Intelligent Information Retrieval Module Based on Learning to Rank
Zhenyu Wang, Yangfei Zhen · 2020
With the rapid expansion of information on the internet nowadays, the query results produced by Relational Database (RDB) can no longer satisfy the complex search requests made by users. Traditional information retrieval modules generally have problems such as low retrieval efficiency, poor scalability, and low intelligence. In this paper, this paper proposed an intelligent information retrieval module based on the Learning to Rank (L TR) to optimizes the relevance score in the information retrieval process. Experimental results conducted in production environment show that our proposed method improved accuracy of relevance score by 15.84% and accelerate the retrieval speed by 4.15 times. In addition, the interpretability of the retrieval results and the selection of key features are also analyzed and realized.