Intelligent information system for trading on the exchange of construction materials
Students Name: Tsizh Ostap Tarasovych
Qualification Level: master (ESP)
Speciality: Information Systems and Technologies
Institute: Institute of Computer Science and Information Technologies
Mode of Study: full
Academic Year: 2020-2021 н.р.
Language of Defence: ukrainian
Abstract: Tsizh O.T., Ryshkovets Yu.V. (supervisor). Intelligent information system for trading on the exchange of construction materials Master’s thesis. - Lviv Polytechnic National University, Lviv, 2021. Extended abstract. The master’s qualification work is devoted to the development of an intelligent information system for the organization of trading on the exchange of construction materials. It provides the study and practical use of real-time data processing in large volumes, creating a fault-tolerant and accessible service. The main purpose of the work is to design and create a system for the purchase or sale of building materials, with real-time data updates for other users, as well as forecasting the price of goods based on historical data using an artificial intelligence module. The bidding system reflects the quantity and price of goods. The system allows users to purchase or sell building materials in bulk and view the historical dynamics of changes in the price of goods. The intellectual part, with the help of the artificial intelligence module of the system, analyzes historical data and can predict the price of the product in the future. The module is built on the basis of microservice architecture, and is not tied to a specific designed system of organization of trading on the exchange of construction materials. Thanks to this, the module can be easily integrated and used by many other systems, which makes it extremely relevant and useful, especially today, when stock markets are experiencing bad times. The research methods are system analysis and comparison of analog systems, construction and construction of a program in which certain third-party services are integrated and the possibility of integration with other platforms. One of the goals was to ensure that there were no delays in displaying the actual quantity of goods in the stock exchange, the ability to process large amounts of data, to improve scalability, and so on. Also a very important goal of the work was to design a separate intelligent module, the main function of which is to process historical data and calculate the possible so-called predicted price of the product. To select an algorithm for constructing the predicted value, the paper analyzes some possible variants of different algorithms. The methods of prediction of time series are considered and investigated in detail in the work, on the basis of which the algorithm of forecasting the price of the goods is developed. The developed artificial intelligence is based on the architecture of a recurrent neural network, in particular the LTSCM model. The server part of the system that processes the input data was built using the C # programming language and the ASP.NET framework. Entity Framework Core was used as an ORM database to store and process all data. On the client side, the default standard for the ASP.NET framework was Razor pages technology. During the development of Microsoft Visual Studio was used as an IDE. The system is created in the C # programming language. The use of a system of distributed independent isolated containers allowed to increase the speed of the application and made it possible to withstand heavy loads. The system architecture is built in such a way that the developer can expand the functionality and cover the software with tests in the future. Expansion methods include displaying more information about a particular product, such as adding a photo, improving a product pricing algorithm, and so on. In the future, the system can be improved and expanded to implement a wider range of tasks. It is worth noting that there are already many such solutions on the market and the number of analogues continues to grow, but there is always the possibility of developing towards optimal solutions for specific users - the so-called customization. The object of the study is the process of transactions on the trading exchange, as well as forecasting the price of goods. The subject of research is the methods and means of organizing trading on the exchange of construction materials. The purpose of this work is to develop an information system for the organization of trading on the stock exchange of construction materials, which would allow the buyer to find the product he needs, and the seller to sell their product. During the work the following tasks were set: 1) to design a conceptual model of the information system of the organization of trades on the exchange of construction materials; 2) develop the architecture of the exchange system; 3) develop an algorithm for changing the price of goods according to supply and demand for it; 4) develop a product price forecasting function based on artificial intelligence. The main problems of bidding are considered. The main advantages and disadvantages of the created system are determined. During the implementation, few services with similar functionality, advantages and disadvantages were analyzed. The developed system covers all the basic needs of its users so that they can track the price of goods in real time using a browser, and carry out their exchange activities. Confirmation of the concept of the system with the completed life cycle of IT project development is the result of this master’s thesis. Keywords: exchange, trade organization, commodity exchange, time series, value forecasting on the basis of historical data of presented magazines nearby, process optimization, real-time data processing, component program interface.