Utilizing emotion scores for improving classifier performance for predicting customer's intended ratings from social media posts

Ray, Arghya; Bala, Pradip Kumar; Jain, Rashmi

doi:10.1108/bij-01-2020-0004

Cited by 19 publications

(29 citation statements)

References 53 publications

(82 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An increasing body of research has been adopting the Tripadvisor platform as a useful source of customer feedback (e.g. Ray et al, 2020;Giglio et al, 2020).…”

Section: Data Collectionmentioning

confidence: 99%

How to use spontaneous customer comments to identify nonlinear background of satisfaction with restaurant services

Tontini

Santos

Kroenke

et al. 2021

BIJ

View full text Add to dashboard Cite

PurposeThe purpose of this study is to demonstrate how to use customer spontaneous comments to identify which aspects influence the overall customer satisfaction with restaurant services from a nonlinear perspective.Design/methodology/approachThe authors collected data from 399 spontaneous comments about a chain of fast-food restaurants in Brazil. The comments are freely available on the TripAdvisor portal and were extracted and classified according to seven dimensions related to the quality of services: tangibles, reliability, responsiveness, assurance, empathy, quality of the food and price. Next, the authors combine the critical incident technique (CIT) and the penalty-reward contrast analysis (PRCA) to investigate the nonlinear relationship between service quality assessment and overall customer satisfaction.FindingsThe method of integrating CIT with PRCA explains 64.7% of the variation in the customer's assessment of the services provided (RAdj2 = 0.647). This shows that spontaneous comments from customers are related to their overall satisfaction with the service provided. Besides, the findings suggest that consumers tend to comment more about positive than negative experiences regarding aspects related to food, attendants' empathy and service assurance, and more negative comments about aspects related to responsiveness and price. However, it was found that negative comments have a stronger influence on overall satisfaction than positive comments.Originality/valueUsing comments available for free on the Internet and evaluating how positive and negative comments can jointly influence customer satisfaction, the proposed methodology demonstrates how restaurants can use their customers' spontaneous comments to identify critical aspects to be managed and improved. To the best of authors’ knowledge, this is the first study presenting how restaurants can use customer spontaneous comments, freely available on the internet, to identify the relevance of different aspects of the services provided from a nonlinear perspective. In addition, the present study shows that although customers spontaneously tend to share more positive than negative comments about restaurant services, events related to negative experiences have a stronger influence on overall satisfaction.

show abstract

“…An increasing body of research has been adopting the Tripadvisor platform as a useful source of customer feedback (e.g. Ray et al, 2020;Giglio et al, 2020).…”

Section: Data Collectionmentioning

confidence: 99%

How to use spontaneous customer comments to identify nonlinear background of satisfaction with restaurant services

Tontini

Santos

Kroenke

et al. 2021

BIJ

View full text Add to dashboard Cite

show abstract

“…Topic modeling helps to create a topic-terms matrix and provides probability scores based on the concept that a topic is composed of several terms and several topics presented in a particular document (Ray & Bala, 2021 ). Topic modeling helps in performing a much deeper analysis of textual data through content analysis (Hagen, 2018 ), path analysis (Ray & Bala, 2019 ; Ray et al, 2020a , 2020b ), etc. For the quantitative analysis sentiment, emotional analysis, and NLP-SEM technique (Ray & Bala, 2021 ) was utilized on the online reviews.…”

Section: Methodsmentioning

confidence: 99%

“…help to understand customers’ feelings better (Cavanaugh et al, 2015). Ray et al, ( 2020a , 2020b ) found an increase in algorithm performance while considering emotional aspects present in online reviews for predicting ratings. Thus emotions can also help in understanding customer behaviour.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Although UGC mainly refers to online customer reviews, in this study the information posted in hotel websites by hotel managers is also referred as UGC. UGC can be analyzed using various techniques, like, word-frequency analysis (Taft, 1979 ), thematic analysis (Ray & Bala, 2021 ), sentiment and emotion analysis (Ray et al, 2020a , 2020b ), natural language processing based structural equation modeling (NLP-SEM) (Ray & Bala, 2021 ), etc. In this study, word frequency analysis, thematic analysis, sentiment and emotional analysis and NLP-SEM is used.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Operational Changes and Performance Outcomes: Analysis on Hotels of Five Asia–Pacific Countries

Ray

2021

JGBC

Self Cite

View full text Add to dashboard Cite

“…In order to create equal-sized training sets, each fold is split into two halves, which are then divided again. Gams [16] primarily depended on this strategy, which they further improved, in order to generate neural network ensembles [19,20]. To make matters even better, Domingos [17] employed cross-validation to speed up the development of his proposed rule induction system, known as RISE [21].…”

Section: X-validationmentioning

confidence: 99%

An Ensemble and Filtering-Based System for Predicting Educational Data Mining

Hananto¹

2021

J. Appl. Data Sci.

View full text Add to dashboard Cite

When developing a prediction paradigm, an ensemble technique such as boosting is used. It is built on a heuristic framework. Generally speaking, engineering ensemble learning is more accurate than individual classifiers when it comes to making predictions. Consequently, numerous ensemble strategies have been presented in this work, particularly to provide a more complete understanding of the essential methods in general. Researchers have experimented with boosting methods to forecast student performance as part of a variety of ensemble techniques. The researchers employed improvement approaches to construct an accurate predictive educational model, which was based on a key phenomena seen in categorization and prediction operations. In light of the uniqueness and originality of the suggested strategy in educational data mining, the researchers used augmentation strategies in order to construct an accurate predictive pedagogical model. Tenfold cross-validation was performed to evaluate the effectiveness of the basic classifiers, which included the random tree, the j48, the knn, and the Naive Bayes. The random tree was found to be the most effective classifier. Several additional screening techniques, including oversampling (SMOTE) and undersampling (Spread subsampling), were utilized to analyze any statistically significant variations in results between the meta and base classifiers that had been identified between the meta and base classifiers. The use of ensemble and screening strategies, as compared to the use of standard classifiers, has demonstrated considerable gains in predicting student performance, as has the use of either strategy alone. Furthermore, after the completion of a performance research on each approach, two new prediction models have been established on the basis of the improved results gained thus far.

show abstract

Utilizing emotion scores for improving classifier performance for predicting customer's intended ratings from social media posts

Cited by 19 publications

References 53 publications

How to use spontaneous customer comments to identify nonlinear background of satisfaction with restaurant services

How to use spontaneous customer comments to identify nonlinear background of satisfaction with restaurant services

Operational Changes and Performance Outcomes: Analysis on Hotels of Five Asia–Pacific Countries

An Ensemble and Filtering-Based System for Predicting Educational Data Mining

Contact Info

Product

Resources

About