Multinomial Relation Prediction in Social Data: A Dimension Reduction Approach

The recent popularization of social web services has made them one of the primary uses of the World Wide Web. An important concept in social web services is social actions such as making connections and communicating with others and adding annotations to web resources. Predicting social actions would improve many fundamental web applications, such as recommendations and web searches. One remarkable characteristic of social actions is that they involve multiple and heterogeneous objects such as users, documents, keywords, and locations. However, the high-dimensional property of such multinomial relations poses one fundamental challenge, that is, predicting multinomial relations with only a limited amount of data. In this paper, we propose a new multinomial relation prediction method, which is robust to data sparsity. We transform each instance of a multinomial relation into a set of binomial relations between the objects and the multinomial relation of the involved objects. We then app...

• Type: Journal
• Year: 2012
• Where: AAAI
• Authors: Nozomi Nori, Danushka Bollegala, Hisashi Kashima
• Keywords: AAAI 2012, Fundamental Web, Generalized Eigenvalue Problem, Global Optimal Solutions, Intelligent Agents