Impact of Machine Learning on Business Predictive Analytics in Telecommunication Firms in South-South, Nigeria

This study investigated the impact of machine learning on business predictive analytics in telecommunication firms in South-South, Nigeria. The study adopted the simple correlation research design. The population of the study consisted of the managers and supervisors from four telecommunication firms (MTN, Globacom, Airtel and 9Mobile) in South-South, Nigeria, which is 162. A sample size of 115 was determined using the Taro Yamane formula. The instrument used for the study was a structured questionnaire using 4-pointLikert scales. The Cronbach Alpha statistic was used to obtain index coefficient values of 0.874, 0.864, 0.865 for the dependent variables and 0.885 for the independent variable as the instrument reliability ratios. The parametric assumptions were diagnosed: outliers were checked using Boxplot and the results indicated potential outliers; and Kolmogorov-Smirnov (KS) and Shapiro-Wilk (SW) statistics of examining normality revealed that the assumption of normality was not met; hence, the need for Spearman’s Rank Correlation Coefficient as the method of data analysis. The research questions and research hypotheses were answered and tested with Spearman correlation statistic so as to establish and measure the “significance” of the relationship between the dependent and independent variables in the study. The analysis was enabled by the use of IBM SPSS version 25.0 software package. . The results of the study revealed a strong and positive correlation between the adoption of machine learning algorithms and the accuracy of business predictive analytics (r = 0.883), speed of decision-making (r = 0.842), and business outcomes (r = 0.722). The study recommended among others that telecommunication firms in South-South, Nigeria should adopt machine learning algorithms to improve the accuracy, speed, and business outcomes of their predictive analytics