Author(s)

Tao Wang

Corresponding Author

Tao Wang

Abstract

Abstract: Investor sentiment, as a core concept in the field of behavioral finance, has been widely studied by scholars. Through scholars' long-term efforts, investor sentiment indexes have undergone a development process from direct-type sentiment indexes to indirect-type sentiment indexes. Although sentiment indices are currently able to better portray investor sentiment, they also have some shortcomings. This paper focuses on how machine learning and text mining techniques can identify investor sentiment. Firstly, it introduces how to transform text into data that can be used for computer analysis through methods such as word separation and vectorization, followed by how to use machine learning methods to mine investor sentiment. A large amount of practice shows that machine learning methods are better applied in the field of investor sentiment.

Keywords

Keywords: Behavioral Finance; Sentiment Indexes; Machine Learning; Text Mining