DATA MINING IN VULNERABLE SYSTEMS WITH DATABASE SECURITY IMPLEMENTATIONS

ABSTRACT

Data mining is the process of obtaining useful information or knowledge from a rather ambiguous dataset. Administrations, businesses, organisations and research teams rely on this process to make informed decisions, forecasters also study trends in a dataset to predict future events. Data mining is a complex endeavour and sometimes overwhelmingly tedious, it involves a lot of technicalities and relies on integrity of the source data to effectively carry out the exercise with accuracy and prospect. In data mining, misleading assumptions or inordinate inferences from analytical models have proven to be more disastrous than the problem it intends to solve. Irrespective of the sanity of data and the credibility of the methodology applied, the information or knowledge drawn from such hypothetical studies matters most. Thus, this study espouses the need to inculcate proficient security and data loss prevention architecture into data mining processes, such as threat modelling using STRIDE model, access control, firewall, VPN for data transmission, forensic data recovery, etc; as veritable tools in data mining. This study also elaborates the various data loss prevention methods as well as their application areas; the study also examines and advocates the usefulness of Data Loss Prevention tools and techniques in classical data mining; this study further dwells on the need to apply various data analytical methods to petrify, coerce and wrangle raw data to obtain the minutest vital information from a rather ambiguous dataset.For the purpose of design implementation, this study collected data from a renowned online global data repository, the gapminders and painstakingly analyzed the trend in life expectancy for global data, as well as categorization by continents to study Africa trend and distinctively studied the trend in Nigeria. The study also attempted to evaluate and investigate other economic and demographic factors that influence life expectancy and inferences was drawn with assumptions. The hypothesis drawn from this study is strictly the perception of the author.