Presentation 2013-09-13
Prediction of Growth Rate of Operating Income from Securities Reports by Sentence Based Analysis
Sachio HIROKAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A business analysis is needed in many scenes, such as selection of a promising company for the investment and for job-hunting activities. Conventionally, a business analysis was mainly conducted on financial numerical data. In recent years, thanks to the development of the natural language processing and related tools, financial text data are gaining hot attention as the target of text mining. The present paper proposes a method to predict the growth rate of operating income of the next accounting period using the securities report. The proposed method applies feature selection to construct models to predict if a sentence appears in securities reports with high growth rate. The models are then used for prediction with respect to reports. Empirical evaluation is conducted for the securities reports of pharmaceutical companies and confirmed that the proposed method outperforms a baseline method that uses document based SVM (support vector machine) with the optimal parameter.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Securities Report / Growth Rate of Operating Income / SVM / Feature Selection
Paper # NLC2013-29
Date of Issue

Conference Information
Committee NLC
Conference Date 2013/9/5(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Prediction of Growth Rate of Operating Income from Securities Reports by Sentence Based Analysis
Sub Title (in English)
Keyword(1) Securities Report
Keyword(2) Growth Rate of Operating Income
Keyword(3) SVM
Keyword(4) Feature Selection
1st Author's Name Sachio HIROKAWA
1st Author's Affiliation Research Institute for Information Technology, Kyushu University()
Date 2013-09-13
Paper # NLC2013-29
Volume (vol) vol.113
Number (no) 213
Page pp.pp.-
#Pages 6
Date of Issue