Construction of an Innovation Information Mining Platform Integrating Knowledge Graph
Deng Wan, Peisen Huang · 2025
With the continuous expansion of data mining scenarios, an accurate data analysis algorithm framework is urgently needed. In this study, the construction of an innovation information mining platform integrating knowledge graph is discussed. The designed platform combines multiple layers of knowledge fusion, matching and problem solving. This architecture can greatly improve the intelligent extraction and mining of innovative information. The designed platform mainly consists of four key parts: 1. At the data acquisition level, network logs are analyzed through Web data mining. At the same time, GoAccess tool is used to extract access logs and conduct preliminary analysis. 2. At the data mining level, this study uses a probabilistic graph model based on a word pair model to perform cluster analysis on large-scale text data. At the same time, the innovative information points are classified through calorific value calculation and information entropy theory to obtain the mining effect. 3. At the system improvement level, we integrate knowledge graphs and data mining algorithms, and use the Jena inference engine to reason and complete data relationships. 4. At the platform optimization level, the cloud platform architecture of this study further optimizes the allocation of computing resources and application deployment, while using Kubernetes and Docker container technologies to ensure the high performance of the system. Through data mining accuracy test, K-mean and FCM algorithms were used to verify the algorithm effect of this study.