ai

Description

systematic literature review

Steps to reproduce

The authors followed a systematic approach to achieve their goal of obtaining the most relevant articles for their study (as indicated in the Figure 1). We conducted a comprehensive search across seven scientific databases, including articles, scientific papers, books, and book chapters that focused on HRM and artificial intelligence. The outcome of the investigation yielded a total of 412,632 scholarly publications about humanity and artificial intelligence. Next, we proceeded with a more focused approach by restricting our search to scholarly articles that specifically mention both HRM and AI in their title, abstract, and keywords. Additionally, we conducted a thorough examination of the data and eliminated any instances of duplicate and irrelevant articles, resulting in a reduction of the search outcome to 703 papers. Subsequently intensifying our attention and refining our criteria, we progressed to the third phase of the study, when we only identified scientific publications published solely within certain disciplines while excluding papers published as chapters, conference papers, or reports. We exclusively chose publications that are readily available and can be downloaded in PDF format. These articles were limited to the publishing period between 2000 and 2023. There were 54 scientific publications obtained from the last step that directly aligned with the existing study. Regarding the research results, it is important to mention that no studies involving variables were conducted prior to the beginning of 2018, except for a study titled " AI in HRM: An experimental study of an expert system " by (Lawler & Elliot, 1996). However, this study was excluded from the selected list due to its outdated nature. The specified search encompasses a certain range of dates. 3/ Result The data collection process relied on the PRISMA chart, which encompasses reporting elements for systematic reviews and meta-analyses. This involved extracting, analysing, examining, and refining the data to identify papers directly relevant to the study topic with increased accuracy. The outcome was a conclusive compilation of 54 articles .We continued to implement the further stages of SLR methodology by inputting the compilation of 54 chosen articles into VOSviewer. This allowed us to construct a bibliometric network and visually depict it by utilising bibliographic associations. VOSviewer is a no-cost software application used for generating, displaying, and investigating network data maps(van Eck & Waltman, 2017), as well as term maps derived from location-based text mining. It is employed to visually represent citation relationships, co-citations, and network groups among chosen articles (Jan van Eck & Waltman, 2012). The bibliometric data acquired with VOSviewer, including keywords, author names, affiliations, and citation , can offer a substantial amount of information when analysed collectively in a sample of papers within a certain field or topic .

Institutions

Categories

Funding

Licence