A new graphic kernel method of stock price trend prediction based on financial news semantic and structural similarity

作者：

Highlights：

• Both contents and structures information in news text can help to stock price predicting.

• Proposed S&S kernel outperforms the other kernels by at least 5% on predicting accuracy.

• A higher weight is assigned to the structure instead of news contents.

• A clear inverse U-relationship between lag days and predicting accuracy can be found.

摘要

•Both contents and structures information in news text can help to stock price predicting.•Proposed S&S kernel outperforms the other kernels by at least 5% on predicting accuracy.•A higher weight is assigned to the structure instead of news contents.•A clear inverse U-relationship between lag days and predicting accuracy can be found.

论文关键词：Stock price movement prediction,Financial news,Information structure,S&S kernel

论文评审过程：Received 8 April 2018, Revised 31 August 2018, Accepted 3 October 2018, Available online 4 October 2018, Version of Record 18 October 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.10.008