Information system for e-commerce analysis and customized data generation
Students Name: Rudakov Andrii Olehovych
Qualification Level: magister
Speciality: Computer Sciences
Institute: Institute of Computer Science and Information Technologies
Mode of Study: full
Academic Year: 2025-2026 н.р.
Language of Defence: англійська
Abstract: The relevance of the work is driven by the fact that small and medium-sized online stores, despite having significant volumes of transactional data, are mostly limited to standard platform reports and manual processing of CSV files. This does not provide sufficient analytical depth, formalized revenue forecasting, or safe scenario modelling. The aim of the study is to create a modular system which, based on the historical orders of an online store, implements a full data processing cycle: loading and cleaning transactions, generating descriptive analytics, building revenue forecasts, and producing synthetic datasets with configurable scenarios for the structure of demand and sales channels. The thesis presents a system analysis of the subject area and existing solutions (e-commerce reporting modules, BI systems, cloud ML services, synthetic data tools), formulates system requirements, and develops an architectural model, DFD and IDEF0 diagrams, UML use case diagrams, and an ER data model; the system’s logic is organized as a pipeline of modules for loading, descriptive analytics, forecasting, generation, and evaluation of synthetic data. The software implementation is carried out in Python using the pandas library (table processing), scikit-learn (forecasting), and matplotlib (visualization), while the data are stored in CSV format in separate directories for input and output files. Thus, the thesis combines approaches from electronic commerce, applied information system development, data analysis and revenue forecasting with the use of synthetic data, scenario modelling, and the Python programming language.