Extant research has made limited efforts to conduct competitive intelligence analysis based on recruitment texts. To fill the gap, this study proposes a method for deriving and analyzing competitive relationships, identifying competition paths, and calculating asymmetric competitiveness degrees, from the recruitment texts on e-recruiting websites. Specifically, this study developed a competitive evaluation index system for companies’ skill needs and resource base based on 53,171 job descriptions and 42,641 company profiles published by companies across 8 industries (including 35 industry segments) using automated text processing methods. Furthermore, in order to identify competitive paths and calculate the degree of asymmetric competitiveness, this study proposes a modified bipartite graph approach (i.e., MBGA) for competitive intelligence analysis of recruitment texts based on the competition evaluation index system. Experiments on a real-world dataset of the representative companies clearly validated the effectiveness of the method. Compared to the five state-of-the-art methods, MBGA performs better in disclosing the overall competition and is more accurate in terms of the error rating ratio (i.e., ERR) of the competition.
Liu, Lu; Fang, Jiaming; and Li, Liangqiang, "Exploring enterprises competition: From a perspective of massive recruitment texts mining" (2022). ICEB 2022 Proceedings (Bangkok, Thailand). 13.