The Layout of Sentiment Lexica in Word Vector Space

The Layout of Sentiment Lexica in Word Vector Space

Elena Razova (Vyatka State University, Russia) and Evgeny Kotelnikov (Vyatka State University, Russia)
DOI: 10.4018/978-1-7998-3038-2.ch015

Abstract

Sentiment lexicons play an important role in opinion mining systems and cognitive linguistics. Previous work aimed mostly at creating sentiment lexicons, but not thorough research into their fundamental properties. In this chapter, the arrangement of sentiment lexica in the multidimensional space of distributed word representations is studied. A hypothesis on the existence of sentiment lexica concentration areas is introduced, and it is tested on the basis of the joint analysis of the distribution of sentiment words and general lexica. The results of the test allow to confirm the proposed hypothesis for universal and domain-oriented sentiment lexicons. Also, the experiments discover the words which more than 80% of the sentiment lexica is concentrated around.
Chapter Preview
Top

In papers dedicated to the joint application of sentiment lexicons and distributed word representations, there are two main directions: creating dictionaries based on existing models (Blinov & Kotelnikov, 2014; Garten et al., 2016; Ito et al., 2017) and building models taking into account the characteristics of sentiment lexica (Hamilton et al., 2016; Tang et al., 2014; Vo & Zhang, 2016; Wang & Xia, 2017; Deng et al., 2019; Li et al., 2019).

Complete Chapter List

Search this Book:
Reset