Building Lexicon for Sentiment Analysis from Massive Collection of HTML Documents

Recognizing polarity requires a list of polar words and phrases. For the purpose of building such lexicon automatically, a lot of studies have investigated (semi-) unsupervised method of learning polarity of words and phrases. In this paper, we explore to use structural clues that can extract polar sentences from Japanese HTML documents, and build lexicon from the extracted polar sentences. The key idea is to develop the structural clues so that it achieves extremely high precision at the cost of recall. In order to compensate for the low recall, we used massive collection of HTML documents. Thus, we could prepare enough polar sentence corpus.

• Type: Conference
• Year: 2007
• Where: EMNLP
• Authors: Nobuhiro Kaji, Masaru Kitsuregawa
• Keywords: EMNLP 2007, Natural Language Processing, Polar Sentence, Polar Words, Structural Clues