This site provides supplemental material and information about the paper "TUMS: Twitter-based User Modeling Service".


The TUMS service allows end-users to inspect Twitter-based profiles and enables other applications to re-use these profiles. People can overview their personal Twitter activities or profiles of other users to explore the topics those users were concerned with in the past. Entity-based, hashtag-based tag clouds allow to quickly grasp the content of the profiles and charts further visualize the profiles. Visualizing the profiles is important to make people aware of what can be inferred from their Twitter activities so that they can reconsider how they publish tweets on Twitter. TUMS is also of interest for other services on the Social Web as it enables them to consume Twitter-based profiles in RDF format. Profiles can be used for personalization and are particularly interesting for other applications that suffer from sparsity problems (e.g. services that cannot collect sufficient data about their users) and services that are interested in "realtime" or very fresh profile information.

Web Application

A demo version of web application of TUMS is accessible at HERE. Just type your Twitter username and click the button, then you get your visualized profile!

Quick start for Using RDF Endpoint

The general pattern of the URL is following:

http://wisserver.st.ewi.tudelft.nl:8080/um-twitter-service/profile/[username]/[profile type]/[weighting]/[enrichment]/?[temporal constraints]

(1) Username: Twitter username that you want to request, e.g. USAGodG20.

(2) Profile type: The profile type parameter enables applications to specify which type of profile they would like to retrieve: hashtag-based, topic-based or entity-based. Given the three types of profiles, the corresponding parameter values that can be used in the HTTP request are the following: hashtag, topic, and entity. For example, the hashtag-based profile can be retrieved as follows:


The weighting scheme, enrichment and temporal constraints can also be adjusted via query. Details can be found in our paper: TUMS - Twitter-based User Modeling Service