Speaker
Description
Identifying news that affects financial markets is an important task on the way to predicting financial markets. A large number of articles are devoted to this topic. But the main problem for analyzing news is neural networks what used. These neural networks are created to analyze user reports about a particular object, be it a restaurant, a movie or a purchased item. In such reports, the emotional component of the report prevails, for which neural networks are trained. The use of these networks for the analysis of classic news texts does not give a positive result, articles are defined as neutral, which is often not true.
To solve the problem of analysis, you need to create your own dataset, on which to train neural networks. Tens of thousands of news are published daily, the processing of which is the processing of large volumes of distributed data. It is necessary to highlight the pool of those news that led to an increase in the volatility of financial instruments. This work describes methods of processing large volumes of distributed data in order to isolate a pool of news, as well as text corpuses that are present in this news and affect the volatility of financial instruments.
