The application of data mining techniques and statistical analysis to the sports field has received increasing attention in the last decade. One of the most famous sports in the world is soccer, and the present work deals with it, using data from the 2009/2010 season to the 2015/2016 season from nine European leagues extracted from the Kaggle European Soccer database. Overall performance indicators of the four roles in a soccer team (forward, midfielder, defender and goalkeeper) for home and away teams are used to investigate the relationships between them and the results of matches, and to predict the wins of the home team. The model used to answer both these demands is the Bayesian Network. This study shows that this model can be very useful for mining the relations between players' performance indicators and for improving knowledge of the game strategies applied by coaches in different leagues. Moreover, it is shown that the ability to predict match results of the proposed Bayesian Network is roughly the same as that of the Naive Bayes model.
With the advent of the data era and of new, more intelligent interfaces for supporting decision-making, there is a growing need to define, model and assess human ability and data visualization usability for better encoding and decoding of data patterns. Data Visualization Literacy (DVL) is the ability to encode and decode data into and from a visual language. Although this ability and its measurement are crucial for advancing human knowledge and decision capacity, they have seldom been investigated, let alone systematically. To address this gap, this paper presents a systematic literature review comprising 45 reports on DVL, analyzed using the PRISMA methodology. Our results include the identification of the purposes of DVL, its satellite aspects, the models proposed, and the assessments designed to evaluate the degree of DVL of people. Eventually, we devise many research directions including, among the most challenging, the definition of a (standard) unifying construct and protocol for DVL.
In this paper we assess how intra-day electricity prices can improve the prediction of zonal day-ahead wholesale electricity prices in Italy. We consider linear autoregressive models with exogenous variables (ARX) with and without interactions among predictors, and non-parametric models taken from the machine learning literature. In particular, we implement Random Forests and support vector machines, which should automatically capture the relevant interactions among predictors. Given the large number of predictors, ARX models are also estimated using LASSO regularization, which improves predictions when regressors are many and selects the important variables. In addition to zonal intra-day prices, among the predictors we include also the official demand forecasts and wind generation expectations. Our results show that the prediction performance of the simple ARX model is mostly superior to those of machine learning models. The analysis of the relevance of exogenous variables, using variable importance measures, reveals that intra-day market information successfully contributes to the forecasting performance, although the impact differs among the estimated models.
Background: Investigation of long time series of cancer data can still be very useful in helping to identify Cancer Control priorities and achievements. Since the partition of Ireland into the independent Republic of Ireland and Northern Ireland, which remained part of the United Kingdom, cancer mortality data have been published in an essentially similar format in both countries. The information presented here will contribute to providing a basis for the collaborative Cancer Research programme initiated recently. Patients and methods: Cancer mortality data have been assembled and analysed separately for the Republic of Ireland and Northern Ireland: the data have then been combined to present mortality rates for the whole of Ireland, covering the period from 1926 to 1995. Several rubrics had to be aggregated to provide data continuously over the time span (e.g. colon and rectum and cervix and body of the uterus). When data were only available in 10-year classes of age, the EM algorithm was employed to obtain 5-year age-specific rates. All rates presented are age-standardised, employing the World Standard Population. Results: In women, the death rate from all neoplasms combined increased very slightly from 117 per 100 000 in 1946–1950 to 120 per 100 000 in 1991–1995. In men, the death rate increased from 127 per 100 000 to 172 per 100 000 over the same time period. The overall cancer death rate in Ireland is currently similar to the European average in men, although in women it is among the top fifth of national cancer mortality rates in European countries. While cancer is a major cause of death in Ireland, there is no evidence of an evolving epidemic building up: the death rates from most forms of cancer are declining towards the end of the time period considered. Conclusions: As demonstrated by falling death rates from Hodgkin’s disease and testicular cancer, major treatment advances appear to have been incorporated effectively into clinical practice in Ireland. Progress is apparent in tobacco control and further initiatives in this area must be undertaken since tobacco appears to be the only major new carcinogen introduced recently into the Irish environment during the period covered by this study. Effective population-based screening programmes for cervix and breast cancer and, more controversially, consideration of a National Prostate Cancer Screening programme, offer scope for further improvement in mortality. Examination of this long time series of mortality data from Ireland provides information about the evolving cancer pattern and provides the necessary background to evaluate the impact of the cross-border cancer research activities now being launched.