Dataset presented in this paper is obtained from the top online automobile selling and purchasing websites. A total of 1000 reviews related to hybrid cars in the form of text reviews are extracted with the help of the Web Scraper tool. The dataset presents the customers sentiments in the form of reviews related to hybrid cars. Various aspects are taken into consideration while annotating the reviews such as driving, performance, comfort, safety features, interior, exterior and accessories. The annotation of data is done at three levels by three annotators i.e., (1) overall polarity of a review, (2) segregation of the sentence term in which aspect is discussed, (3) polarity of the discussed aspect. Cohen's Kappa score of 0.90 was achieved among the authors while annotating the reviews. Dataset can be used for sentiment analysis, information retrieving, lexicon analysis, and grammatical and morphological analysis.
Keywords: Aspects; Natural language processing; Opinion mining; Sentiment analysis.
© 2022 The Authors. Published by Elsevier Inc.