Research and Design of Data Security Risk Assessment Model Based on Fusion of Deep Learning and Analytic Hierarchy Process (AHP)

With the rapid development of information technology, data security risks are increasing, which not only brings unprecedented challenges to enterprises and individuals, but also makes it difficult for traditional risk assessment methods to effectively deal with increasingly complex security threats. Therefore, an innovative assessment model that can accurately identify, assess and deal with these security risks is urgently needed. This study proposes a new data security risk assessment model combining deep learning and analytic Hierarchy Process (AHP), which uses deep learning technology to automatically extract potential security risks from massive data and capture key risk factors through pattern recognition. Then, the analytic hierarchy process is used to build a multidimensional risk assessment framework, combining qualitative and quantitative analysis methods to comprehensively evaluate and rank risk factors, and ensure the scientific and accurate evaluation results through expert scoring and weight allocation. In order to further enhance the reliability of the evaluation, this study also introduces the fuzzy comprehensive evaluation method for comprehensive analysis, and optimizes the decision-making process by TOPSIS method to screen out the most suitable security reinforcement scheme. Experiments show that the proposed model runs stably in many practical scenarios, effectively improves the accuracy and operability of decision making, and verifies the effectiveness and practicability of the model. To sum up, this study not only provides innovative technical solutions for data security risk assessment, but also provides enterprises with more scientific and efficient decision support tools for dealing with complex security threats, which has important practical application value, and provides valuable theoretical basis and practical guidance for future research and model optimization.