The Method for Generating Recommended Candidates through Prediction of Multi-Criteria Ratings Using CNN-BiLSTM
1
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
To improve the accuracy of the recommendation system, multi-criteria recommendation systems have beenwidely researched. However, it is highly complicated to extract the preferred features of users and items fromthe data. To this end, subjective indicators, which indicate a user’s priorities for personalized recommendations,should be derived. In this study, we propose a method for generating recommendation candidates by predictingmulti-criteria ratings from reviews and using them to derive user priorities. Using a deep learning model basedon convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM), multi-criteriaprediction ratings were derived from reviews. These ratings were then aggregated to form a linear regressionmodel to predict the overall rating. This model not only predicts the overall rating but also uses the trainingweights from the layers of the model as the user’s priority. Based on this, a new score matrix forrecommendation is derived by calculating the similarity between the user and the item according to the criteria,and an item suitable for the user is proposed. The experiment was conducted by collecting the actual“TripAdvisor” dataset. For performance evaluation, the proposed method was compared with a generalrecommendation system based on singular value decomposition. The results of the experiments demonstratethe high performance of the proposed method.
Abstract
To improve the accuracy of the recommendation system, multi-criteria recommendation systems have beenwidely researched. However, it is highly complicated to extract the preferred features of users and items fromthe data. To this end, subjective indicators, which indicate a user’s priorities for personalized recommendations,should be derived. In this study, we propose a method for generating recommendation candidates by predictingmulti-criteria ratings from reviews and using them to derive user priorities. Using a deep learning model basedon convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM), multi-criteriaprediction ratings were derived from reviews. These ratings were then aggregated to form a linear regressionmodel to predict the overall rating. This model not only predicts the overall rating but also uses the trainingweights from the layers of the model as the user’s priority. Based on this, a new score matrix forrecommendation is derived by calculating the similarity between the user and the item according to the criteria,and an item suitable for the user is proposed. The experiment was conducted by collecting the actual“TripAdvisor” dataset. For performance evaluation, the proposed method was compared with a generalrecommendation system based on singular value decomposition. The results of the experiments demonstratethe high performance of the proposed method.
참고문헌 (25)
- Phonexay Vilakone, "Personalized Movie Recommendation System Combining Data Mining with the k-Clique Method", JIPS(Journal of Information Processing Systems), 15(5), pp.1141~1155 (2019), 한국정보처리학회[DOI]
- V. Mohammadi, "Trust-based recommendation systems in Internet of Things : a systematic literature review", Human-centric Computing and Information Sciences, 9 (2019)
- M. Li, "A hybrid recommendation system for Q&A documents", Expert Systems with Applications, 144 (2020)
- N. Liu, "Attention-based sentiment reasoner for aspect-based sentiment analysis", Human-centric Computing and Information Sciences, 9 (2019)
- 송혜진, "User Modeling Using User Preference and User Life Pattern Based on Personal Bio Data and SNS Data", JIPS(Journal of Information Processing Systems), 15(3), pp.645~654 (2019), 한국정보처리학회[DOI]
- Z. Batmaz, "AE-MCCF : an autoencoder-based multi-criteria recommendation algorithm", Arabian Journal for Science and Engineering, 44(11), pp.9235~9247 (2019)
- M. Nilashi, "Hybrid recommendation approaches for multi-criteria collaborative filtering", Expert Systems with Applications, 41(8), pp.3879~3900 (2014)
- S. M. Al-Ghuribi, "Multi-criteria review-based recommender system : the state of the art", IEEE Access, 7, pp.169446~169468 (2019)
- A. Ebadi, "A hybrid multi-criteria hotel recommender system using explicit and implicit feedbacks", International Journal of Computer and Information Engineering, 10(8), pp.1377~1385 (2016)
- S. N. Alotaibi, "Improving collaborative filtering recommender by using multi-criteria rating and implicit social networks to recommend research papers", Ph.D, pp.2016
- M. K. Najafabadi, "Improving the accuracy of collaborative filtering recommendations using clustering and association rules mining on implicit data", Computers in Human Behavior, 67, pp.113~128 (2017)
- G. Adomavicius, "New recommendation techniques for multicriteria rating systems", IEEE Intelligent Systems, 22(3), pp.48~55 (2007)
- D. Y. Shee, "Multi-criteria evaluation of the web-based e-learning system : a methodology based on learner satisfaction and its applications", Computers & Education, 50(3), pp.894~905 (2008)
- N. Nassar, "A novel deep multi-criteria collaborative filtering model for recommendation system", Knowledge-Based Systems, 187 (2020)
- M. Hassan, "A neural networks approach for improving the accuracy of multi-criteria recommender systems", Applied Sciences, 7(9) (2017)
- 김진아, "Rating and Comments Mining Using TF-IDF and SO-PMI for Improved Priority Ratings", KSII Transactions on Internet and Information Systems, 13(11), pp.5321~5334 (2019), 한국인터넷정보학회
- J. C. De Albornoz, "Advances in Information Retrieval", pp.55~66 (2011), Springer
- Y. Zhang, "Explicit factor models for explainable recommendation based on phrase-level sentiment analysis", Proceedings of the 37th International ACM SIGIR Conference on Research & Development in Information Retrieval, pp.83~92 (2014)
- Y. Luan, "Research on text classification based on CNN and LSTM", Proceedings of 2019 IEEE International Conference on Artificial Intelligence and Computer Applications (ICAICA), pp.352~355 (2019)
- 박호연, "CNN-LSTM 조합모델을 이용한 영화리뷰 감성분석", 지능정보연구, 25(4), pp.141~154 (2019), 한국지능정보시스템학회
- A. Yenter, "Deep CNN-LSTM with combined kernels from multiple branches for IMDb review sentiment analysis", Proceedings of 2017 IEEE 8th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference (UEMCON), pp.540~546 (2017)
- X. Wang, "Combination of convolutional and recurrent neural network for sentiment analysis of short texts", Proceedings of the 26th International Conference on Computational Linguistics:Technical Papers (COLING), pp.2428~2437 (2016)
- H. W. An, "Design of recommendation system for tourist spot using sentiment analysis based on CNN-LSTM", Journal of Ambient Intelligence and Humanized Computing (2019)
- J. Kim, "BiLSTM model based on multivariate time series data in multiple field for forecasting trading area", Journal of Ambient Intelligence and Humanized Computing (2019)
- H. Wang, "Latent aspect rating analysis without aspect keyword supervision", Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.618~626 (2011)