Arjumand Younus

What Do the Average Twitterers Say: A Twitter Model for Public Opinion Analysis in the Face of Major Political Events

Social media platforms have become a forum for giving a voice to the masses with a significant proportion of those masses coming from the developing world. This was largely evidenced through the significant role played by social media platforms in the recent uprisings in the Arab world. In this paper, we take up a study of social media engagement patterns of users from the developing world through a study of Twitters role during the recent Tunisian uprising. Motivated by the results of a user survey conducted mainly for users from the developing world who tweeted heavily during the uprisings in the Arab world, we propose a novel method for subjectivity analysis of tweets corresponding to political events in the developing world. Our proposed method differs from previous subjectivity analysis approaches in that it is the first method that takes into account social features of social media platforms for the subjectivity classification task. Through experimental evaluations, we observe the accuracy of the proposed method to be 83.3% which demonstrates a promising outcome for large-scale application of our proposed subjectivity analysis technique.

A Language Modeling Approach to Personalized Search Based on Users' Microblog Behavior

Personalized Web search offers a promising solution to the task of user-tailored information-seeking, and particularly in cases where the same query may represent diverse information needs. A significant component of any Web search personalization model is the means with which to model a users interests and preferences to build what is termed as a user profile. This work explores the use of the Twitter microblog network as a source of user profile construction for Web search personalization. We propose a statistical language modeling approach taking into account various features of a users Twitter network. The richness of the Web search personalization model leads to significant performance improvements in retrieval accuracy. Furthermore, the model is extended to include a similarity measure which further improves search engine performance.

Discovering Irrelevance in the Blogosphere through Blog Search

Web 2.0 technologies have given birth to the blogosphere, which is an information sharing medium by the users for the users. Furthermore, these technologies have also expanded the search problem to a new form of search known as blog search. Similar to Web search, blog search has been affected by spam which affects the quality of search results. This paper approaches the relevant blog problem in the top search results against the general topic queries. It pursues a study of irrelevant blogs appearing in the top search results of Google Blog Search for the blog spot domains. We define metrics for irrelevant blogs by observing the qualitative relevance of content and by analyzing the link structure of those blogs. Our preliminary results show an overall recall of 0.875 with a precision of 1.0 for finding irrelevant blogs in the top 15 search results against six general topic queries on Google Blog Search.