Multi-model based Attention Mechanism for Stock Movement Prediction

DOI

10.2991/978-94-6463-042-8_183How to use a DOI?

Keywords

stock movement prediction; attention mechanism; deep learning; forecasting

Abstract

Rather than a pure random walk, the stock price changes in the manner of piecewise trend fluctuations. Predictions of stock's future movements have traditionally been based on prior trade data. With the rise of social media, many market participants are opting to make their tactics public, offering a window into the overall's attitude toward future developments by recovering the semantics underlying social media. Social media, on the other hand, includes contradictory information and cannot totally supplant the historical record. In this paper, we present a multimodal attention network that integrates semantic and numerical data to anticipate future stock movements and reduces conflict. We collect semantic information via social media and assess its reliability based on the publisher's name and public reputation. We next design trading strategies by combining semantics from online discussions with numerical characteristics from historical records. The results of our experiments reveal that our strategy exceeds earlier methods in terms of prediction accuracy and trading profit. It demonstrates that our strategy enhances stock movement forecasting performance and guides future multimodal fusion stock forecasting research.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Po-Wei Chen
AU  - Hao Yuan
AU  - Jiaying Huang
AU  - Po-Ju Chen
PY  - 2022
DA  - 2022/12/29
TI  - Multi-model based Attention Mechanism for Stock Movement Prediction
BT  - Proceedings of the 2022 International Conference on mathematical statistics and economic analysis (MSEA 2022)
PB  - Atlantis Press
SP  - 1278
EP  - 1282
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-042-8_183
DO  - 10.2991/978-94-6463-042-8_183
ID  - Chen2022
ER  -