstats from Uber's @emilywbailey at #EARL2017 pic.twitter.com/w3TNiw0vWv";;@emilywbailey;#rstats #EARL2017;"872534840725225473";https://twitter.com/LouBajuk/status/872534840725225473 57 | LouBajuk;2017-06-07 20:06;1;2;"Advantages of moving from Excel to R #rstats #EARL2017 pic.twitter.com/MI4tyrPYWY";;;#rstats #EARL2017;"872530226575626240";https://twitter.com/LouBajuk/status/872530226575626240 58 | LouBajuk;2017-06-07 20:04;1;1;"What does an analyst do? #EARL2017 pic.twitter.com/ASt7Ql5z3z";;;#EARL2017;"872529738291490816";https://twitter.com/LouBajuk/status/872529738291490816 59 | grousselle;2017-06-07 19:21;1;1;"Cool user model from Pandora #EARL2017 pic.twitter.com/q9JAEeokfG";;;#EARL2017;"872518817972342784";https://twitter.com/grousselle/status/872518817972342784 60 | grousselle;2017-06-07 18:25;2;3;"Saturday night life in various cities based on Uber requests data #EARL2017 pic.twitter.com/vAWRWaXbvW";;;#EARL2017;"872504719960195072";https://twitter.com/grousselle/status/872504719960195072 61 | LouBajuk;2017-06-07 17:15;2;3;". @MangoTheCat presenting on the various #DataScience teams at Uber #EARL2017 pic.twitter.com/qKwW6PpMks";;@MangoTheCat;#DataScience #EARL2017;"872487156840710144";https://twitter.com/LouBajuk/status/872487156840710144 62 | daroczig;2017-06-07 00:36;2;12;"Thanks for having me at #EARL2017 , fantastic audience and questions! Please find my #rstats slides at http:// bit.ly/EARL-2017-SF-K inesis … if interested https:// twitter.com/sellorm/status /872234209569525760 …";;;#EARL2017 #rstats;"872235657858433024";https://twitter.com/daroczig/status/872235657858433024 63 | LouBajuk;2017-06-06 22:41;0;0;"No Databricks, I'm sorry, I will not accept this as a verb. "Dogfooding our own solution" #earl2017";;;#earl2017;"872206871372812289";https://twitter.com/LouBajuk/status/872206871372812289 64 | MangoMattGlover;2017-06-06 21:41;2;2;"Dennis Salguero talking about predictive policing in R! #earl2017 #earlconf #rstats pic.twitter.com/YLZ9ZtkYaF";;;#earl2017 #earlconf #rstats;"872191709496524800";https://twitter.com/MangoMattGlover/status/872191709496524800 65 | johncwhitmer;2017-06-06 21:36;1;6;""It takes a big man to admit that his data is small" brilliant. Cited to @hadleywickam at #EARL2017";;@hadleywickam;#EARL2017;"872190537364844547";https://twitter.com/johncwhitmer/status/872190537364844547 66 | daroczig;2017-06-06 20:27;5;8;"The first session at #earl2017 just ended with lot of interesting thoughts on building effective infra in production​ on the top of #rstats pic.twitter.com/WYYKiMIdWx – bei Holiday Inn Golden Gateway Hotel";;;#earl2017 #rstats;"872172996785446912";https://twitter.com/daroczig/status/872172996785446912 67 | LouBajuk;2017-06-06 20:02;3;6;". @SelfScore is hiring for full stack #DataEngineer . Unicorns welcome. #EARL2017 pic.twitter.com/JuGPcDXO5u";;@SelfScore;#DataEngineer #EARL2017;"872166904328101888";https://twitter.com/LouBajuk/status/872166904328101888 68 | LouBajuk;2017-06-06 19:46;4;6;"And the #rstats version #EARL2017 pic.twitter.com/YUvgEuoQXV";;;#rstats #EARL2017;"872162912818708480";https://twitter.com/LouBajuk/status/872162912818708480 69 | LouBajuk;2017-06-06 19:45;3;4;". @SelfScore view of the full software stack, as an analogy for #DataScience at #EARL2017 pic.twitter.com/Ax7Upmw1rS";;@SelfScore;#DataScience #EARL2017;"872162549722103808";https://twitter.com/LouBajuk/status/872162549722103808 70 | LouBajuk;2017-06-06 19:13;4;6;"Provocative view from @RStudioJoe #EARL2017 : there is no innocent #DataScientist pic.twitter.com/tX0ptxxvq2";;@RStudioJoe;#EARL2017 #DataScientist;"872154476747960320";https://twitter.com/LouBajuk/status/872154476747960320 71 | LouBajuk;2017-06-06 19:06;6;5;"Nice summary of