Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Data Mining Practical Machine Learning Tools and Techniques

We explore the effect of dimensionality on the nearest neighbor problem. We show that under a broad set of conditions (much broader than independent and identically distributed dimensions), as dimensionality increases, the distance to the nearest data point approaches the distance to the farthest data point. To provide a practical perspective, we present empirical results on both real and synthetic data sets that demonstrate that this effect can occur for as few as 10-15 dimensions. These results should not be interpreted to mean that high-dimensional indexing is never meaningful; we illustrate this point by identifying some high-dimensional workloads for which this effect does not occur. However, our results do emphasize that the methodology used almost universally in the database literature to evaluate high-dimensional indexing techniques is flawed, and should be modified. In particular, most such techniques proposed in the literature are not evaluated versus simple linear scan, and are evaluated over workloads for which nearest neighbor is not meaningful. Often, even the reported experiments, when analyzed carefully, show that linear scan would outperform the techniques being proposed on the workloads studied in high (10-15) dimensionality!.

When is nearest neighbor meaningful

Relief-based feature selection: Introduction and review.

Gamification of education is a developing approach for increasing learners’ motivation and engagement by incorporating game design elements in educational environments. With the growing popularity of gamification and yet mixed success of its application in educational contexts, the current review is aiming to shed a more realistic light on the research in this field by focusing on empirical evidence rather than on potentialities, beliefs or preferences. Accordingly, it critically examines the advancement in gamifying education. The discussion is structured around the used gamification mechanisms, the gamified subjects, the type of gamified learning activities, and the study goals, with an emphasis on the reliability and validity of the reported outcomes. To improve our understanding and offer a more realistic picture of the progress of gamification in education, consistent with the presented evidence, we examine both the outcomes reported in the papers and how they have been obtained. While the gamification in education is still a growing phenomenon, the review reveals that (i) insufficient evidence exists to support the long-term benefits of gamification in educational contexts; (ii) the practice of gamifying learning has outpaced researchers’ understanding of its mechanisms and methods; (iii) the knowledge of how to gamify an activity in accordance with the specifics of the educational context is still limited. The review highlights the need for systematically designed studies and rigorously tested approaches confirming the educational benefits of gamification, if gamified learning is to become a recognized instructional approach.

/pdf/gamifying-education-what-is-known-what-is-believed-and-what-ghobpeqha0.pdf

Gamifying education: what is known, what is believed and what remains uncertain: a critical review

Introduction Traditional schooling is perceived as ineffective and boring by many students. Although teachers continuously seek novel instructional approaches, it is largely agreed that today's schools face major problems around student motivation and engagement (Lee & Hammer, 2011). The use of educational games as learning tools is a promising approach due to the games' abilities to teach and the fact that they reinforce not only knowledge but also important skills such as problem-solving, collaboration, and communication. Games have remarkable motivational power; they utilize a number of mechanisms to encourage people to engage with them, often without any reward, just for the joy of playing and the possibility to win. Creating a highly engaging, full-blown instructional game however is difficult, time consuming, and costly (Kapp, 2012a), while typically targeting only a single set of learning objectives as chosen by the game designer. In addition, their effective classroom adoption requires certain technical infrastructure and appropriate pedagogical integration. As opposed to using elaborate games requiring a large amount of design and development efforts, the "gamification" approach suggests using game thinking and game design elements to improve learners' engagement and motivation. Gamification, defined by Deterding et al. (2011) as the use of game design elements in non-game contexts, is a fairly new and rapidly growing field. The concept of gamification is different from that of an educational or serious game. While the latter describes the design of full-fledged games for non-entertainment purposes, "gamified" applications merely employ elements of games. The term "gamification" is quite recent: According to (Deterding et al., 2011) its first documented use is in 2008 but it did not see widespread adoption before the second half of 2010. Nevertheless, the concept itself is not new. For example, badges and ranks have been long used in the military, in the early Soviet era, game elements were used by the Soviet Union leaders as a substitute for monetary incentives for performing at work, etc. In recent years gamification has seen rapid adoption in business, marketing, corporate management, and wellness and ecology initiatives. This is driven by its potential to shape users' behavior in a desirable direction. Loyalty programs such as the frequent-flyer programs, Foursquare, and Nike+ are often given as examples of successful gamified mass-market products. Stackoverflow.com provides another example in which users' reputations increase as they answer questions and receive votes for their answers. Online education sites such as codeacademy.com and khanacademy.org use game elements to better engage users. The more courses and lessons that users complete, the more badges they earn. Sites like eBay and Fitocracy use game elements to keep people engaged and to encourage friendly competition between users. Gamification is still rising in popularity. According to Gartner's Hype Cycle (Gartner, 2013), a research methodology that outlines an emerging technology's viability for commercial success, gamification is at the peak of the Hype Cycle in 2013, with an expectation for reaching the productivity plateau in five to ten years. This position, however, mainly reflects its use in business contexts. The penetration of the gamification trend in educational settings seems to be still climbing up to the top, as indicated by the amount and annual distribution of the reviewed works. This paper presents the results of a study of the published works on the application of gamification to education, which aims to shed light on the tendencies and emerging practices in this area. There are few literature reviews on gamification (see Xu, 2012; Hamari, Koivisto, & Sarsa, 2014; Nah, Zeng, Telaprolu, Ayyappa, & Eschenbrenner, 2014), with only the last one focusing on education. This study differs from the latter by presenting a thematic analysis instead of narrative summaries that focus on a qualitative review. …

