Opinion Question Answering (Opinion QA), which aims to find the authors ’ sentimental opinions on a specific target, is more challenging than traditional factbased question answering problems. To extract the opinion oriented answers, we need to consider both topic relevance and opinion sentiment issues. Current solutions to this problem are mostly ad-hoc combinations of question topic information and opinion information. In this paper, we propose an Opinion PageRank model and an Opinion HITS model to fully explore the information from different relations among questions and answers, answers and answers, and topics and opinions. By fully exploiting these relations, the experiment results show that our proposed algorithms outperform several state of the art baselines on benchmark data set. A gain of over 10 % in F scores is achieved as compared to many other systems. 1

In TAC 2008 we participated in the main task (Update Summarization) as well as the Sentiment Summarization pilot task. We modified the FastSum system (Schilder and Kondadadi, 2008) and added more aggressive filtering in order to adapt the system to update summarization and sentiment summarization. For the Update Summarization task, we show that a classifier that identifies sentences that are similar to typical first sentences of a news article improves the overall linguistic quality of the generated summaries. For the Sentiment Summarization pilot task, we use a simple sentiment classifier based on a gazetteer of positive and negative sentiment words derived from the General Inquirer and other sources to produce opinion-based summaries for a collection of blog posts given a set of positive and negative questions. 1

In this paper we begin to investigate how to automatically determine the subjectivity orientation of questions posted by real users in community question answering (CQA) portals. Subjective questions seek answers containing private states, such as personal opinion and experience. In contrast, objective questions request objective, verifiable information, often with support from reliable sources. Knowing the question orientation would be helpful not only for evaluating answers provided by users, but also for guiding the CQA engine to process questions more intelligently. Our experiments on Yahoo! Answers data show that our method exhibits promising performance.

Problem. Multi-document summarization saves time when dealing with large document quantities, but state of the art only handles fact summarization. What about sentiment (attitudes held by somebody about something)? Solution. Query-based Sentiment summarization, a form of (multi-document) summarization

As more and more people are expressing their opinions on the web in the form of weblogs (or blogs), research on the blogosphere is gaining popularity. As the outcome of this research, different natural language tools such as querybased opinion summarizers have been developed to mine and organize opinions on a particular event or entity in blog entries. However, the variety of blog posts and the informal style and structure of blog entries pose many difficulties for these natural language tools. In this paper, we identify and categorize errors which typically occur in opinion summarization from blog entries and compare blog entry summaries with traditional news text summaries based on these error types to quantify the differences between these two genres of texts for the purpose of summarization. For evaluation, we used summaries from participating systems of the TAC 2008 opinion summarization track and updated summarization track. Our results show that some errors are much more frequent to blog entries (e.g. topic irrelevant information) compared to news texts; while other error types, such as content overlap, seem to be comparable. These findings can be used to prioritize these error types and give clear indications as to where we should put effort to improve blog summarization.

An increasingly popular method for finding information online is via the

journal homepage: www.elsevier.com/locate/infosys