Abstract

In the past years, we have witnessed an explosion of web applications, and in particular of electronic commerce websites. This has led to unquestionable benefits for both producers and consumers of goods. On the other hand, however, untrusted companies have the opportunity to bypass checks and regulations imposed by relevant bodies. This problem is prevalent in the context of online commerce of pharmaceutical products, where it is essential that such products are safe, of good quality, and only used with a proper prescription. In this work, we study the problem of internet pharmacy verification. To this effect, we build a classifier, able to find patterns and predict the class of unseen data. Moreover, we devise algorithms that give a trust score to each pharmacy, in order to have a legitimacy indicator usable by human reviewers. We experimentally evaluate the proposed approach with real data coming from two different time periods. The results demonstrate the effectiveness of our approach, as well as the potential of using similar techniques for automatically checking regulation compliance in electronic commerce.

Source code

You may use this code as is for research purposes, provided that you properly acknowledge the authors using the following reference: Alberto Cordioli, Themis Palpanas. An Automated System for Internet Pharmacy Verification. International Conference on Extending Database Technology (EDBT), Vienna, Austria, March 2018.

• Zip file with source code for all the algorithms used in the paper.