Recognizing and Modeling Influence in Social Media Language
Abstract
Social influence has a profound impact on peoples emotion, opinion, and behavior. Prior work has exploited different information, from network analysis and the content. This project aims to focus on the language aspect in addition to social interaction to study influence. We have investigated the following problems in this reporting period:- predict review helpfulness. The review usefulness used in previous work has not been systematically tested. Second, whether a user rates a review as helpful depends on not only the review content but also the context in which it was written, such as whether the user agrees with the reviewer's opinion. Merely using features taken from the review limits a systems predictive power. Our work so far has focused on identifying these key research issues in existing work.- extract implicit aspects in reviews. So far we have found some characteristics of implicit aspects (they are common, overlap with explicit aspects ,functionality of opinions, attached to specific attributes) and just started data collection effort.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 23, 2018
- Accession Number
- AD1060777
Entities
People
- Yang Liu
Organizations
- University of Texas at Dallas