Gamification in Education: A Systematic Mapping Study

Learning is a goal driven social activity determined by motivational factors. To be able to efficiently gamify learning for improved student motivation and engagement, the educators have to understand the related aspects studied in games, motivational psychology and pedagogy. This will help them to identify the factors that drive and explain desired learning behaviors. This paper presents a survey of the main approaches employed in gamification and the emerging new directions in the context of the relevant motivational psychology and pedagogy. The focus is on the motivational factors that impact learning and understanding of behavior change. The purpose of the paper is two-fold: on one side, to provide analysis and guide to relevant works related to gamification along with outlining the emerging trends, and on the other, to provide foundation for evaluation and identification of the areas of possible improvements.

/pdf/from-gamification-to-gameful-design-and-gameful-experience-3waj1y5rar.pdf

From Gamification to Gameful Design and Gameful Experience in Learning

The aim of the present study is to apply support vector machines (SVM) and artificial neural network (ANN) as sex classifiers and to generate useful classification models for sex estimation based on cranial measurements. Besides, the performance of the generated sub-symbolic machine learning models is compared with models developed through logistic regression (LR). The study was carried out on computed tomography images of 393 Bulgarian adults (169 males and 224 females). The three-dimensional coordinates of 47 landmarks were acquired and used for calculation of the cranial measurements. A total of 64 measurements (linear distances, angles, triangle areas and heights) and 22 indices were calculated. Two datasets were assembled including the linear measurements only and all measurements and index, respectively. An additional third dataset comprising all possible interlandmark distances between the landmarks was constructed. Two machine learning algorithms—SVM and ANN and a traditional statistical analysis LR—were applied to generate models for sex estimation. In addition, two advanced attribute selection techniques (Weka BestFirst and Weka GeneticSearch) were used. The classification accuracy of the models was evaluated by means of 10 × 10-fold cross-validation procedure. All three methods achieved accuracy results higher than 95%. The best accuracy (96.1 ± 0.5%) was obtained by SVM and it was statistically significantly higher than the best results achieved by ANN and LR. SVM and ANN reached higher accuracy by training on the full datasets than the selection datasets, except for the sample described by the interlandmark distances, where the reduction of attributes by the GeneticSearch algorithm improved the accuracy.

Machine learning approaches for sex estimation using cranial measurements

The paper deals with the problem of improving problem-solving behavior of traditional KBS in the course of its real operation which is a part of the maintenance task. The solution of the problem is searched in integration of the KBS with a specially designed case-based reasoning module used for correcting solutions produced by the KBS. Special attention is paid to the methods of case matching and reconciling conflicts between CBR and RBR. The proposed solution for both problems is based on treating the maintenance task as a problem for learning two-tiered domain representation. From this view point rules form the first domain tier reflecting existing strong patterns in the representation of domain concepts, while the second tier is formed by the newly solved cases along with a special domain-dependent procedure for case matching. The main ideas of the approach are illustrated by the results of some experiments with the experimental system CoRCase.

KBS Maintenance as Learning Two-Tiered Domain Representation

The paper presents a casual-probabilistic approach to the technical diagnosis in which the solution of the technical diagnostic problem is considered as a probabilistic inference on a special kind of Bayesian networks called Diagnostic Bayesian Networks. A mechanism of probabilistic inference and an algorithm for inference control are described. It is provided that a diagnostic problem represented by a singly connected Diagnostic Bayesian network can be decomposed to a sequence of subproblems with directed tree of multitree topology which are exactly solved in the sense of minimizing the average number of executed tests. The applications of the approach and future trends are briefly discussed.

Gennady Agre

Papers

Gamification in Education: A Systematic Mapping Study

From Gamification to Gameful Design and Gameful Experience in Learning

Machine learning approaches for sex estimation using cranial measurements

KBS Maintenance as Learning Two-Tiered Domain Representation

Diagnostic Bayesian Networks.