Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A multi-label ensemble predicting model to service recommendation from social media contents

A multi-label ensemble predicting model to service recommendation from social media contents Consumer sentiment is one of the essential measures of predictive recommendations in travel and tourism. Nowadays, a massive amount of data is available on the online platform related to consumer sentiment, which may help draw insights into how consumers provide feedback and how we can use that feedback to predict recommendations using machine learning techniques. In this study, we have designed a predictive recommendations method that predicts the consumer recommendations in travel and tourism, particularly in the case of the airline. We developed our predictive methods as a multi-label classification system. We implemented K-Nearest Neighbors, Support Vector Machine, Multi-layer Perceptron, Logistic Regression, Random Forest, and Ensemble Learning as basic classification models to train our model. Further, we boosted our predictive model by implementing state-of-the-art partitioning methods to partition the label space in lower spaces and utilized label space partitioning approaches, namely RAkELo and Louvain, as a transformation technique to transform every label set into a multi-class classification problem. The suggested model obtained higher performance in terms of accuracy using various evaluation measures compared to other binary classifications. Furthermore, the valuable traveler may get benefited from this approach in making their predictive decisions before travel. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journal of Supercomputing Springer Journals

A multi-label ensemble predicting model to service recommendation from social media contents

Loading next page...
 
/lp/springer-journals/a-multi-label-ensemble-predicting-model-to-service-recommendation-from-izLhg0zur8

References (51)

Publisher
Springer Journals
Copyright
Copyright © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021
ISSN
0920-8542
eISSN
1573-0484
DOI
10.1007/s11227-021-04087-7
Publisher site
See Article on Publisher Site

Abstract

Consumer sentiment is one of the essential measures of predictive recommendations in travel and tourism. Nowadays, a massive amount of data is available on the online platform related to consumer sentiment, which may help draw insights into how consumers provide feedback and how we can use that feedback to predict recommendations using machine learning techniques. In this study, we have designed a predictive recommendations method that predicts the consumer recommendations in travel and tourism, particularly in the case of the airline. We developed our predictive methods as a multi-label classification system. We implemented K-Nearest Neighbors, Support Vector Machine, Multi-layer Perceptron, Logistic Regression, Random Forest, and Ensemble Learning as basic classification models to train our model. Further, we boosted our predictive model by implementing state-of-the-art partitioning methods to partition the label space in lower spaces and utilized label space partitioning approaches, namely RAkELo and Louvain, as a transformation technique to transform every label set into a multi-class classification problem. The suggested model obtained higher performance in terms of accuracy using various evaluation measures compared to other binary classifications. Furthermore, the valuable traveler may get benefited from this approach in making their predictive decisions before travel.

Journal

The Journal of SupercomputingSpringer Journals

Published: Mar 1, 2022

Keywords: Ensemble model; Multi-label classification; Sentiment analysis; Consumer recommendations prediction; Natural language processing

There are no references for this article.