A Review Text-based Recommendation System in Text Mining

Abstract

Digitization makes data more readily available, leading to data inflation in recent years. However, the large amount of data makes it difficult for users to find information that suits their needs. The recommendation system is an option to provide accurate information so that it can find and determine information based on needs. This paper proposed to identify and determine the dataset, methods, and objectives of text-based recommendations in Text Mining. Based on several search strategies, datasets mainly from user reviews such as Yelp, Amazon, TripAdvisor, and IMDb are the most used datasets. The recommendation system approach used is Latent Dirichlet Allocation (LDA), Machine Learning Approach, and Hybrid Recommendation to make product recommendations, travel, tourism, and several other recommendations. This paper can be used as a basis and comparison when other researchers want to develop a text-based recommendation system in Text Mining by considering the dataset, approach, and purpose of the recommendation.