A Proposal for Extraction Method of Emotional Expression that Depended on Clusters  

Oleh:

Kaneishi, Shougo ; Tajima, Takuya

Jenis: Article from Proceeding
Dalam koleksi: The 14th Asia Pacific Industrial Engineering and Management Systems Conference (APIEMS), 3-6 December 2013 Cebu, Philippines, page 1-6.
Topik: Natural Language Processing; Context Analysis; Text Mining
Fulltext: 1187_Oyabu.pdf (182.15KB)

Isi artikel

A purpose of this study is to understand sensitivity expression depending on clusters. Communications by text are increasing. In addition, users of Social Networking Service are increasing too. The communications with the Social Networking Service are carried out in short sentences. On the one hand, Misunderstandings are often happened from the communications of short sentences. The causes of these misunderstandings are that we skip defined words in our community, we change words to fewer words. On the other hand, we use context which we can understand only within friends, work offices, hobby groups. In existing study to interpret the context correctly, Researchers attempted to guess omitted words from the context in conversations. However there is no possibility to know what they want to say or what is summarized from no existing information. In this paper, we propose an emotion extraction method that depended on clusters. In this proposal, we explored processes to produce emotion lexicon collecting emotional expressions that depended on clusters. We checked usefulness of evaluation information on blog websites. The users of blog websites are classified based on different evaluationpolarity by the holder. Finally, we report evaluation expressions in the lexicon collected by these processes. Reputations of products and word of mouth in blog website will be understood exactly by results of this study.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali