Identify the Sentiment Strength of Words in MicroBlog

蔡佳侑(Chia-Yu Tsai)

指導教授 ： 陳宜欣

國立清華大學/電機資訊學院/資訊系統與應用研究所/碩士(2010年)

摘要

在本研究中，我們利用微網誌所提供的四個資訊：驚嘆號、表情符號、重複情緒字、重複字母來分析每個情緒字所隱含的情緒強度。當使用者在微網誌中使用情緒字來表達個人對於某個事物的觀感時，經常會使用這四個資訊來傳達其所隱含的情緒強度。在本研究中，我們首先將每則微網誌訊息中的情緒字及四個相對應的情緒強度資訊擷取出來，形成字與強度資訊的陣列。透過基因演算法的學習方式，針對每個情緒字找出一個最佳的參數，透過參數的設定來計算出每個情緒字在微網誌中所隱含的強度。在實驗中，我們蒐集使用者的主觀意見，比較實驗結果與使用者的主觀意見使否符合。情緒強度可以用來提供使用者在進行決策時一個重要的參考依據，同時也提供情緒研究上另一個層面的思考。實驗結果顯示，透過微網誌的四個資訊，可以判斷出每個字所隱含的情緒強度，這些強度和使用者的主觀認定相符合。

關鍵字

情緒分析 ； 微網誌 ； 情緒強度

並列摘要

MicroBlogs have become a popular and important communication platform for Internet users. Many people frequently use the platform to share opinions about different topics. Although a few research papers have employed the use of MicroBlogs for sentiment analysis, the focus has been limited to classifying user sentiment into positive and negative classes. In this paper, we introduce an approach to identify the sentiment strength of words in MicroBlogs. Given a sentiment word, we extract the corresponding strength attributes from MicroBlogs: repeated exclamations, emoticons, repeated words, and repeated characters. By integrating these attributes with genetic algorithms, comparable word strength is provided. The experimental results demonstrate that the proposed approach satisfies the expectations of users and effectively provides comparable metrics on the sentiment strength of words.

並列關鍵字

Sentiment Analysis ； MicroBlog Analysis ； Sentiment Strength Analysis

參考文獻

[4] V. Hatzivassiloglou and K. McKeown. Predicting the semantic orientation of adjec-

ceeding of the Fourth International AAAI Conference on Weblogs and Social Media,

for Computational Linguistics, page 181. Association for Computational Linguistics,

[5] V. Hatzivassiloglou and J. Wiebe. Effects of adjective orientation and gradability

on sentence subjectivity. In Proceedings of the 18th conference on Computational