No abstract

Companies have realized they need to hire data scientists, academic institutions are scrambling to put together data-science programs, and publications are touting data science as a hot-even "sexy"-career choice. However, there is confusion about what exactly data science is, and this confusion could lead to disillusionment as the concept diffuses into meaningless buzz. In this article, we argue that there are good reasons why it has been hard to pin down exactly what is data science. One reason is that data science is intricately intertwined with other important concepts also of growing importance, such as big data and data-driven decision making. Another reason is the natural tendency to associate what a practitioner does with the definition of the practitioner's field; this can result in overlooking the fundamentals of the field. We believe that trying to define the boundaries of data science precisely is not of the utmost importance. We can debate the boundaries of the field in an academic setting, but in order for data science to serve business effectively, it is important (i) to understand its relationships to other important related concepts, and (ii) to begin to identify the fundamental principles underlying data science. Once we embrace (ii), we can much better understand and explain exactly what data science has to offer. Furthermore, only once we embrace (ii) should we be comfortable calling it data science. In this article, we present a perspective that addresses all these concepts. We close by offering, as examples, a partial list of fundamental principles underlying data science.

This paper surveys the intersection of two fascinating and increasingly popular domains: swarm intelligence and data mining. Whereas data mining has been a popular academic topic for decades, swarm intelligence is a relatively new subfield of artificial intelligence which studies the emergent collective intelligence of groups of simple agents. It is based on social behavior that can be observed in nature, such as ant colonies, flocks of birds, fish schools and bee hives, where a number of individuals with limited capabilities are able to come to intelligent solutions for complex problems. In recent years the swarm intelligence paradigm has received widespread attention in research, mainly as Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO). These are also the most popular swarm intelligence metaheuristics for data mining. In addition to an overview of these nature inspired computing methodologies, we discuss popular data mining techniques based on these principles and schematically list the main differences in our literature tables. Further, we provide a unifying framework that categorizes the swarm intelligence based data mining algorithms into two approaches: effective search and data organizing. Finally, we list interesting issues for future research, hereby identifying methodological gaps in current research as well as mapping opportunities provided by swarm intelligence to current challenges within data mining research.Editor: Foster Provost.

Classifier calibration is the process of converting classifier scores into reliable probability estimates. Recently, a calibration technique based on isotonic regression has gained attention within machine learning as a flexible and effective way to calibrate classifiers. We show that, surprisingly, isotonic regression based calibration using the Pool Adjacent Violators algorithm is equivalent to the ROC convex hull method.

Rules are commonly used for classification because they are modular; intelligible and easy to learn. Existing work in classijication rule learning assumes the goal is to produce categorical classifications to maximize classijication accuracy Recentiwork in machine learning has pointed out the limitations of classijication accuracy: when class distributions are skewed, or error costs are unequal, an accuracy maximizing rule set can perform poorly. A more flexible use of a rule set is to produce instance scores indicating the likelihood that an instance belongs to a given class. With such an abilit): we can apply rulesets effectively when distributions are skewed or error costs are unequal. This paper empirically investigates different strategies for evaluating rule sets when the goal is to maximize the scoring (ROC) performance.