Misc Web Science Topics

The comment sections of online newspapers are an important space to indulge in political discussions and discuss various opinions. These discussion forums have to be moderated due to the misuse by spammers, haters, trolls, and means of propaganda. This moderation process is very expensive and many online news providers have discontinued their comment sections. With more and more political campaigning, or even agitation being distributed over the internet, serious and safe platforms to discuss political topics are increasingly important.

In this master thesis, you will analyze comments, users, and articles to understand the dynamics, the information flow, and the interactions in the comment sections. You will work on a subtask of comment analysis, such as detecting inappropriate comments, predicting popular news topics, recommending information, and others. Our recent seminar Text Mining in Practice covered a selection of possible tasks.

For more information please contact Ralf Krestel.

Granted patents form an extensive knowledge base for information retrieval, which is an interesting research field for academia and industry. A common task is to identify patents and patent applications that claim similar inventions. A particular challenge for state-of-the-art approaches is to model and understand domain-specific terminology.

In this master thesis, you will develop a machine learning approach that supports patent officers in finding such similar patents and patent applications, also known as prior art). In order to improve current retrieval models, you can use datasets of millions of patents and patent applications together with their metadata and information about actions taken by patent officers.

The master's thesis can also include a thorough data exploration and analysis. For example, the inventor’s address can be geocoded to the actual geolocation, so that regional patterns can be found. Besides regional patterns, you can analyse patent topics with regard to changes over time. Therefore, you can deal with topic modeling, document embedding, and geocoding. Feel free to come up with a master's thesis topic in the patent domain on your own.

For more information please contact Ralf Krestel.