Unsupervised product feature extraction for feature-oriented opinion determination

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

“…Non-ontological approaches can further be divided into topic modeling based [107,185,189,191,240] and non-topic modeling based approaches [25,37,59,67,186,190,193,218,241,243].…”

“…They proposed to augment the proposed method with clustering in order to extract explicit and implicit aspects and opinions. 37 www.cs.uic.edu/~liub/FBS/opinion-lexicon-English.rar Quan and Ren [67] coined a novel similarity measure, PMI-TFIDF, to identify association between products and its aspects. They performed aspect-level SA based on aspect-opinion pair extraction and aspect oriented opinion lexicon generation.…”

Hybrid optimization for feature selection in opinion mining

“…Less generic candidate features whose EDR score is less than threshold and domain-specific whose IDR is greater than another threshold is selected as opinion features. Point-wise Mutual Information (PMI)-TFIDF which is a point similarity measure is introduced by Quan& Ren [11] in evaluating the relevance of candidate features and domain entities. Outcome of the study shows that feature extraction approach performs better than other modern methods and comparative domain corpora remains the lone external resources used.…”

EXPRS: An extended pagerank method for product feature extraction from online consumer reviews

“…The three products included one digital camera, one mobile phone, and one DVD player, with 8901, 11,291, and 9,000 reviews, respectively. Those three product categories are the most commonly used ones in previous studies [3,36,41]. Digital camera and cell phone are search goods [32], while DVD player is experience goods [40].…”

“…As a fundamental and important step of consumer review analysis, feature extraction is aimed to identify product features commented in reviews automatically [11,36]. Existing approaches to feature extraction have several limitations.…”