Big data usage, natural language processing, and ethics in developing countries: instrument development using systematic literature review

Abstract

Artificial intelligence (AI) and big data analytics technology are widely used globally. However, the challenges both present to organizations and individuals differ in developing and developed countries. This study aims to develop measurement instruments using a systematic literature review to assess the usage of big data analytics in developing countries, considering the ethical and AI components of natural language processing (NLP) challenges. The PRISMA method, a systematic review protocol, was considered for developing measurement instruments using structural (that relates the latent constructs) and measurement (that relates the latent constructs to their indicators) frameworks. About 136 articles were extracted from the publication databases. After exclusion and inclusion criteria were used, 99 papers were extensively processed and analyzed. The findings revealed 29 items measuring big data analytics usage in developing countries when NLP is moderating. Specifically, we found five items measuring the latent construct of smartphone record data, six items measuring the latent construct of big data location analytics, five items measuring the latent construct of big data analytics family ties, seven items measuring the latent construct of big data analytics ethics, and six items measuring the latent construct NLP in developing countries. Theoretical and practical contributions were offered .