Sentiment Analysis: Does Coreference Matter?
نویسندگان
چکیده
We investigate the boost in sentiment performance by taking coreference information into account. We mine user generated content for market research. Frequent topic shifts in the data lead us to investigate sentiment algorithms which look at a window of text around topic key phrases. We describe and implement a lightweight sentiment analysis system and show how the system performance can be improved by about 10% when taking into account nominal and pronominal coreference elements.
منابع مشابه
Partially Supervised Coreference Resolution for Opinion Summarization through Structured Rule Learning
Combining fine-grained opinion information to produce opinion summaries is important for sentiment analysis applications. Toward that end, we tackle the problem of source coreference resolution – linking together source mentions that refer to the same entity. The partially supervised nature of the problem leads us to define and approach it as the novel problem of partially supervised clustering...
متن کاملTargeting Sentiment Expressions through Supervised Ranking of Linguistic Configurations
User generated content is extremely valuable for mining market intelligence because it is unsolicited. We study the problem of analyzing users’ sentiment and opinion in their blog, message board, etc. posts with respect to topics expressed as a search query. In the scenario we consider the matches of the search query terms are expanded through coreference and meronymy to produce a set of mentio...
متن کاملSentiment Analysis of Social Networking Data Using Categorized Dictionary
Sentiment analysis is the process of analyzing a person’s perception or belief about a particular subject matter. However, finding correct opinion or interest from multi-facet sentiment data is a tedious task. In this paper, a method to improve the sentiment accuracy by utilizing the concept of categorized dictionary for sentiment classification and analysis is proposed. A categorized dictiona...
متن کاملResolving Object and Attribute Coreference in Opinion Mining
Coreference resolution is a classic NLP problem and has been studied extensively by many researchers. Most existing studies, however, are generic in the sense that they are not focused on any specific text. In the past few years, opinion mining became a popular topic of research because of a wide range of applications. However, limited work has been done on coreference resolution in opinionated...
متن کاملSentiment Mining for Natural Language Documents
The wide variety of possible applications for sentiment mining has made it the focus of considerable research in recent years. High accuracy classification has been achieved by using a variety of techniques, most of which are heavily reliant on machine learning. At its core, sentiment mining involves correct interpretation of natural language, and all the complexities and challenges involved in...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2008