Date and LocationNov 27, 2017 - 2:00pm to 3:00pm
Learning from Labeled and Unlabeled Vertices in Networks (presented by Andrew Huang, Computer Science Department)
Ye, W., Zhou, L., Mautz, D., Plant, C., & Böhm, C. (2017, August). Learning from Labeled and Unlabeled Vertices in Networks. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1265-1274). ACM.
Networks such as social networks, citation networks, protein-protein interaction networks, etc., are prevalent in real world. However, only very few vertices have labels compared to large amounts of unlabeled vertices. For example, in social networks, not every user provides his/her profile information such as the personal interests which are relevant for targeted advertising. Can we leverage the limited user information and friendship network wisely to infer the labels of unlabeled users?
In this paper, we propose a semi-supervised learning framework called weighted-vote Geometric Neighbor classier (wvGN) to infer the likely labels of unlabeled vertices in sparsely labeled networks. wvGN exploits random walks to explore not only local but also global neighborhood information of a vertex. en the label of the vertex is determined by the accumulated local and global neighborhood information. Specifically, wvGN optimizes a proposed objective function by a search strategy which is based on the gradient and coordinate descent methods. e search strategy iteratively conducts a coarse search and a ne search to escape from local optima. Extensive experiments on various synthetic and real-world data verify the effectiveness of wvGN compared to state-of-the-art approaches.
GloVe: Global Vectors for Word Representation (presented by Magzhan Zholbaryssov, Computer Science Department)
Pennington, J., Socher, R., & Manning, C. (2014). Glove: Global vectors for word representation. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP) (pp. 1532-1543).
Recent methods for learning vector space representations of words have succeeded in capturing fine-grained semantic and syntactic regularities using vector arithmetic, but the origin of these regularities has remained opaque. We analyze and make explicit the model properties needed for such regularities to emerge in word vectors. The result is a new global log-bilinear regression model that combines the advantages of the two major model families in the literature: global matrix factorization and local context window methods. Our model efficiently leverages statistical information by training only on the nonzero elements in a word-word co-occurrence matrix, rather than on the entire sparse matrix or on individual context windows in a large corpus. The model produces a vector space with meaningful substructure, as evidenced by its performance of 75% on a recent word analogy task. It also outperforms related models on similarity tasks and named entity recognition.