Resource Base of Ancient Chinese Books, National Library of China
- Designed and optimized the information retrieval system for literature resources
- Used Scrapy to extract online literature materials and established text indexing and the search engine framework through PyLucene, reducing search response time by 50%
- Optimized the search response time of the repository and improved the comprehensiveness and accuracy when searching with keywords in classical or early modern Chinese
Natural Language Processing API, BML (Baidu Machine Learning)
- Designed and optimized the information retrieval system for literature resourcesDeveloped visual modeling components and operators for Baidu’s full-featured AI development platform (BML)
- Improved the entity analysis operator, lexical analysis operator, and Chinese DNN language modeling operator to enhance the system for better language understanding, resulting in an average reduction in processing time of more than 70%