AISWare KG helps to answer operation and maintenance questions intelligently

Customer requirement

Based on work orders such as complaint work orders, problem orders and event orders, the data is scattered in the process and logs of each system, and the confirmation of faults and problems is not timely. Meanwhile, it is difficult to find data and confirm problems. Therefore, we hope to solve relevant problems with the help of knowledge graph.

Construction Plan

Introducing knowledge graph building technology, based on work order sample data such as complaining work order, question work order and events work order, we extract knowledge mining of historical data, then reason and sum it up to the knowledge of problem storehouse, to provide the reference for subsequent automation or artificial processing of similar problems, to support operational knowledge response ability, and to promote efficiency of fault handling.AISWare KG stores, displays, and thinks of data in a graphical structure to capture "treasures" hidden in the graph. Graph analysis mining algorithm is the core means of mining and discovering new relationships. By integrating a large number of data mining algorithms and the high quality algorithm provided by the capability mall, the threshold of using the algorithm is lowered, which makes graph data mining easy to understand and thus reduces personnel input.

Application Effect

  • Automatically identify

    Automatically identify the problems input of knowledge base, realize the automatic accumulation of IT operation and maintenance experience, and reduce the difficulty of knowledge precipitation.

  • Automatic processing

    Help with alarm processing and intelligent operation and maintenance assistant scene, realize automatic processing of associated knowledge of repeated problems, reduce repetitive work of operation and maintenance personnel, and improve its efficiency.

  • Intelligent recommendation

    Intelligent recommendation of question answers for operation and maintenance personnel to improve the efficiency of knowledge application and realize knowledge identification and recommendation related to faults.