Textual analysis and machine leaning: Crack unstructured data in finance and accounting

• Main Authors: GUO, Li, SHI, Feng, TU, Jun
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2016
• Subjects: Machine learning; Textual analysis; Finance; Accounting; Media news; Sentiment; Information; Finance and Financial Management
• Online Access: https://ink.library.smu.edu.sg/lkcsb_research/5407; https://ink.library.smu.edu.sg/context/lkcsb_research/article/6406/viewcontent/1_s2.0_S2405918816300496_main.pdf
• Institution: Singapore Management University

In finance and accounting, relative to quantitative methods traditionally used, textual analysis becomes popular recently despite of its substantially less precise manner. In an overview of the literature, we describe various methods used in textual analysis, especially machine learning. By comparing their classification performance, we find that neural network outperforms many other machine learning techniques in classifying news category. Moreover, we highlight that there are many challenges left for future development of textual analysis, such as identifying multiple objects within one single document.