@inproceedings{angelova2008characterizing,
  abstract = {Social networks and collaborative tagging systems are rapidly gaining popularity as a primary means for storing and sharing data among friends, family, colleagues, or perfect strangers as long as they have common interests. del.icio.us is a social network where people store and share their personal bookmarks. Most importantly, users tag their bookmarks for ease of information dissemination and later look up. However, it is the friendship links, that make delicious a social network. They exist independently of the set of bookmarks that belong to the users and have no relation to the tags typically assigned to the bookmarks. To study the interaction among users, the strength of the existing links and their hidden meaning, we introduce implicit links in the network. These links connect only highly "similar" users. Here, similarity can reflect different aspects of the user’s profile that makes her similar to any other user, such as number of shared bookmarks, or similarity of their tags clouds. We investigate the question whether friends have common interests, we gain additional insights on the strategies that users use to assign tags to their bookmarks, and we demonstrate that the graphs formed by implicit links have unique properties differing from binomial random graphs or random graphs with an expected power-law degree distribution. },
  author = {Angelova, Ralitsa and Lipczak, Marek and Milios, Evangelos and Prałat, Paweł},
  booktitle = {Proceedings of the Mining Social Data Workshop (MSoDa)},
  interhash = {f74d27a66d2754f3d5892d68c4abee4c},
  intrahash = {02d6739886a13180dd92fbb7243ab58b},
  month = jul,
  organization = {ECAI 2008},
  pages = {21--25},
  title = {Characterizing a social bookmarking and tagging network},
  url = {http://www.math.ryerson.ca/~pralat/papers/2008_delicious.pdf},
  year = 2008
}

@inproceedings{angelova2006graphbased,
  abstract = {Automatic classification of data items, based on training samples, can be boosted by considering the neighborhood of data items in a graph structure (e.g., neighboring documents in a hyperlink environment or co-authors and their publications for bibliographic data entries). This paper presents a new method for graph-based classification, with particular emphasis on hyperlinked text documents but broader applicability. Our approach is based on iterative relaxation labeling and can be combined with either Bayesian or SVM classifiers on the feature spaces of the given data items. The graph neighborhood is taken into consideration to exploit locality patterns while at the same time avoiding overfitting. In contrast to prior work along these lines, our approach employs a number of novel techniques: dynamically inferring the link/class pattern in the graph in the run of the iterative relaxation labeling, judicious pruning of edges from the neighborhood graph based on node dissimilarities and node degrees, weighting the influence of edges based on a distance metric between the classification labels of interest and weighting edges by content similarity measures. Our techniques considerably improve the robustness and accuracy of the classification outcome, as shown in systematic experimental comparisons with previously published methods on three different real-world datasets.},
  acmid = {1148254},
  address = {New York, NY, USA},
  author = {Angelova, Ralitsa and Weikum, Gerhard},
  booktitle = {Proceedings of the 29th annual international ACM SIGIR conference on Research and development in information retrieval},
  doi = {10.1145/1148170.1148254},
  interhash = {ac882d49fa7067c6a0bfca1583ef29b2},
  intrahash = {f4cffa23d78bab3284db2f233c226425},
  isbn = {1-59593-369-7},
  location = {Seattle, Washington, USA},
  numpages = {8},
  pages = {485--492},
  publisher = {ACM},
  series = {SIGIR '06},
  title = {Graph-based text classification: learn from your neighbors},
  url = {http://doi.acm.org/10.1145/1148170.1148254},
  year = 2006
}