Construction of a data-driven knowledge discovery path from the perspective of intelligent academic evaluation: an empirical study based on institutional repository and solr technology

To address the problems of low efficiency, long turnaround cycles and difficulty integrating data in manual statistics of traditional academic evaluation (especially for professional title evaluation and award assessment), this study constructs a data-driven intelligent academic evaluation support system based on an Institutional Repository (IR). The system integrates dynamic data and a configurable rule engine to enhance the automation, standardization and intelligence of academic evaluation workflows. It provides accurate data support for academic evaluation, and enables knowledge discovery and efficient management throughout the academic evaluation process. By systematically analyzing the bottlenecks in manual statistics, the “Intelligent Academic Evaluation” system was designed and implemented. The system adopts the Java Web hierarchical architecture and integrates Solr indexing technology (with comparative justification of search technology selection) to achieve dynamic matching of multi-source heterogeneous data, providing efficient data retrieval support for intelligent evaluation.It combines the Spring Boot framework with the Redis caching mechanism to improve data processing performance; We have developed functional modules such as flexible configuration of academic evaluation conditions, qualification review, and batch export of reports, covering the entire process of achievement collection, statistics, and review, providing functional guarantees for the implementation of intelligent evaluation.Meanwhile, in combination with the actual business scenarios of a certain scientific research institution, the corresponding modules were deployed in the institutional knowledge base system and empirical verification was carried out. After verification, the system has achieved the following main results: (1)The professional title evaluation cycle has been shortened by approximately 70%, and the statistical accuracy rate has been increased to 98%, significantly reducing the cost of manual intervention in academic evaluation; (2)The system supports custom reward schemes and automatic report generation to achieve structured output of evaluation data; (3)The system delivers a second-level response to millions of academic literature records (supported by stress test data) to ensure the efficient operation of academic evaluations. Empirical results show that this system can significantly enhance the efficiency and quality of academic evaluation, providing a replicable technical pathway for universities and research institutions to implement a data-driven intelligent academic evaluation system.Limitations such as data quality dependence and scalability under high concurrency are discussed, and future improvement directions are proposed.