Enriching SenticNet Polarity Scores through Semi-Supervised Fuzzy Clustering

Abstract

Abstract-SenticNet 1.0 is one of the most widely used freelyavailable resources for concept-level opinion mining, containing about 5700 common sense concepts and their corresponding polarity scores. Specific affective information associated to such concepts, however, is often desirable for tasks such as emotion recognition. In this work, we propose a method for assigning emotion labels to SenticNet concepts based on a semi-supervised classifier trained on WordNet-Affect emotion lists with features extracted from various lexical resources.

Dependency-Based Semantic Parsing for Concept-Level Text Analysis

“…Concept-level text analysis [24,26,25] focuses on a semantic analysis of text [12] through the use of web ontologies or semantic networks, which allow the aggregation of conceptual and affective information associated with natural language opinions. By relying on large semantic knowledge bases, such approaches step away from blind use of keywords and word co-occurrence count, but rather rely on the implicit features associated with natural language concepts.…”

Music Genre Classification: A Semi-supervised Approach

“…We followed a procedure suggested in [1] for a quite different task: effective classification of words [2]. The procedure consists in the following steps:…”

“…In this paper we propose to use for genre classification a methodology that was proven to work well in a similar situation: affective labeling of words in natural language texts, where, similarly, unlabeled texts abound but few words have a manually assigned affective label [1]. For brevity we refer to this methodology as semi-supervised learning, to emphasize that it uses two kinds of data: few labeled examples and a large quantity of unlabeled data; however, internally our two-step procedure works differently from a typical semi-supervised learner.…”

Sentiment Analysis in Turkish

“…Domain-Dependence: Words may have different meanings in different domains: for instance, the word "small" has a negative meaning in the hotel domain whereas it is in general positive in the cellphone domain. Since domain-independent lexicons such as SentiWordNet (Esuli and Sebastiani, 2006) and SenticNet (Poria et al, 2012) do not contain homonyms (a word that has diverse meanings in different contexts), they may mislead the sentiment analysis system. Hence, one may need a domain-specific lexicon which can be constructed by using a corpus of labeled reviews in a specific domain.…”

“…SentiWordnet (Esuli and Sebastiani, 2006) and SenticNet (Poria et al, 2012) are two of the most commonly used domain-independent polarity lexicons, for sentiment analysis. The lexicon-based approach obtains the polarities of the words or phrases in a document from a polarity lexicon, towards the goal of determining the semantic orientation of the document (Turney, 2002).…”