About

This site provides supplemental material and information about the paper "Analyzing Cross-System User Modeling on the Social Web".

Abstract. In this article, we analyze tag-based user profiles, which result from social tagging activities in Social Web systems and particularly in Flickr, Twitter and Delicious. We investigate the characteristics of tag-based user profiles within these systems, examine to what extent tag-based profiles of individual users overlap between the systems and identify significant benefits of cross-system user modeling by means of aggregating the different profiles of a same user.
We present a set of cross-system user modeling strategies and evaluate their performance in generating valuable profiles in the context of tag and resource recommendations in Flickr, Twitter and Delicious. Our evaluation shows that the cross-system user modeling strategies outperform other strategies significantly and have tremendous impact on the recommendation quality in cold-start settings where systems have sparse information about their users.


Slides presented at ICWE:

Analyzing Cross-System User Modeling on the Social Web
View more presentations from Web Information Systems, TU Delft.

1. Dataset Crawling

We crawled user profiles via the Mypes service, which we developed for aggregating user data from different Social Web systems. In particular we performed the following steps.

  1. Search for Google profile URIs via Google Profiles Search with popular given names such as John, Marry: http://www.google.com/profiles?q=john
  2. Input Google Profile URIs into Mypes service: http://mypes.groupme.org/mypes/aggregate?google=http://www.google.com/profiles/109472200289645138170
  3. Mypes then performs a Google Social Graph lookup to obtain the different accounts of the corresponding user
  4. Finally, Mypes crawls the profiles from Twitter, Delicious, Flickr, etc. (if the user holds an account at the corresponding service).

2. Twitter + Delicious Dataset

Tag assignments of 1500 users, who hold an account at Twitter and Delicious in SQL format (SQL table structure + INSERT statements).

name #users #tags #resources #tag assignments description
tas-delicious-td-sample.tar.gz (8MB) 1500 72901 124520 619856 tag assignments performed in Delicious
tas-twitter-td-sample.tar.gz (1MB) 1500 24492 57236 80464 (hash) tag assignments performed by the same users in Twitter

3. Flickr + Delicious Dataset

Tag assignments of 1467 users, who hold an account at Flickr and Delicious in SQL format (SQL table structure + INSERT statements).

name #users #tags #resources #tag assignments description
tas-delicious-fd-sample.tar.gz (9MB) 1467 59275 109242 683665 tag assignments performed in Delicious
tas-flickr-fd-sample.tar.gz (8MB) 1476 72671 166423 892378 tag assignments performed by the same users in Flickr