Home // International Journal On Advances in Intelligent Systems, volume 15, numbers 3 and 4, 2022 // View article

Generating Market Comments on Stock Price Fluctuations Using Technical Analysis Features

Authors:
Ibuki Sekino
Minoru Sasaki

Keywords: Generate market comments; Stock price fluctuation; Nikkei Stock Average; Dow Jones Industrial Average; Technical Analysis;

Abstract:
Recently, there has been significant interest in techniques for generating market comments from stock prices automatically. However, it takes a multitude of time and effort for analysts to generate full-text market comments from stock prices. In this paper, we propose a method for generating "comments on stock price fluctuations" included in market comments to reduce the workload of analysts. The proposed method learns stock price fluctuations and the corresponding expressions, generates comments, and completes market condition comments by assigning them to prepared canned sentences. So far, this is the result of my previous study [1], and this paper improves the results by adding new features to the training data. The new data to be added to the training data in this paper are the "Dow Jones Industrial Average" and "Technical Analysis." The new data to be added to the training data in this paper are "Dow Jones Industrial Average" and "Technical Analysis", both of which are expected to improve the results. Because of our experiments, we found that the features used to generate them are effective and the proposed method can accurately generate market comments.

Pages: 83 to 92

Copyright: Copyright (c) to authors, 2022. Used with permission.

Publication date: December 31, 2022

Published in: journal

ISSN: 1942-2679