Developing a concept-level knowledge base for sentiment analysis in\n Singlish

2017/07/14 by Rajiv Bajpai, Soujanya Poria, Bajpai, Rajiv +5
Computer Science · #Advanced Text Analysis Techniques #Computation and Language (cs.CL) #FOS: Computer and information sciences #Sentiment Analysis and Opinion Mining #Text and Document Classification Technologies

paper · pdf · doi:10.48550/arxiv.1707.04408

openalex publication_date 2017/07/14 · openalex created_date 2022/10/06 · openalex updated_date 2026/07/28

Abstract

In this paper, we present Singlish sentiment lexicon, a concept-level\nknowledge base for sentiment analysis that associates multiword expressions to\na set of emotion labels and a polarity value. Unlike many other sentiment\nanalysis resources, this lexicon is not built by manually labeling pieces of\nknowledge coming from general NLP resources such as WordNet or DBPedia.\nInstead, it is automatically constructed by applying graph-mining and\nmulti-dimensional scaling techniques on the affective common-sense knowledge\ncollected from three different sources. This knowledge is represented\nredundantly at three levels: semantic network, matrix, and vector space.\nSubsequently, the concepts are labeled by emotions and polarity through the\nensemble application of spreading activation, neural networks and an emotion\ncategorization model.\n

Related

• USA: Universal Sentiment Analysis Model & Construction of Japanese Sentiment Text Classification and Part of Speech Dataset
• Sentiment analysis using lexico-semantic features
• Representation Mapping: A Novel Approach to Generate High-Quality Multi-Lingual Emotion Lexicons
• Sentiment analysis is not solved! Assessing and probing sentiment\n classification
• Massively Multilingual Corpus of Sentiment Datasets and Multi-faceted Sentiment Classification Benchmark