Abstract:
Big data is the term used to refer to any group of datasets so huge and composite that it is difficult to process the same using traditional data processing applications. Big data analytics is a set of procedures and technologies that entail new forms of integration to uncover large unknown values from large datasets that are various, complex, and of an immense scale. Analysing big data is a challenging task as it contains huge dispersed file systems which should be fault-tolerant, flexible and scalable. There is an immense need of constructions, platforms, tools, techniques and algorithms to handle big data. Some of the tools used to anlyse big data are Hadoop, Map Reduce, Apache Hive, and No SQL, among others. Techniques for big data analytics include descriptive, diagnostic and prescriptive analytics. This chapter compares the techniques and tools used for big data analytics by the Technical University of Kenya with those used by Strathmore University. The study on which this chapter is based was conducted as a mixed method research to enable deep understanding of the concept. Primary data was collected through structured questionnaires and interviews with clientele and information communication technology staff from the two institutions in Nairobi, Kenya. Secondary data was collected through document analysis. Data was analysed and presented using descriptive statistics. The findings revealed that the tools used frequently for big data analytics were SQL and Java. The two academic institutions mostly used descriptive big data analytics techniques. There was variance in the use of some techniques where SU applied predictive and TUK diagnostic techniques, SU used rules and algorithms to detect the patterns. They also employed statistical analysis, data mining and machine learning to get meaning from data. On the other hand, TUK employed diagnostic analytics to examine their big data.
