Kanga(ROO): Handling the micro-DST of the BaBar experiment with ROOT

A system based on ROOT for handling the micro-DST of the BaBar experiment is described. The purpose of the Kanga system is to have micro-DST data available in a format well suited for data distribution within a world-wide collaboration with many small sites. The design requirements, implementation and experience in practice after three years of data taking by the BaBar experiment are presented.     

The Open International Soccer Database for machine learning

How well can machine learning predict the outcome of a soccer game, given the most commonly and freely available match data? To help answer this question and to facilitate machine learning research in soccer, we have developed the Open International Soccer Database. Version v1.0 of the Database contains essential information from 216,743 league soccer matches from 52 leagues in 35 countries. The earliest entries in the Database are from the year 2000, which is when football leagues generally adopted the “three points for a win” rule. To demonstrate the use of the Database for machine learning research, we organized the 2017 Soccer Prediction Challenge. One of the goals of the Challenge was to estimate where the limits of predictability lie, given the type of match data contained in the Database. Another goal of the Challenge was to pose a real-world machine learning problem with a fixed time line and a genuine prediction task: to develop a predictive model from the Database and then to predict the outcome of the 206 future soccer matches taking place from 31 March 2017 to the end of the regular season. The Open International Soccer Database is released as an open science project, providing a valuable resource for soccer analysts and a unique benchmark for advanced machine learning methods. Here, we describe the Database and the 2017 Soccer Prediction Challenge and its results.

     

Discovering relevance knowledge in data: a growing cell structuresapproach

Both information retrieval and case-based reasoning systems rely on effective and efficient selection of relevant data. Typically, relevance in such systems is approximated by similarity or indexing models. However, the definition of what makes data items similar or how they should be indexed is often nontrivial and time-consuming. Based on growing cell structure artificial neural networks, this paper presents a method that automatically constructs a case retrieval model from existing data. Within the case-based reasoning (CBR) framework, the method is evaluated for two medical prognosis tasks, namely, colorectal cancer survival and coronary heart disease risk prognosis. The results of the experiments suggest that the proposed method is effective and robust. To gain a deeper insight and understanding of the underlying mechanisms of the proposed model, a detailed empirical analysis of the models structural and behavioral properties is also provided.     

Modelling codependence in biological systems

A central aim of systems biology is to elucidate the complex dynamic structure of biological systems within which functioning and control occur. The success of this endeavour requires a dialogue between the two quite distinct disciplines of life science and systems theory, and so drives the need for graphical notations which facilitate this dialogue. Several methods have been developed for modelling and simulating biochemical networks, some of which provide notations for graphicall4y constructing a model. Such notations must support the full panoply of mechanisms of systems biology, including metabolic, regulatory, signalling and transport processes. Notations in systems biology tend to fall into two groups. The first group derives its orientation from conventional biochemical pathway diagrams, and so tends to ignore the role of information processing. The second group focuses on the processing of information, incorporating information-processing ideas from other systems-oriented disciplines, such as engineering and business. This, however, can lead to the two crucial and related difficulties of impedance mismatch and conceptual baggage. Impedance mismatch concerns the rift between non-biological notations and biological reality, which forces the researcher to employ awkward workarounds when modelling uniquely biological mechanisms. Conceptual baggage can arise when, for instance, an engineering notation is adapted to cater for these distinctively biological needs, since these adaptations will, typically, never completely free the notation of the conceptual structure of its original engineering motivation. A novel formalism, codependence modelling, which seeks to combine the needs of the biologist with the mathematical rigour required to support computer simulation of dynamics is proposed here. The notion of codependence encompasses the transformation of both chemical substance and information, thus integrating both metabolic and gene regulatory processes within a single conceptual schema.     

Improving clinical decision support through case-based data fusion

This paper presents an information fusion technique based on a knowledge discovery model, and the case-based reasoning decision framework. Using signal data and database records from the heart disease risk estimation domain, three data fusion methods are discussed. Two of these methods combine information at the retrieval-outcome level, and one method merges data at the discovery-input level. The result of these three models are compared and evaluated against the performance of single-source models. It is shown that the methods that fuse information at the retrieval-outcome level are significantly superior.     

Incorporating domain knowledge in machine learning for soccer outcome prediction

The task of the 2017 Soccer Prediction Challenge was to use machine learning to predict the outcome of future soccer matches based on a data set describing the match outcomes of 216,743 past soccer matches. One of the goals of the Challenge was to gauge where the limits of predictability lie with this type of commonly available data. Another goal was to pose a real-world machine learning challenge with a fixed time line, involving the prediction of real future events. Here, we present two novel ideas for integrating soccer domain knowledge into the modeling process. Based on these ideas, we developed two new feature engineering methods for match outcome prediction, which we denote as recency feature extraction and rating feature learning. Using these methods, we constructed two learning sets from the Challenge data. The top-ranking model of the 2017 Soccer Prediction Challenge was our k-nearest neighbor model trained on the rating feature learning set. In further experiments, we could slightly improve on this performance with an ensemble of extreme gradient boosted trees (XGBoost). Our study suggests that a key factor in soccer match outcome prediction lies in the successful incorporation of domain knowledge into the machine learning modeling process.

     

Towards Automated Case Knowledge Discovery in the M 2 Case-Based Reasoning System

In this paper we present the M ² Case-Based Reasoning (CBR) system. The M ² system addresses a number of issues that present methodologies for CBR systems have shied away from. We discuss techniques for removing the knowledge acquisition bottleneck when acquiring case knowledge. Here, case knowledge refers to the complementary knowledge structures, cases (more specific in nature) and adaptation rules (more general). We address the use of negative cases for updating the case knowledge as well as for refining the similarity measures. In particular we discuss in detail, showing experimental results, the use of Data Mining within the M ² system to build the case base from a database containing operational data, and discover adaptation rules. A methodology to monitor the competence of the CBR system and to utilize negative cases for updating the CBR system to enhance its competence is also discussed. The M ² CBR system also employs Rough Set and Fuzzy Set theories to further enhance its capabilities within real-world applications as well as providing a richer and truer model of human reasoning.