KR-20260062274-A - Method and Apparatus for Predicting Sports Lottery Outcomes

KR20260062274AKR 20260062274 AKR20260062274 AKR 20260062274AKR-20260062274-A

Abstract

The present invention discloses a method and apparatus for predicting the outcome of a sports lottery. A method for predicting the outcome of a sports lottery using an ensemble model according to one embodiment may include the steps of: receiving sports result predictions for a sports lottery and lottery-linked match data (hereinafter referred to as "match data") that were performed at any one of a predetermined, selected, current, or past date from a sports lottery-related server (not shown); performing data preprocessing on the match data; training a plurality of match prediction models using the match data on which data preprocessing has been performed; and obtaining a predicted value of the match result using the trained plurality of match prediction models.

Inventors

류항용

Assignees

류항용

Dates

Publication Date: 20260507
Application Date: 20241029

Claims (14)

In a method for predicting the outcome of a sports lottery match in Sports Toto soccer using an ensemble model in which each step is performed by a match prediction device implemented as a computing device, A step of receiving sports result predictions for a sports lottery and lottery-linked match data (hereinafter referred to as "match data") performed during any one of a predetermined, selected, present, or past period (hereinafter referred to as "N period") from a sports lottery-related server (not shown); A step of performing data preprocessing on the above match data; A step of training a plurality of match prediction models using match data on which the above data preprocessing has been performed; and It includes a step of obtaining a predicted value of the match result using multiple learned match prediction models, and The weights are values set for each of the plurality of match prediction models to maximize the prediction accuracy of the predicted values, and are characterized by being determined based on the match data. The step of performing the above data preprocessing is, A step of calculating team-specific statistical data from the above match data; and The method includes the step of performing normalization on the above-mentioned team-specific statistical data and labeling it with integer values. The step of performing the above labeling includes the step of classifying each team included in the above match data into a home team and an away team and performing labeling. How to predict the outcome of a sports lottery.
In paragraph 1, The step of receiving game data for a sports lottery performed during the above N period is, The method includes the step of receiving game data for a sports lottery performed during the above N period in the form of time-series data arranged in chronological order. How to predict the outcome of a sports lottery.
In paragraph 1, The step of training the above-mentioned multiple match prediction models is, A step of inputting match data for one or more first match seasons performed during the above N period into the plurality of match prediction models as training data to obtain output values for the prediction of match results corresponding to the second match season performed during the above N period; and A method comprising the step of training the plurality of match prediction models based on output values for the prediction of match results corresponding to the second match season and the match results for the second match season. How to predict the outcome of a sports lottery.
In paragraph 3, The step of training the above-mentioned multiple match prediction models is, A method comprising the step of calculating a result value by applying a weight corresponding to each of the plurality of match prediction models to each of the output values corresponding to each of the plurality of match prediction models, and determining the parameters of each of the plurality of match prediction models and the weight based on the result value and the training data. How to predict the outcome of a sports lottery.
In paragraph 4, The step of determining the parameters and weights of each of the plurality of match prediction models is A step comprising determining the parameters of each of the plurality of match prediction models and the weights corresponding to each of the plurality of match prediction models based on the above result value and the match results for the second match season. How to predict the outcome of a sports lottery.
In paragraph 1, The above multiple match prediction models are, Including at least two of MLP (Multilayer Perceptrons), LSTM (Long Short-Term Memory), SVM (Support Vector Machine), Random Forest, and Logistic Regression, How to predict the outcome of a sports lottery.
In paragraph 3, The above first game season is, A game season conducted during a period N prior to the aforementioned second game season, How to predict the outcome of a sports lottery.
In paragraph 1, The step of obtaining the above predicted value is, A step comprising obtaining a predicted value of the match result for a third match season to be performed in the future beyond the second match season, How to predict the outcome of a sports lottery.
In paragraph 8, The step of obtaining the above predicted value is, A step of calculating a weighted sum with each weight applied based on output values obtained from multiple learned match prediction models; and A step comprising obtaining a predicted value of the match result for the third match season based on the above weighted sum, How to predict the outcome of a sports lottery.
In paragraph 1, The above sports lottery is, Sports lotteries linked to sports result prediction and lotteries, such as the aforementioned sports (soccer, baseball, basketball, volleyball, golf, wrestling, etc.) and lotteries (Sports Toto, lottery, horse racing ticket, cycling race ticket, boat race ticket, cattle ticket, etc., which are called sports lotteries), How to predict the outcome of a sports lottery.
In paragraph 1, The above team-specific statistical data is, The above-mentioned team-specific statistical data includes at least one of the lottery name, lottery date, lottery number, odds, odds rate, match name, player number, match date, record, win percentage, save percentage, highest record, lowest record, ball possession, number of shots on target, total number of shots, total number of ball touches, total number of passes, total number of tackles, total number of successful saves, sports lottery match result, and league ranking of the sports lottery match related to the match data for each team corresponding to the above-mentioned team-specific statistical data. How to predict the outcome of a sports lottery.
In a sports toto soccer match prediction device using an ensemble model that performs a sports lottery match prediction method, A receiving unit that receives sports result predictions for a sports lottery and lottery-linked match data (hereinafter referred to as "match data") performed during any one of a predetermined, selected, present, or past period (hereinafter referred to as "N period") from a sports lottery-related server (not shown); A preprocessing unit that performs data preprocessing on the above game data; Multiple match prediction models that obtain predicted match results using multiple learned match prediction models; and It includes a learning unit that trains a plurality of match prediction models using match data on which the above data preprocessing has been performed, and The weights are values set for each of the plurality of match prediction models to maximize the prediction accuracy of the predicted values, and are characterized by being determined based on the match data. The above preprocessing unit is, Calculate team-specific statistical data from the above match data, perform normalization on the calculated team-specific statistical data, and perform labeling with integer values. In the process of performing the above labeling, based on the above team-specific statistical data, each team corresponding to the above team-specific statistical data is classified into home team and away team to perform labeling. Match prediction device.
In Paragraph 12, The above data receiving unit receives game data for a sports lottery performed during the above N period in the form of time-series data arranged in chronological order. Match prediction device.
In Paragraph 12, The above multiple match prediction models are, Including at least two of MLP, LSTM, SVM, Random Forest, and Logistic Regression, Match prediction device.

