A Review of Research on the Application of Deep Learning in Library Text Classification
Text categorization is a major research direction in the field of library.Classifying user-generated content based on deep learning methods can help libraries understand user behavior more accurately and evaluate the quality of library services.This paper makes recommendations for future research by investigating and critically reviewing research on the use of deep learning methods for text classification in the library field and in the computer science field using library data.The results show that the current research mainly focuses on text feature classification,text sentiment classification and text rating classification.Most research still uses traditional deep learning methods such as feedforward neural networks and artificial neural networks.Deep learning algorithms with better classification performance in computer science have not been introduced to the library domain.The study recommends introducing a methodological framework for deep learning algorithms,building and developing deeper algorithms,clarifying detailed steps for text classification,specifying data annotation specifications and annotation steps,and actively using multi-model classification methods that combine their respective strengths.
万方数据
维普
