6 January 2020 in Journal of Big Data
Sentiment analysis is recognized as one of the most important sub-areas in Natural Language Processing (NLP) research, where understanding implicit or explicit sentiments expressed in social media contents is valuable to customers, business owners, and other stakeholders. Researchers have recognized that the generic sentiments extracted from the textual contents are inadequate, thus, Aspect Based Sentiment Analysis (ABSA) was coined to capture aspect sentiments expressed toward specific review aspects. Existing ABSA methods not only treat the analytical problem as single-label classification that requires a fairly large amount of labelled data for model training purposes, but also underestimate the entity aspects that are independent of certain sentiments. In this study, we propose a transfer learning based approach tackling the aforementioned shortcomings of existing ABSA methods. Firstly, the proposed approach extends the ABSA methods with multi-label classification capabilities. Secondly, we propose an advanced sentiment analysis method, namely Aspect Enhanced Sentiment Analysis (AESA) to classify text into sentiment classes with consideration of the entity aspects. Thirdly, we extend two state-of-the-art transfer learning models as the analytical vehicles of multi-label ABSA and AESA tasks. We design an experiment that includes data from different domains to extensively evaluate the proposed approach. The empirical results undoubtedly exhibit that the proposed approach outperform all the baseline approaches.
26 September 2022 in Information Systems Frontiers
2 September 2020 in IEEE Transactions on Engineering Management
Business closure is a critical stage in the lifecycle of any business. Despite a body of literature on the factors influencing business failure and/or success, studies on business closure prediction are next to none. Additionally, the identification of those factors has focused primarily on the managerial and economic dimensions rather than the consumer dimension. This research investigates the prediction of business closure from online consumer reviews, and validates the predictive models in the restaurant industry, which is subject to a relatively high attrition rate. Our proposed method for predicting business closure includes several novel artifacts, including integrating deep learning and time-series analysis techniques, extracting information embedded in online reviews using a hybrid classification method, and incorporating a novel triple word embedding model for text representation. The evaluation results of the proposed method using Yelp online reviews demonstrate its superior performance in business closure prediction, which indicate that online reviews provide strong signals for predicting business closure. We conduct another experiment with review data collected from TripAdvisors.com, and the experiment results are consistent, which provide evidence for the generalizability of the proposed method. The findings of this research have important managerial implications for business and investment decision making.
18 March 2022 in Journal of the Association for Information Science and Technology
9 June 2015 in Information Systems Frontiers
1 October 2015 in Future Generation Computer Systems