Kiduk et al. ICWSM 2007
This a Paper discussed in Social Media Analysis 11-772 in Autumn 2012.
Citation
Fusion Approach to Finding opinions In Blogosphere Kiduk Yang, Ning Yu, Alejandro Valerio Hui Zhang and Weimao Ke. ICWSM'2007.
Online version
Fusion Approach to Finding opinions In Blogosphere
Summary
This paper describe a fusion approach to finding opinion about a given target in blog postings. They opinion retrieval approach was to first apply traditional IR methods to retrieve on-topic blogs, and then boost the ranks of opinionated blogs using combined opinion scores generated by four opinion assessment methods. They opinion module consists of Opinion Term Module, which identify opinions based on the frequency of opinion terms (i.e., terms that only occur frequently in opinion blogs), Rare Term Module, which uses uncommon/rare terms (e.g., “sooo good”) for opinion classification, IU Module, which uses IU (I and you) collocations, and Adjective-Verb Module, which uses computational linguistics’ distribution similarity approach to learn the subjective language from training data. They use their system to participate TREC-2006 blog track and got the best results.
They test on TREC Blogs06 dataset.
Evaluation
The main performance evaluation metric for blog opinion retrieval task is mean average precision (MAP), which is the sum of precision at rank where relevant item is tetrieved averaged over topics. Mean R-precision(MRP), which is the precision at rank same as the total number of relevant items averaged over topics and precision at rank N were also used to evaluate the system performances.
Discussion
In most situation, fusion of different approaches helps to improve the performance of the whole system, yet the reason behind this is still not clear. It would be great if we can have some theoretical explanation about this.
