Abstract:
Online product reviews have become eminent in the purchase decision-making process. With progress in web 2.0 technologies, huge volumes of unstructured text data are generated as reviews on e-commerce platforms and third-party web portals. Opinion review mining has become a critical area of research in language processing and applied machine learning. Opinion reviews available across various portals are perceived primarily to understand the sentiment polarity expressed by the reviewer at multiple granularities. The opinion review may also contain suggestions or tips for manufacturers and peer customers. Suggestion Mining refers to the automatic extraction of suggestions from opinionated text. The applications include product quality improvement, peer customer suggestions, summarizing collected surveys and feedback, a recommender system, and enhancing sentiment polarity classification. Suggestion mining is considered a sentence classification task, such as classifying a given review as suggestive intent or not. Various linguistic, syntactic, and semantic features with core machine learning and neural network approaches are used for suggestion mining. This paper presents a comprehensive and systematic review of suggestion mining from opinion reviews and their facets in the literature.
Page(s):
6061-6072
DOI:
DOI not available
Published:
Journal: Journal of Theoretical and Applied Information Technology, Volume: 100, Issue: 20, Year: 2022
Keywords:
deep learning
,
word embedding
,
Suggestion Mining
,
Opinion Reviews