Article Preview
TopIntroduction
Electronic government (e-government) is the use of information and communication technology as a process of interaction between government and citizens to increase the service to citizen, for example, e-government application in the legislative and judicative area can improve internal efficiency of democratic governance (Jacob et al., 2017a). However, technological, governing and social issues have to tread carefully in order to adopt these phenomena. Carter and Weerakkody (2008) stated that one important factor for the success of e-government services is the acceptance and willingness of people to use e-government. Meanwhile, the other scholars Al-hujran et al. (2015) and Lian (2015) stated that e-government leads to better transparency, accountability and public services.
E-government initiatives are still in the early stages in most developing countries (Alomari et al., 2014; Chartier et al., 2015; Chen et al., 2015), and face many issues regarding the adoption and implementation (Rana et al., 2013). Adoption and implementation are fundamental stages in terms of measuring the success in using of e-government systems. While governments develop e-government systems to provide e-services to citizens, the adoption and usage level is still low especially in developing countries (Kim & Grant, 2010; Stier, 2015; Sharma, 2015). Successful implementation of electronic government processes and satisfactory usage level by all government stakeholders are the main goals, thus, analyzing the significant factors that influence the adoption and utilization of e-government becoming a necessity. The traditional main objectives of analyzing significant factors of e-government service are to deal with the uncertainty due to designing the e-government adoption model in order to improve public service. To achieve this objective, certain clustering techniques are also being applied. Clustering a set of objects into homogeneous classes is an important data mining operation. Furthermore, clustering is a process for grouping data into multiple clusters or groups so that data in one cluster has a maximum degree of similarity and the data between clusters has a minimum similarity (Qin et al., 2012; Herawan et al., 2010; Yanto et al., 2012). However, certain set theory is not well suited for analyzing uncertainty information systems, as demonstrated in the previous work on constructing student models through mining student’s classification (Wang et al., 2001). Meanwhile, the other work studied about mining significant association rules and rough set theory for clustering the e-government data set in Indonesia (Jacob et al., 2017b; Jacob et al., 2017c). Their results showed that attention should be given to handle the uncertainty information in order to reach a satisfactory prediction accuracy.
This work applied maximum attribute relative as the clustering technique for grouping e-government data set. It is based on a concept of attribute relative where the comparison of attributes is made by taking into account the relative of the attribute at the category level (Mamat et al., 2013). The data were taken from a survey aimed to identify the citizen behavior in using e-government. Furthermore, descriptive statistics is used to find out the Mean (M) and Standard Deviation (SD). Therefore, the nine variables, namely: (1) Performance Expectancy (PE), (2) Effort Expectancy(EE), (3) Social Influence (SI), (4) Facilitating Condition (FC), (5) Behavior Intention (B), (6) User Behavior, (7) Trust (TR), (8) System Quality (SQ), and (9) Information Quality (IQ) are examined to identify the variables to select the best clustering attribute.