Network Security Situation Cognition Ability Assessment Based on Commonsense Reasoning Calculation model

Konferenz: CAIBDA 2022 - 2nd International Conference on Artificial Intelligence, Big Data and Algorithms
17.06.2022 - 19.06.2022 in Nanjing, China

Tagungsband: CAIBDA 2022

Seiten: 6Sprache: EnglischTyp: PDF

Autoren:
Sun, Jing; Liu, Yifei; Zhang, Xiujian; Cheng, Zhonghao (Key Laboratory of Artificial Intelligence Measurement and Standards for State Market Regulation, Beijing Aerospace Institute for Metrology and Measurement Technology, Beijing, China)

Inhalt:
With the continuous evolution and development of information technology, the pattern of network threat attacks and the source of cyber security situation cognition are increasing, leading to more stringent speed and accuracy requirements for understanding and cognition of network security situation. The quality of the commonsense reasoning calculation model directly affects the intelligence processing and utilization capabilities of a cognitive system. Focusing on the comprehensive and accurate requirements of network security situation cognition, this paper researches on the evaluation technology of network security situation cognition ability based on commonsense reasoning calculation model. The construction method of the reasoning ability test dataset for the commonsense reasoning calculation model has been proposed. The multiple sets of commonsense test questions with different difficulty degree have been developed. The selection of commonsense test questions and the update of the test datasets have been studied. The research will provide theoretical support for further optimizing the commonsense reasoning calculation model, and improving the capabilities of situation element acquisition, situation assessment, experience learning and logical problem solving.