Data Mining. New Trends, Applications and Challenges

Data mining involves extracting interesting patterns from data to create and enhance decision support systems. Whereas in the early days of data mining, some tasks already relied on statistical and operations research methods such as linear programming and forecasting, data mining methods nowadays are based on a variety of methods including linear and quadratic optimisation, as well as on concepts such as genetic algorithms and artificial ant colonies. Their use has quickly become widespread, with applications in domains ranging from credit risk, marketing, or fraud detection to counter-terrorism. In all of these, data mining is increasingly forming a key part in the actual decision making. Nonetheless, many challenges still need to be tackled, ranging from data quality issues to e.g. the problem of how to include domain experts’ knowledge, or how to monitor the performance of the obtained models. In this paper, we outline a series of upcoming trends and challenges within data mining.