the process of #DataScience from @rstudio @RStudioJoe at #EARL2017 pic.twitter.com/YkbZ6g6X9x";;@rstudio @RStudioJoe;#DataScience #EARL2017;"872152796887269376";https://twitter.com/LouBajuk/status/872152796887269376 72 | tomaz_tsql;2017-06-06 18:35;1;3;"Have a great time! #rstats #earl2017 //cc: @earlconf @MangoTheCat @daroczig https:// twitter.com/daroczig/statu s/872126931524763648 …";;@earlconf @MangoTheCat @daroczig;#rstats #earl2017;"872144886329376770";https://twitter.com/tomaz_tsql/status/872144886329376770 73 | johncwhitmer;2017-06-06 18:29;1;4;"Impressive approach to #datascience at @Airbnb - especially widespread education beginning with stats concepts #Earl2017";;@Airbnb;#datascience #Earl2017;"872143405706563585";https://twitter.com/johncwhitmer/status/872143405706563585 74 | LouBajuk;2017-06-06 18:19;8;5;". @ricardobion at #EARL2017 on @Airbnb internal data university, teaching every employee (some) #DataScience pic.twitter.com/YIquNiyjHW";;@ricardobion @Airbnb;#EARL2017 #DataScience;"872140824930209792";https://twitter.com/LouBajuk/status/872140824930209792 75 | groundwalkergmb;2017-06-06 17:58;1;2;". @ricardobion talking about the difference between data-informed and data-driven companies. airbnb is data informed. #EARL2017";;@ricardobion;#EARL2017;"872135603990802433";https://twitter.com/groundwalkergmb/status/872135603990802433 76 | groundwalkergmb;2017-06-06 17:45;1;2;"@hspter I think CodeDepends can be used to make the dependency DAG you talked about in your #EARL2017 talk. We should chat about that.";;@hspter;#EARL2017;"872132339379445760";https://twitter.com/groundwalkergmb/status/872132339379445760 77 | LouBajuk;2017-06-06 17:38;12;21;". @hspter #earl2017 : Assume your colleagues have good intentions. If they make an error, blame (& fix) the process they used, not the person";;@hspter;#earl2017;"872130574604840960";https://twitter.com/LouBajuk/status/872130574604840960 78 | AndySugs;2017-06-06 17:24;0;1;"RT: (daroczig) The #earl2017 conf in San Francisco is just starting, looking forward to the next two days full of #rstats talks and networ…";;;#earl2017 #rstats;"872127054749331456";https://twitter.com/AndySugs/status/872127054749331456 79 | daroczig;2017-06-06 17:23;5;15;"The #earl2017 conf in San Francisco is just starting, looking forward to the next two days full of #rstats talks and networking pic.twitter.com/z4pNaoqCVK – bei Holiday Inn Golden Gateway Hotel";;;#earl2017 #rstats;"872126931524763648";https://twitter.com/daroczig/status/872126931524763648 80 | LouBajuk;2017-06-06 17:19;6;11;". @hspter at #earl2017 on the myth of the perfect #DataScientist vs. Analytic Development pic.twitter.com/EGof1xQTsi";;@hspter;#earl2017 #DataScientist;"872125892260052993";https://twitter.com/LouBajuk/status/872125892260052993 81 | LouBajuk;2017-06-06 17:12;2;4;". @richatmango from @MangoTheCat kicks off #earl2017 In #SanFrancisco pic.twitter.com/xrahtcWVSl";;@richatmango @MangoTheCat;#earl2017 #SanFrancisco;"872124149866479616";https://twitter.com/LouBajuk/status/872124149866479616 82 | DataScienceLA;2017-06-01 20:27;6;16;"I'll be speaking at @earlconf #EARL2017 next week about using R for machine learning in production for business applications #rstats pic.twitter.com/DURGwM9ZOs";;@earlconf;#EARL2017 #rstats;"870361295542931457";https://twitter.com/DataScienceLA/status/870361295542931457 83 | richatmango;2017-05-31 16:32;5;8;"I aint getting on no plane! The cats are on their way to #earl2017 #sanfrancisco . See u there! #rstats #datascience @earlconf @MangoTheCat pic.twitter.com/OxocTPyj2n";;@earlconf @MangoTheCat;#earl2017 #sanfrancisco #rstats #datascience;"869939631005069317";https://twitter.com/richatmango/status/869939631005069317 84 | richatmango;2017-05-31 15:15;7;10;"if you know companies who still use SAS you may want to send them to #earl2017 next week in #sanfrancisco #rstats #datascience @earlconf https:// twitter.com/nic_crane/stat us/869647044310433792 …";;@earlconf;#earl2017 #sanfrancisco #rstats #datascience;"869920342336253953";https://twitter.com/richatmango/status/869920342336253953 85 | richatmango;2017-05-24 17:54;0;1;"maybe in future attendees score talks then throw some optimisation at it to optimise schedule? *digs out old OR notes* #earl2017 #rstats";;;#earl2017 #rstats;"867423586054803456";https://twitter.com/richatmango/status/867423586054803456 86 | richatmango;2017-05-24 17:25;2;1;"2 streams at #earl2017 #SanFrancisco - need to plan my session hoping! suggestions? :) #rstats #datascience https:// earlconf.com/downloads/sanf rancisco/EARL-SF-2017-Agenda.pdf …";;;#earl2017 #SanFrancisco #rstats #datascience;"867416351626989568";https://twitter.com/richatmango/status/867416351626989568 87 | richatmango;2017-05-19 12:16;9;25;"THE fashion accessory for this summer in #sanfrancisco @hadleywickham @earlconf @MangoTheCat #earl2017 #rstats #datascience pic.twitter.com/bgIP1te2b8";;@hadleywickham @earlconf @MangoTheCat;#sanfrancisco #earl2017 #rstats #datascience;"865526647688093696";https://twitter.com/richatmango/status/865526647688093696 88 | richatmango;2017-05-16 20:54;4;5;"Not long to go until #earl2017 in #sanfrancisco !!! See you there!! #rstats #datascience http:// earlconf.com @MangoTheCat @earlconf";;@MangoTheCat @earlconf;#earl2017 #sanfrancisco #rstats #datascience;"864569689372061696";https://twitter.com/richatmango/status/864569689372061696 89 | earlconf;2017-05-11 15:03;2;0;"London #EARL2017 Speakers announced: https:// earlconf.com/london/ Get your earlybird tickets now pic.twitter.com/V5dLrE5lyf";;;#EARL2017;"862669497685159936";https://twitter.com/earlconf/status/862669497685159936 90 | aimeegott_R;2017-05-02 20:15;1;3;"Not long to go now! Can't wait for the first #rstats #earl2017 in SF https:// twitter.com/earlconf/statu s/859420656844361728 …";;;#rstats #earl2017;"859486496742678528";https://twitter.com/aimeegott_R/status/859486496742678528 91 | richatmango;2017-04-28 21:19;0;2;"#earl2017 #sanfrancisco agenda announced! And it's R-some!!! #rstats #datascience https:// twitter.com/MangoTheCat/st atus/857995453840990208 …";;;#earl2017 #sanfrancisco #rstats #datascience;"858052958314999808";https://twitter.com/richatmango/status/858052958314999808 92 | earlconf;2017-04-18 15:54;3;5;"Submitted an abstract for #EARL2017 ? We are wading through them now and hope to have the acceptance notices out asap #rstats";;;#EARL2017 #rstats;"854347341834452993";https://twitter.com/earlconf/status/854347341834452993 93 | aimeegott_R;2017-04-11 10:20;4;4;"Enjoying starting to read through submissions for #earl2017 #london #sanfrancisco still a couple more days to get your abstract in! #rstats";;;#earl2017 #london #sanfrancisco #rstats;"851726618192216065";https://twitter.com/aimeegott_R/status/851726618192216065 94 | Springcoil;2017-03-24 12:32;0;4;"@MangoTheCat I've just submitted a talk for #EARL2017";;@MangoTheCat;#EARL2017;"845252027953373185";https://twitter.com/Springcoil/status/845252027953373185 95 | daroczig;2017-03-15 21:04;0;1;"That would be great and I hope we can meet next time! I plan to attend #EARL2017 SF in June, so that might an option.";;;#EARL2017;"842119398840180736";https://twitter.com/daroczig/status/842119398840180736 96 | earlconf;2017-03-14 16:38;2;0;"@polbots - if you're using #rstats in your fab work and have a story to tell #EARL2017 we'd love to hear from you https:// earlconf.com /";;@polbots;#rstats #EARL2017;"841690015281401856";https://twitter.com/earlconf/status/841690015281401856 97 | earlconf;2017-03-14 16:28;8;4;"Hey @RLadiesLA @RLadiesSF & @RLadiesGlobal -the abstracts for #EARL2017 SF & London are mostly from guys! http:// bit.ly/2nk7Ng5 #rstats pic.twitter.com/mYmZyq6myz";;@RLadiesLA @RLadiesSF @RLadiesGlobal;#EARL2017 #rstats;"841687435474010113";https://twitter.com/earlconf/status/841687435474010113 98 | earlconf;2017-03-10 12:44;0;0;"@ChrisBeeley Great blogposts on your website Chris - can we persuade you to turn one into an abstract for #EARL2017 ?";;@ChrisBeeley;#EARL2017;"840181597487976448";https://twitter.com/earlconf/status/840181597487976448 99 | MangoTheCat;2017-03-08 15:55;0;0;"@Sportradar we're looking for abstracts for R programming language conference #EARL2017 if you're interested? https:// earlconf.com /";;@Sportradar;#EARL2017;"839504970521116672";https://twitter.com/MangoTheCat/status/839504970521116672 100 | LukeMckerr;2017-03-08 15:22;0;0;"@DeanoGNR #earl2017";;@DeanoGNR;#earl2017;"839496479546953728";https://twitter.com/LukeMckerr/status/839496479546953728 101 | martycorden;2017-03-08 15:12;0;0;"@DeanoGNR #earl2017 #TheYearOfJonto";;@DeanoGNR;#earl2017 #TheYearOfJonto;"839494112285962240";https://twitter.com/martycorden/status/839494112285962240 102 | DavidLoughrey;2017-03-08 15:09;0;0;"#earl2017 @DeanoGNR";;@DeanoGNR;#earl2017;"839493294799339521";https://twitter.com/DavidLoughrey/status/839493294799339521 103 | earlconf;2017-03-08 12:34;0;0;"could you please encourage your membership to submit abstracts for #EARL2017 in SF in June https:// earlconf.com/abstracts/ by 24 Mar?";;;#EARL2017;"839454226308808705";https://twitter.com/earlconf/status/839454226308808705 104 | earlconf;2017-03-08 11:52;0;0;"Hey #Rstats Ladies - how about submitting an abstract for #EARL2017 https:// earlconf.com/abstracts/ (London or San Fran?) #InternationalWomensDay";;;#Rstats #EARL2017 #InternationalWomensDay;"839443739076669440";https://twitter.com/earlconf/status/839443739076669440 105 | earlconf;2017-03-07 14:04;3;1;"Thanks to everyone who has submitted abstracts for #EARL2017 London & San Francisco so far - if you haven't yet: https:// earlconf.com/abstracts/ pic.twitter.com/lGgV9cf0rX";;;#EARL2017;"839114595667230721";https://twitter.com/earlconf/status/839114595667230721 106 | LearnRinaDay;2017-02-06 21:14;0;0;"Share your knowledge at EARL 2017 – call for abstracts, #San #Francisco and #London : https://www. r-bloggers.com/share-your-kno wledge-at-earl-2017-call-for-abstracts-san-francisco-and-london/ … #rstats #EARL2017 #conferences";;;#San #Francisco #London #rstats #EARL2017 #conferences;"828713557810348033";https://twitter.com/LearnRinaDay/status/828713557810348033 107 | paoloberdin;2017-01-26 03:36;0;1;"01/25/17 #EArl2017 pic.twitter.com/JS7Tf2m2f5";;;#EArl2017;"824460995556839425";https://twitter.com/paoloberdin/status/824460995556839425 108 | JLdelda;2017-01-25 20:14;2;1;"#R Conference is looking for abstracts: #EARL2017 [Effective Applications of R language] @MangotheCat http:// buff.ly/2jjTXYb";;@MangotheCat;#R #EARL2017;"824349654309601280";https://twitter.com/JLdelda/status/824349654309601280 109 | earlconf;2017-01-24 16:00;3;2;"SO exciting to see the first abstracts coming in for #EARL2017 - have you submitted yours yet? https:// earlconf.com/2017/abstracts/ #rstats pic.twitter.com/VLZvcvTr0u";;;#EARL2017 #rstats;"823923536859426816";https://twitter.com/earlconf/status/823923536859426816 110 | IMAROOSTAR;2016-08-27 14:18;0;2;"@rvpredv @sandorearl bring back #earl2017";;@rvpredv @sandorearl;#earl2017;"769524479823323136";https://twitter.com/IMAROOSTAR/status/769524479823323136 --------------------------------------------------------------------------------