Heterogeneous ensembles of classifiers in predicting Bundesliga football results

Kozák, Jan; Głowania, Szymon

doi:10.1016/j.procs.2021.08.161

Cited by 10 publications

(6 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the literature, works based on the use of the various algorithms can be found. The most popular approaches are as follows: Support Vector Machine [10]; Artificial Neural Network [14]; Random Forest [5]; Decision Tree [15]; Logistic Regression [16]. In Table 1 has been presented a comparison of the approaches used in individual sports disciplines.…”

Section: Machine Learning Algorithms In Sports Datamentioning

confidence: 99%

“…Some sports have found opportunities to apply machine learning, from predicting sports results to planning team lineups [5,6]. In the literature, articles can be found that present issues related to monitoring fitness and injuries in sports such as basketball or speedway [7,8].…”

Section: Introductionmentioning

confidence: 99%

“…Invariably, the most popular and most effective algorithms used in sports prediction are artificial neural networks, logistic regression, support vector machine, random forests and naive Bayes classifiers. In the current research, artificial neural networks have shown great potential [9], along with random forests [10] and heterogeneous ensembles of classifiers [5]. The last of the mentioned solutions have only recently been considered for use in sports-related prediction, but the results provided are very promising.…”

Section: Introductionmentioning

confidence: 99%

“…In order to verify the given approach, to prepare the dataset and to check its usefulness in the application of different algorithms and the limited feature space, the algorithms previously used in this problem and presented in [5] were selected as a base. On the other hand, in terms of specific implementations of the algorithms, solutions from [48] were selected; the exact machine learning algorithms used were the following:…”

mentioning

confidence: 99%

See 3 more Smart Citations

Knowledge Discovery in Databases for a Football Match Result

2023

Self Cite

View full text Add to dashboard Cite

The analysis of sports data and the possibility of using machine learning in the prediction of sports results is an increasingly popular topic of research and application. The main problem, apart from choosing the right algorithm, is to obtain data that allow for effective prediction. The article presents a comprehensive KDD (Knowledge Discovery in Databases) approach that allows for the appropriate preparation of data for sports prediction on sports data. The first part of the article covers the subject of KDD and sports data. The next section presents an approach to developing a dataset on top football leagues. The developed datasets are the main purpose of the article and have been made publicly available to the research community. In the latter part of the article, an experiment with the results based on heterogeneous groups of classifiers and the developed datasets is presented.

show abstract

Section: Machine Learning Algorithms In Sports Datamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Knowledge Discovery in Databases for a Football Match Result

2023

Self Cite

View full text Add to dashboard Cite

show abstract

“…A quotient, integer variable ranging from six to 34 for Germany, and to 38 for Spain and Italy. Based on conducted experiments and the arguments indicated in the literature, the data for the first five rounds of each of the seasons were excluded from the analysis [ 62 ]. Team1 : The name of the first team.…”

Section: Data Preparation and Preprocessingmentioning

confidence: 99%

Real-World Data Difficulty Estimation with the Use of Entropy

Juszczuk

Kozák

Dziczkowski

et al. 2021

Entropy

Self Cite

View full text Add to dashboard Cite

In the era of the Internet of Things and big data, we are faced with the management of a flood of information. The complexity and amount of data presented to the decision-maker are enormous, and existing methods often fail to derive nonredundant information quickly. Thus, the selection of the most satisfactory set of solutions is often a struggle. This article investigates the possibilities of using the entropy measure as an indicator of data difficulty. To do so, we focus on real-world data covering various fields related to markets (the real estate market and financial markets), sports data, fake news data, and more. The problem is twofold: First, since we deal with unprocessed, inconsistent data, it is necessary to perform additional preprocessing. Therefore, the second step of our research is using the entropy-based measure to capture the nonredundant, noncorrelated core information from the data. Research is conducted using well-known algorithms from the classification domain to investigate the quality of solutions derived based on initial preprocessing and the information indicated by the entropy measure. Eventually, the best 25% (in the sense of entropy measure) attributes are selected to perform the whole classification procedure once again, and the results are compared.

show abstract