PROJECT OF INFORMATION SYSTEM FOR COLLECTION, PROCESSING AND DATA ANALYSIS

Abstract

The task of accurate and efficient weather forecasting in different geographical regions of the state is becoming increasingly important and in demand in various applied areas of human activity, including the agricultural, economic, construction, manufacturing, trade and tourism sectors of modern business. The aim of the article is to develop a project of an information system for the collection, processing and analysis of large-scale meteorological data. Realization of the set purpose provides the decision of a number of problems: 1) definition and the characteristic of specificity of construction of meteorological forecasts and processing of statistical data; 2) development of the concept of system operation and its architecture; 3) formalization and testing of data processing to identify strengths in the created information system. Scientific novelty. The article considers the possibility of using a number of machine learning algorithms, in particular models of artificial neural networks of different architectural types, to automate the process of solving the regression problem and when working with time series. In conclusion, the article emphasizes that the developed information system for the collection, processing and analysis of large-scale meteorological data implements a full cycle of functional processes for building weather forecasts for several days in advance. A significant advantage of the proposed system is its web orientation, which allows to make the final software product cross-platform, scalable and distributed software solutions for use by different types of users to perform research or application tasks for meteorological data processing. A further way to develop the system may be to add support for new types of source data on the studied objects on various indicators and the integration of electronic tools and platforms for direct data acquisition from sources (sensors and sensors).