Description

Method and Apparatus for Predicting Sports Lottery Outcomes The following embodiments relate to a technology for predicting the outcome of a sports lottery. Generally, the information and knowledge related to sports lotteries, which link sports result prediction, are vast and complex. Since predicting sports lotteries involves data that only each sports lottery club can receive or requires significant cost and time to collect, it has been nearly impossible for general users. Machine learning is a type of Artificial Intelligence (AI) that refers to the process in which a computer performs prediction tasks, such as regression, classification, and clustering, based on what it has learned autonomously from data. Deep learning is a field of machine learning that teaches computers human thinking and can be defined as a set of machine learning algorithms that attempt high levels of abstraction (the task of summarizing key content or functions from a large amount of data or complex materials) through a combination of various non-linear transformation techniques. Deep learning architecture is a concept designed based on artificial neural networks (ANNs). An artificial neural network is an algorithm that aims to achieve learning capabilities similar to the human brain by mathematically modeling and simulating virtual neurons, and it is primarily used for pattern recognition. The artificial neural network models used in deep learning have a structure built by repeatedly stacking linear fitting and nonlinear transformations (or activations). Examples of neural network models used in deep learning include Deep Neural Networks (DNNs), Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Restricted Boltzmann Machines (RBMs), Deep Belief Networks (DBNs), and Deep Q-Networks. Meanwhile, although neural network-based prediction models have been utilized to predict sports lottery results, existing methods have primarily calculated prediction values using a single prediction model. There are clear limitations to accurately predicting sports lottery results using a single prediction model. Therefore, research is needed to address these limitations. FIG. 1 is a diagram illustrating the overall configuration of a match prediction system according to one embodiment. FIG. 2 is a flowchart for explaining the operation of a sports lottery match prediction method according to one embodiment. FIG. 3 is a diagram illustrating an operation for obtaining a predicted value of a game result according to one embodiment. FIG. 4 is a diagram illustrating the configuration of a match prediction device according to one embodiment. Hereinafter, embodiments are described in detail with reference to the attached drawings. However, various modifications may be made to the embodiments, and thus the scope of the patent application is not limited or restricted by these embodiments. It should be understood that all modifications, equivalents, and substitutions to the embodiments are included within the scope of the rights. The terms used in the embodiments are for illustrative purposes only and should not be interpreted as intended to be limiting. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as "comprising" or "having" are intended to indicate the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof. Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the embodiments pertain. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application. In addition, when describing with reference to the attached drawings, identical components are assigned the same reference numeral regardless of drawing symbols, and redundant descriptions thereof are omitted. In describing the embodiments, if it is determined that a detailed description of related prior art could unnecessarily obscure the essence of the embodiments, such detailed description is omitted. Before proceeding with the explanation, in the embodiments of the present invention, a sports lottery is defined as a sports game lottery linked to sports result prediction, such as sports (soccer, baseball, basketball, volleyball, golf, wrestling, etc.), Sports Toto, lottery, horse racing ticket, cycling race ticket, boat race ticket, and cattle ticket. FIG. 1 is a diagram illustrating the overall configuration of a match prediction system acc