Information technology has revolutionized data collection and analysis, offering unprecedented opportunities to study human behavior. Various information registers, the internet of things, and electronic demographic platforms that collect and analyze user data from various online sources provide a unique opportunity to predict human behavior using machine learning methods. This study applies machine learning to analyze textual data derived from diverse sources: demographic data, scientific articles, employee documents, and social media content. The primary goal is to identify a person's area of interest and predict their behavior. We propose using Support Vector Machines (SVM) as a robust and versatile machine learning algorithm for text data analysis. SVM's ability to handle diverse data types makes it well-suited for analyzing complex human behavior patterns. By classifying documents into relevant topics, SVM can help assess how employee behavior aligns with organizational goals and performance metrics. This research aims to contribute to human behavior analysis by demonstrating the effectiveness of machine learning techniques, particularly SVM, in extracting meaningful insights from textual data.

The development of Internet technologies influences the activities of libraries and changes their nature. The volume of content collected in digital libraries is growing rapidly. This requires the use of new technologies to search and obtain electronic materials (text, video, images, sound files) stored in the e-library. Today, using the new capabilities of network technologies and intelligent systems, the proper organization of the digital library, and increasing the efficiency of library services are the main factors leading to an increase in the number of readers and their satisfaction. The main objectives of digital libraries are to ensure efficient retrieval of electronic resources and collaboration between users. While researching various scientific articles on library and information sciences (LIS), we did not encounter approaches using cluster analysis in combination with wiki technologies. To collaborate users in digital libraries and their involvement in organizing electronic resources, we propose using an open database managed by wiki technologies. To effectively search for electronic resources in these open databases, it is proposed to use the clear clustering method. The clear clustering method also allows you to control the quality of clustering. The proposed method is important when creating intelligent (smart) libraries that are easy to manage and automate certain tasks. The research aims to create not just a smart library, but a smart library based on wiki technologies.