Objectives: The present study aims to provide a framework for machine learning applications in organizational knowledge management at the Keshavarzi Bank, using the fuzzy network analysis process technique. Methods: The present study is applied in terms of the purpose and in terms of data collection, it is a descriptive survey. The statistical population includes IT and human resources development experts and managers of Keshavarzi Bank branches, from which 10 people have been selected through a purposive sampling method. Data collection tools are three questionnaires: Delphi, Dematel, and paired comparison questionnaire. Descriptive and inference methods were used to analyze data and information. After data collection, raw data is first encoded and classified and converted to research variables during the fuzzy Delphi technique. Then, using descriptive statistics, the index of the center and the dispersal of the sample members of the research were calculated, and used the fuzzy and fuzzy Delphi network analysis process to examine the criteria. In the analysis of the research data, Excel and SuperDecision software are used. Results: Eeffective criteria and sub-criteria with knowledge management in the banking system with machine learning approach in 4 general criteria and 17 sub-criteria were identified as follows: Infrastructural factors (Protection of personal information, hardware and software, communications, physical protection), environmental factors (organizational culture and atmosphere, organizational rules and regulations, knowledge delivery time, managerial support); Content factors (type of knowledge, specialized and knowledgeable human resources, knowledge presentation position, content development); Knowledge management process (knowledge creation, knowledge acquisition, knowledge conversion, knowledge application, knowledge needs assessment). In order to prioritize the indicators with the fuzzy network analysis technique, after calculating the limit supermatrix, the results of the cluster matrix results and the normalization of the coefficient of the sub-criteria in the limit supermatrix by the clustering coefficient are based on the calculations and the limit supermatrix, The final priority of infrastructure factors with a limit weight of 0.43683 have the most impact and content factors with a limit weight of 0.04817 have the least impact on maintaining knowledge management in the banking system by considering the machine learning approach. Conclusions: According to the obtained results, infrastructure factors have the most impact and content factors have the least impact on maintaining knowledge management in the banking system by considering the machine learning approach.