Development of a Corporate Knowledge Management System Using Machine Learning Techniques

Modern knowledge management systems are often essentially data libraries that differ only in the degree of structuring and processing methodology. From the point of view of management theory, the management process consists of such stages as analysis, goal setting, planning, implementation and control. In knowledge management, as a rule, only the analytical stage is presented. The process of management itself is limited by the difficulty of forecasting the processes of development of the system of corporate and public knowledge. In our study, we attempted to solve this problem by applying machine learning techniques. The term Machine Learning was used for the first time by Arthur Lee Samuel (1959). Machine learning is a class of artificial intelligence methods, the characteristic feature of which is not a direct solution to the problem, but training in the process of applying solutions to many similar problems. The science of machine learning itself studies methods for constructing algorithms capable of learning from various inputs. As part of this study, we will be interested only in learning with a teacher, which, in turn, is divided into the following subtasks: - Classification tasks; Regression tasks; Ranking tasks; Prediction tasks. The specific feature of learning with a teacher is that there is both a lot of data in which the model searches for patterns, and answers to the forecast of the model as part of its training. We have simulated the forecasting process in relation to the knowledge system of a company operating in the securities market. Machine learning methods have shown high efficiency in solving the entire range of tasks related to knowledge management.