CSR Image Construction of Chinese Construction Enterprises in Africa Based on Data Mining and Corpus Analysis

Abstract

Since there is negative coverage of some western media on the business activities of Chinese overseas enterprises, which has adverse impact on the image of Chinese enterprises and even the national image of China, this study aims to detect the corporate social responsibility image (hereafter CSR image) of Chinese construction enterprises in Africa (hereafter CCEA) through analyzing the coverage of Financial Times (hereafter FT) from the UK and The Wall Street Journal (hereafter WSJ) from the US and dig up the motives behind their coverage. Octopus is first applied to mine and collect the reports data on CCEA from 2011 to 2019 by the two media. Two small corpora including the reports are then built. NVivo is next used to do the statistical analysis and clustering analysis of the keywords in two corpora as a whole and AntConc is finally utilized to do the statistics of high-frequency evaluative adjectives and nouns modified by evaluative adjectives as well as the concordance of the low-frequency words but closely relevant to corporate social responsibility (hereafter CSR) in two corpora, respectively. The results of the detailed analyses of the keywords are combined to unveil the CSR image of CCEA, which is followed by a discussion about the motives behind the coverage and finally some suggestions are put forward to improve the CSR image of CCEA. Theoretically, the present study promotes the interaction among data science, management, communications, and linguistics; practically it offers some advice to CCEA to elevate their CSR image.