Scholarlink

<p>A recommendation system plays a pivotal role in assisting users by providing personalized suggestions based on their preferences. However, these systems often neglect the crucial aspect of user sentiment. To address this limitation, this work focuses on integrating sentiment analysis into recommendation systems. Using a dataset of 5000 reviews from Yelp, we employ preprocessing tasks such as Stop Words Removal, Tokenization, Part-of-Speech Tagging, WordNet Part-of-Speech Information, and Lemmatization to facilitate sentiment analysis. Leveraging the power of BERT, our sentiment analysis model achieves an impressive accuracy rate of 91%. We then incorporate the sentiment analysis component into collaborative filtering, utilizing cosine similarity and significantly reducing the root mean square error (RMSE) to 1.2478. Additionally, we integrate sentiment analysis into content-based recommendation using clustering, resulting in improved recommendations with a higher silhouette score of 0.2270. To further enhance the system’s performance, we propose a novel approach that combines these sentiment-aware components using NMF with DecisionTreeRegressor, resulting in an even lower RMSE of 1.1955. This integration of sentiment analysis into the recommendation system demonstrates its effectiveness in improving accuracy and personalization, providing users with more meaningful and relevant recommendations based on their sentiments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Enhancing recommendation systems with collaborative filtering and sentiment analysis: dimensionality reduction for improved content-based approaches

  • 期刊论文

  • 发表日期：

    2025年5月5日

查看全文
  • Nossayba Darraz,
  • Ikram Karabila,
  • Anas El-Ansari,
  • Nabil Alami,
  • Mostafa El Mallahi

摘要

A recommendation system plays a pivotal role in assisting users by providing personalized suggestions based on their preferences. However, these systems often neglect the crucial aspect of user sentiment. To address this limitation, this work focuses on integrating sentiment analysis into recommendation systems. Using a dataset of 5000 reviews from Yelp, we employ preprocessing tasks such as Stop Words Removal, Tokenization, Part-of-Speech Tagging, WordNet Part-of-Speech Information, and Lemmatization to facilitate sentiment analysis. Leveraging the power of BERT, our sentiment analysis model achieves an impressive accuracy rate of 91%. We then incorporate the sentiment analysis component into collaborative filtering, utilizing cosine similarity and significantly reducing the root mean square error (RMSE) to 1.2478. Additionally, we integrate sentiment analysis into content-based recommendation using clustering, resulting in improved recommendations with a higher silhouette score of 0.2270. To further enhance the system’s performance, we propose a novel approach that combines these sentiment-aware components using NMF with DecisionTreeRegressor, resulting in an even lower RMSE of 1.1955. This integration of sentiment analysis into the recommendation system demonstrates its effectiveness in improving accuracy and personalization, providing users with more meaningful and relevant recommendations based on their sentiments.