Introduction

This is the home of the master thesis work Automatic Filtering of Abuse Reports. On this page you can read the abstract and download the full report. Some information about StarClassifier, a program that has been developed during this work, is also available. Contact information can be found at the bottom of this page.

Abstract

This master thesis investigates how abuse reports can be automatically filtered in order to save time, lower costs and increase safety. Abuse reports are reports that users on a website file when they encounter content they find inappropriate. These reports are then generally handled by the Customer Service who decides if the content should be removed from the website. The reports that potentially can be automatically removed are the ones that do not result in deletion of content.

The study that is presented in this report took place at the community website Stardoll. A decision tree was built to classify reports as either good or bad. The over 200 attributes that were used to train the tree contained information about the user who filed the report, the user who the report was filed against and the report itself. Unfortunately no data could be extracted from the reported content.

On unseen data the decision tree correctly removed 22% of the reports that should be removed while incorrectly removing 8% of the reports that should not. These numbers are not good enough to make it feasible to start using the system without further refinements. Such refinements are outlined in this report together with suggestions for how other automated methods can be used at Stardoll and similar communities.

Download the report

Use the link below to download the report.

Download (PDF, 798 kB)

StarClassifier

StarClassifier is a light-weight software tool that can perform classifications on a set of examples in a database and write the result back to the database in question. The classifications made are based on a decision tree that needs to be presented in textual form to the tool. StarClassifier is built in Java and has no dependencies in itself, but in order for it to function properly the appropriate database driver must be available. The tool is not specific to Stardoll and can be used for all tasks where examples need to be classified according to a decision tree.

More information about StarClassifier can be found in Appendix A in the report.

Read more and download (starclassifier.udd.be)

Contact

Feel free to contact me if you have any questions or comments. You can reach me at "thesis", followed by an at sign (@), followed by "udd.be".