We develop an algorithm based on Supervised Random Walks that naturally combines the information from the network structure with node and edge level attributes. We achieve this by using these attributes to guide a random walk on the graph. We formulate a supervised learning task where the goal is to learn a function that assigns strengths to edges in the network such that a random walker is more likely to visit the nodes to which new links will be created in the future. We develop an efficient training algorithm to directly learn the edge strength estimation function.
Our experiments on the Facebook social graph and large collaboration networks show that our approach outperforms state-of-the-art unsupervised approaches as well as approaches that are based on feature extraction. ER - TY - GEN AU - Backstrom, L. AU - Leskovec, J. A2 - T1 - Supervised Random Walks: Predicting and Recommending Links in Social Networks JO - PB - AD - PY - 2010/ VL - IS - SP - EP - UR - http://arxiv.org/abs/1011.4071 M3 - KW - predicting KW - random KW - recommending KW - supervised KW - toread KW - walks L1 - N1 - Supervised Random Walks: Predicting and Recommending Links in Social Networks N1 - AB - Predicting the occurrence of links is a fundamental problem in networks. In the link prediction problem we are given a snapshot of a network and would like to infer which interactions among existing members are likely to occur in the near future or which existing interactions are we missing. Although this problem has been extensively studied, the challenge of how to effectively combine the information from the network structure with rich node and edge attribute data remains largely open. We develop an algorithm based on Supervised Random Walks that naturally combines the information from the network structure with node and edge level attributes. We achieve this by using these attributes to guide a random walk on the graph. We formulate a supervised learning task where the goal is to learn a function that assigns strengths to edges in the network such that a random walker is more likely to visit the nodes to which new links will be created in the future. We develop an efficient training algorithm to directly learn the edge strength estimation function. Our experiments on the Facebook social graph and large collaboration networks show that our approach outperforms state-of-the-art unsupervised approaches as well as approaches that are based on feature extraction. ER - TY - CONF AU - Ramage, Daniel AU - Hall, David AU - Nallapati, Ramesh AU - Manning, Christopher D. A2 - T1 - Labeled LDA: A Supervised Topic Model for Credit Attribution in Multi-labeled Corpora T2 - Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing: Volume 1 - Volume 1 PB - Association for Computational Linguistics CY - Stroudsburg, PA, USA PY - 2009/ M2 - VL - IS - SP - 248 EP - 256 UR - http://dl.acm.org/citation.cfm?id=1699510.1699543 M3 - KW - lda KW - model KW - supervised KW - topic KW - toread L1 - SN - 978-1-932432-59-6 N1 - Labeled LDA N1 - AB - A significant portion of the world's text is tagged by readers on social bookmarking websites. Credit attribution is an inherent problem in these corpora because most pages have multiple tags, but the tags do not always apply with equal specificity across the whole document. Solving the credit attribution problem requires associating each word in a document with the most appropriate tags and vice versa. This paper introduces Labeled LDA, a topic model that constrains Latent Dirichlet Allocation by defining a one-to-one correspondence between LDA's latent topics and user tags. This allows Labeled LDA to directly learn word-tag correspondences. We demonstrate Labeled LDA's improved expressiveness over traditional LDA with visualizations of a corpus of tagged web pages from del.icio.us. Labeled LDA outperforms SVMs by more than 3 to 1 when extracting tag-specific document snippets. As a multi-label text classifier, our model is competitive with a discriminative baseline on a variety of datasets. ER - TY - CONF AU - Schmidt-Thieme, Lars A2 - T1 - Compound Classification Models for Recommender Systems. T2 - Proceedings of the 5th IEEE International Conference on Data Mining (ICDM 2005), 27-30 November 2005 PB - IEEE Computer Society CY - Houston, Texas, USA PY - 2005/ M2 - VL - IS - SP - 378 EP - 385 UR - http://doi.ieeecomputersociety.org/10.1109/ICDM.2005.46 M3 - KW - recommender KW - learning KW - supervised L1 - SN - 0-7695-2278-5 N1 - N1 - AB - ER - TY - RPRT AU - Zhu, Xiaojin A2 - T1 - Semi-Supervised Learning Literature Survey PB - Computer Sciences, University of Wisconsin-Madison AD - PY - 2005/ VL - IS - 1530 SP - EP - UR - http://pages.cs.wisc.edu/~jerryzhu/pub/ssl_survey.pdf M3 - KW - clustering KW - learning KW - semi KW - supervised KW - toread L1 - N1 - N1 - N1 - AB - ER - TY - JOUR AU - Basu, Sugato AU - Banerjee, Arindam AU - Mooney, Raymond J. T1 - Active Semi-Supervision for Pairwise Constrained Clustering JO - PY - 2004/04 VL - IS - SP - 333 EP - 344 UR - http://www.cs.utexas.edu/users/ml/papers/semi-sdm-04.pdf M3 - KW - active KW - clustering KW - semi KW - supervised L1 - SN - N1 - N1 - AB - Semi-supervised clustering uses a small amount of supervised
data to aid unsupervised learning. One typical approach
specifies a limited number of must-link and cannotlink
constraints between pairs of examples. This paper
presents a pairwise constrained clustering framework and a
new method for actively selecting informative pairwise constraints
to get improved clustering performance. The clustering
and active learning methods are both easily scalable
to large datasets, and can handle very high dimensional data.
Experimental and theoretical results confirm that this active
querying of pairwise constraints significantly improves the
accuracy of clustering when given a relatively small amount
of supervision. ER -