Intellectual model of knowledge management in the conditions of the heterogeneity of information space

Data retrieval processes have shifted towards open processes with visualization and parameter setting and a predictive model. Data and models in hyperspace can be visualized for end users using popular data mining platforms. Numerous studies have shown how adjusting and even creating decision tree classifiers can help end users better understand the dataset and the context in which the data was collected. In order to use the possibilities of such an open approach, the article presents a method of extended intelligence, as well as a bioinspired algorithm based on the adaptive behavior of bats. This method will allow end users to analyze data in an iterative process. Based on the proposed method, knowledge discovery and the accuracy of the predictive model generated by the algorithm increase over time due to interactions between models and end users. The article describes methods of information extraction in data mining. An extended intelligence is described, including algorithms for machine learning and deep learning networks, as well as methods of rational and augmented machine learning, on the basis of which own data will be created, having a limited amount of information for training.