Earticle

초록

영어

In this paper, we present an e-book recommendation system using topic-modeling based on Latent Dirichlet Allocation (LDA) that is a probabilistic model increasingly used in various textual data analysis. Through the generated topics, we found the latent themes of the topics by estimating probability distributions for the topics in eBooks and words. We also address the basic concept of micro-segmentation used mainly in customer marketing field, which ensures that a variety of eBooks are recommended to users. The primary aim of our proposed method is to integrate the effective and efficient techniques with only using textual data of eBooks to improve recommendation performance in Content-Based Filtering (CBF) recommendation when it is unable to rely on the Collaborative Filtering (CF) utilizing ratings and reviews data obtained from a user's own past information. The experiment demonstrates the robustness of the presented method, and also shows that the method provides explainable recommendation results.

목차

Abstract
1. Introduction
2. Topic Modeling based on LDA
3. Proposed System
3.1. System Architecture
3.2. Process of Proposed Method
4. Experiments
4.1. Evaluation
4.2. Results
5. Conclusion and Future Research Directions
References

키워드

저자

• Byounghee Kim [ Edulab AI Lab ]
• Jungah An [ Edulab AI Lab ]
• Bokyeong Kang [ Edulab AI Lab ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  한국경영정보학회 정기 학술대회 [KMIS Conference]

• 간기

  반년간

• 수록기간

  1990~2025

• 십진분류

  KDC 325 DDC 658