No abstract

Managers make decisions on team tactics, formations, and player selection based on their own experiences. The managers have limitations in understanding the team's situation and sometimes they can think wrong. The purpose of this study is to make decisions on player selection and tactical formation according to the level of the opponent based on the data, not on the intuition of the manager. In our previous study, the Boruta algorithm was used to extract important features from 69 features in soccer player data by position. The detailed roles of each position were defined by using K-means algorithm. For example, the detailed roles of each position were defined as Mezzala, Shadow Striker, Deep-lying playmaker, and so on. That is, forward positions are classified as Target Man (TM) and Shadow Striker (SS). TM is a high-goal, high-competitive forward, and SS is a high-dribble, high-pass forward. In this study, we analyze a clustering dataset and the game appearance dataset. The game appearance dataset are divided into CL (Champions league Level), EL (Europa league Level), ML (Middle Level), and RL (Relegation Level). Association rule mining algorithm analyzes the synergy between positions, and selects a position with high synergy. Weighted association rule mining algorithm establishes player selection and tactical formation with the weight, which is the player's rating data. Finally, using the obtained results, we visualize the synergy between positions, tactical formation, and player characteristics depending on the level of the opponent.

No abstract

In modern sports, strategy and tactics are important in determining the game outcome. However, many coaches still base their game tactics on experience and intuition. The aim of this study is to predict tactics such as formations, game styles, and game outcome based on soccer dataset. In this paper, we propose to use Deep Neural Networks (DNN) based on Multi-Layer Perceptron (MLP) and feature engineering to predict the soccer tactics of teams. Previous works adopt simple machine learning techniques, such as Support Vector Machine (SVM) and decision tree, to analyze soccer dataset. However, these often have limitations in predicting tactics using soccer dataset. In this study, we use feature selection, clustering techniques for the segmented positions and Multi-Output model for Soccer (MOS) based on DNN, wide inputs and residual connections. Feature selection selects important features among features of soccer player dataset. Each position is segmented by applying clustering to the selected features. The segmented positions and game appearance dataset are used as training dataset for the proposed model. Our model predicts the core of soccer tactics: formation, game style and game outcome. And, we use wide inputs and embedding layers to learn sparse, specific rules of soccer dataset, and use residual connections to learn additional information. MLP layers help the model to generalize features of soccer dataset. Experimental results demonstrate the superiority of the proposed model, which obtain significant improvements comparing to baseline models.