Metaheuristic Optimization Review MOR 3066-280X 10.54216/MOR https://www.americaspg.com/journals/show/4197 2024 2024 Tashkent State University of Economics Tashkent, Uzbekistan Bahodir Bahodir Tashkent University of Information Technology, Uzbekistan Ziyoda Norqulova Today, information systems handle large volumes of data from various sources. These data may differ in both form and meaning. Such data diversity is one of the main problems in network integration and analysis. This research paper analyzes the main types of data: digital, integer, text, categorical, temporal, logical, and spatial. Today, information systems work with large volumes of information obtained from various sources. This data can differ in both form and meaning. This diversity of data is one of the main problems in the processes of integration and analysis in the network. This research paper analyzes the main types of data: digital, integer, text, categorical, temporal, logical, and spatial. For each type of data, a normalization approach is selected that corresponds to it. In particular, we will study the min-max scaling and Z-score standardization methods for digital data, one-hot and label encoding for category attributes, as well as lemmatization and normalization based on Unicode for text data. The analysis shows that choosing the right approach for each data type increases the efficiency of unification, ontological mapping, and visualization. The article analyzes the advantages and limitations of existing normalization methods and provides practical recommendations for selecting optimal methods for processing network data. The proposed approach can be effectively used in the processes of semantic integration of multi-source network data, as well as to its visual analysis. 2025 2025 53 61 10.54216/MOR.040206 https://www.americaspg.com/articleinfo/41/show/4197