Using of Public Sentiment Gradient Analysis in Data Forecasting System

Abstract

Data forecasting data in financial markets is a critical task in today’s world. The ability to predict the market movements helps the investors avoid obvious risks and reduce potential losses. Various trading platforms have been developed to provide quick access to vast volumes of historical data, allowing real-time market analysis from anywhere in the world using just a laptop or personal computer. These platforms enable the creation of unique strategies based on fundamental or technical analysis, which take into account news about the specific companies, their profits, capitalization, and dividend payments.

News about different companies helps potential investors identify various risks, including personnel, production, and, most commonly in modern times, reputation risks. Therefore, textual news analysis plays a crucial role in forming fundamental analysis, which is most effectively conducted using neural networks.

Sentimental analysis by means of neural networks is a powerful tool for the markets forecasting as it enables to analyze and understand deep sentiments and emotions of the consumers and investors , based on textual information , such as feedback, social media , news, etc.

Over the past decade, due to technological innovations and advances in neural networks, these networks have become instrumental in analyzing large datasets, including textual data. As each news piece about a company targeted by potential investors or traders carries emotional sentiment, such as positive or negative, this sentiment can be determined using specially trained neural networks. This enables making accurate predictions in financial markets and developing effective strategies. When combined with technical analysis, the development and investigation of such an approach to forecasting can yield precise results. Hence, scientific research in this field remains relevant.

This article substantiates the sentiment analysis approach for forecasting historical data in financial markets, describes similar approaches, and outlines their advantages and drawbacks. Solutions are provided for a data forecasting system using selected sentiment analysis methods.