Data-driven integration of multiple sentiment dictionaries for lexicon-based sentiment classification of product reviews

A survey on opinion mining and sentiment analysis: Tasks, approaches and applications

“…The paper is concluded in section 5. Document level 73 [13], [18], [22], [32], [33], [36], [40], [43], [45], [48], [50], [51], [53], [54], [61], [64], [66], [77], [81], [80], [85], [88], [90], [91], [94], [96], [101], [111], [117], [121], [123], [130], [131], [132], [148], [155], [156], [157], [158], [167], [168], [169], [175], [176], [177], [179], [180], [182], [194], [195], …”

“…The learners in the second group of study are trained on source domain and tested on target domain [98,99]. These studies were carried out using lexicon based [36,98], machine learning based [121], and hybrid approaches [99].…”

“…These four thesauri were compared against three baseline methods and yielded better performance over all methods. Future work needs to ensure wide applicability in other domains as well.Cho et al[36] suggested cross-domain sentiment classification by integrating multiple sentiment dictionaries viz. WordNet-Affect[4], SentiWordNet[5,6], WordNet[133], Opinion Lexicon (OL)[38], AFINN[55], SO-CAL[69], Subjectivity Lexicon[104], General Inquirer[105], SenticNet[106], SentiSense[113], and Micro-WNOp[119] together.…”

A hybrid approach to the sentiment analysis problem at the sentence level

“…As per their creators [6], sentic computing "relies on the ensemble application of common-sense computing and the psychology of emotions to infer the conceptual and affective information associated with natural language." Other articles worth mentioning explore topics around sentiment lexicon-based techniques, like the contributions of Cho et al [14] and Huang et al [31]. The work by Bravo-Márquez et al [5], on the use of multiple techniques and tools in SA, offers a complete study on how several resources that "are focused on different sentiment scopes" can complement each other.…”

Survey on Tasks and Approaches of Sentiment Analysis

“…Seki et al [38] dealt with multi-lingual based opinion identification using the view of author. [39] used cross domain opinion analysis by using different dictionaries such as Word net affect [40], Senti word net [41], Word net [49], Opinion-lexicon [3], Sentic net [42], Senti sense [50] together. 5) Lexicon based polarity identification: Turney [33] have implemented two word phrase extraction using two pattern of tags.…”