(Benford's) Law and Order (Fraud) by Rhys Elsmore
72 | 73 | 74 |
Description
100 |"On the internet, fraudulent and abusive behavior is 101 | considered especially heinous. At Heroku, the dedicated detectives who 102 | investigate these vicious felonies are members of an elite squad armed 103 | with large amounts of data and spare CPU cycles. These are their 104 | stories."
105 |Bad behavior can wreak havoc on your web application. It might be 106 | mass-signups, fraudulent orders, spammy posts, right up to automated 107 | bots designed to work around restrictions you have set in place; this 108 | can cost you time, resources, and lots of money. All is not lost though. 109 | Despite the ongoing efforts of abusers, their activity still leaves 110 | fingerprints and clues, which you can use to your advantage.
111 |This talk is a 101 introduction to some of the methods which you can 112 | use to separate good from bad users using a combination of data mining, 113 | statistics, and some some basic machine learning. Basically, I want to 114 | get you thinking like an internet detective.
115 |Some of the topics I will be covering include:
116 |-
117 |
- Collecting and preparing data sources. 118 |
- Effective methods for classifying existing users. 119 |
- Feature extracting; what works and what doesn't. 120 |
- Analyzing user-provided data to profile your users, and weed out the bad operators. 121 |
- Determining a user's intentions by looking at their access patterns. 122 |
- Making use of 'outliers' to find suspicious users and transactions. 123 |
- Stopping bad users before they can wreak havoc. 124 |
As this is a 101 topic I will provide some basic examples, as well as 126 | links to more in-depth resources for further reading. I would recommend 127 | this talk to developers of web applications, especially those with a 128 | large number of users, the ability to process credit cards, or with a 129 | 'free' offering. Attendees should have a basic understanding of topics 130 | such as SQL, Pandas, and some basic understanding of mathematics and 131 | statistics, although this is not essential as I will be providing links 132 | to further reading.
133 |