We introduce a method for learning to predict reader interest. In our approach, social interaction content and both syntactic and semantic features of words are utilized. The proposed method involves estimating topical interest preferences and determining the informativity between articles and their social content. In interest prediction, we integrate articles’ quality social feedback represent...

With the dawn of social network and its attractiveness, people are interested to share their experience, such as rating, reviews, etc. which helps to recommend the items of user interest. The potential growth of the internet results the use of social networks such as Face book, Twitter, linked-in etc. which produces huge amount of information (data), which leads to overwhelming. To overcome ove...

s should be submitted to Research Conference, Joint Center for Poverty Research, 2046 Sheridan Rd., Northwestern University, Evanston, IL 60208-4105. For further information, write to the same address or, by e-mail, to povcen @nwu.edu Order forms for Focus and other Institute publications are

Recommender systems have grown extremely in day to day life, and it is very useful in a plenty of applications. Although the recommender system is very popular, there are some serious problems like cold start (unavailability of data for modeling algorithms) and sparsity (initial rating not available). This made people to step back in analyzing the functionality of those algorithms which leads t...

Preferences for redistribution may be influenced by values and beliefs about distributive justice as well as by self-interest. People may prefer more redistribution to the poor if they believe that poverty is caused by circumstances beyond individual control. Therefore, beliefs about the causes of income may affect demand for redistribution. Alternatively, the effect of these beliefs on redistr...