Personalisation framework for Malaysian m-government service

Abstract

Mobile Government (m-Government) is an implementation strategy involving all types of mobile technology services and applications to enhance the benefits of citizens, businesses and all government units. Due to the escalating number of m-Government services developed annually, citizens face difficulties finding the appropriate government services according to their needs which indirectly lead to information overload. There is a gap in the existing m-Government personalisation framework which is deficient in personalisation efficiency and intelligence and is weak at data management. The current m-Government services are limited to simple online presentations, and intelligent services are highly desirable. The aim of this study was to propose a Personalisation Framework for Malaysian m-Government services (Pe-mGov) to better understand the needs of citizens toward the m-Government services. Three research objectives guided this study; firstly, to identify and categorise the m-Government services for the Malaysian citizens; secondly, to design a personalisation framework for Malaysian m-Government services (Pe-mGov); and thirdly, to evaluate the applicability of personalisation framework through the development of predictive model. A Design Science Research methodology was used to solve the problems to ensure this framework would be rigour and relevance. In this study, five steps were involved in developing the framework namely, firstly, categorisation of m-Government services; secondly, data collection; thirdly, storage of data regarding citizen profiles and feedback; fourthly, cluster analysis and predictive model; and fifthly, model evaluation and validation. The demographic and services variables were the dependent variables used for utilizing the two-step cluster technique. The multinomial logistic regression was used to estimate the independent associations, obtaining the odds ratios and 95% confidence intervals. Three clusters were generated, namely, firstly, working people; secondly, non-working people and thirdly, students. The findings showed that the accuracy of the predictive model was 92.0% and the model is an excellent fit to recommend the m-Government service. Besides, this proposed framework can be used as a guide to assist government agencies in promoting m-Government services among citizens.