Paper Info
Title
The Impact Of Csr'S Multi-Dimensions On Performance Assessment: A Joint Utilization Of Latent Topic Modelling And Support Vector Machine
Abstract
This study examines the impact of corporate social responsibility (CSR) news reports on corporate operating performance forecasting using a large database of publicly-listed electronics firms in Taiwan. By executing latent topic modelling, we can break down CSR news reports into multiple dimensions and then examine which dimension(s) affects operating performance. To offer decision-makers with a comprehensive, overarching view of the corporate's operations, the data envelopment analysis (DEA), that can handle multiple input and multiple output variables simultaneously, was conducted to generate the performance rank. The analyzed results were then fed into support vector machine (SVM) to construct the model for performance forecasting. The inherent parameters were decided by harmony search (HS) algorithm. The proposed approach, supported by real samples, can assist both internal and external stakeholders in allocating scarce resources to specific CSR dimensions to enhance a corporate's growth potential as well as to achieve a win-win situation.
Year
DOI
Venue
2018
10.1109/SMC.2018.00624
2018 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC)
Keywords
Field
DocType
corporate social responsibility, latent topic modelling, decision making, data envelopment analysis, support vector machine
Corporate social responsibility,Scarcity,Industrial engineering,Computer science,Support vector machine,Electronics,Artificial intelligence,Data envelopment analysis,Harmony search,Topic model,Multiple time dimensions,Machine learning
Conference
ISSN
Citations 
PageRank 
1062-922X
0
0.34
References 
Authors
0
4
Name
Order
Citations
PageRank
Fu Hsiang Chen1133.27
Te-Min Chang2346.29
Sin-Jin Lin3707.51
Ming-Fu Hsu412411.40