Method for producing go questions

By collecting and analyzing Go game data and using algorithms to generate Go questions, the problems of low generation efficiency, limited quantity and poor difficulty grading in the existing technology are solved, efficient and personalized question generation are achieved, and the Go learning effect is improved and cultural communication.

CN119005880APending Publication Date: 2024-11-22YOUTH ZONGHENG (JIANGSU) TECH DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410972685.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-19
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The existing Go questions are inefficient in generation, limited in number, and the difficulty grading is not meticulous enough to meet the needs of large-scale teaching and training, which affects the learning effect.

Method used

Collect a large amount of game data through the Internet or Go competition database, use advanced chess game analysis algorithm to extract key features, design Go question generation algorithm, and automatically generate questions of different difficulty levels, including key chess game detection, difficulty evaluation and question generation modules, providing online and Question bank for offline use.

Benefits of technology

It has improved the efficiency and quantity of questions to generate, achieved refinement of difficulty grading, provided personalized learning experience, enriched learning content, improved learning effect, expanded the audience of Go education, and promoted the development of Go culture and industry.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a method for producing go questions, which comprises the steps of data acquisition, data analysis, question generation algorithm and question output, and is characterized in that the go questions are automatically generated through the algorithm, so that the time and labor cost for manually writing the questions are greatly reduced; difficulty and types of questions can be quickly adjusted according to user feedback and data analysis results, and adaptability and flexibility of question generation are improved; customized questions are provided according to segments and skill levels of players, and the players are helped to learn and progress at the difficulty suitable for themselves; according to the automatic go question generation method, various types of questions including death and livability, hand tendons, layout and the like are generated, the learning content is enriched, the requirements of different players are met, the go learning efficiency and effect can be improved, go culture propagation and go industry development can be promoted, and remarkable social and economic values are achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field related to go education and training, and specifically relates to a method for producing go questions. Background Art

[0002] The generation of existing Go questions usually relies on manual writing, which has the following shortcomings and defects:

[0003] Low efficiency: Manually writing Go questions requires a lot of time and effort. The writer not only needs to have a high level of Go skills, but also needs to have creative thinking to design questions of various types and difficulties. This process is complicated and tedious, resulting in very low efficiency in question generation.

[0004] Limited number of questions: Since writing Go questions requires a high level of expertise and a lot of time, human resource limitations make it impossible to increase the number of questions on a large scale. Existing Go question banks are often limited in number and cannot meet the needs of large-scale Go teaching and training, especially in the context of the rapid development of modern Go education. This problem is particularly prominent.

[0005] Difficulty grading is not precise and is subjective: The difficulty grading of manually written Go questions usually relies on the subjective judgment of the author, which is difficult to be accurate and consistent. This subjectivity leads to a rough difficulty grading of questions, which is difficult to accurately match the needs of players of different levels. For beginners and advanced players, questions without precise difficulty grading are difficult to provide effective training, which affects learning and progress. Summary of the invention

[0006] The purpose of the present invention is to provide a method for producing Go problems to solve the problems of low efficiency, limited number of problems, and insufficient and subjective difficulty grading proposed in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for generating Go questions, the method is as follows:

[0009] Step 1: Data Collection

[0010] Collect a large amount of Go game data from the Internet or Go competition database as the basis for question generation. The data includes but is not limited to the ranking information of both players, game records, and win and loss results;

[0011] Step 2: Data Analysis

[0012] Use advanced chess game analysis algorithms to process the collected game data and extract key features. The analysis includes:

[0013] The rank of both players: understand the level of the players so as to facilitate the difficulty classification of the questions;

[0014] Win rate and point difference of chess board situations: Evaluate the situational advantage to determine the key points of the question;

[0015] Key moves on the chessboard: Identify the moves that have a decisive impact on the game;

[0016] Winning or losing result: Determine the final result of the game;

[0017] Step 3: Question generation algorithm

[0018] Design and implement a Go problem generation algorithm, and automatically generate Go problems of different difficulty levels based on data analysis results;

[0019] Step 4: Output the title

[0020] Output the generated questions to the database or Go teaching software for users to use. The output forms may include:

[0021] Online question bank: Through Go teaching software or websites, users can access and practice generated questions;

[0022] Question bank file: export questions into standard format files for offline use and printing;

[0023] Application integration: Integrate with existing Go teaching and training applications to achieve seamless connection.

[0024] Preferably, the algorithm includes the following main parts: a key chess game detection module, a difficulty assessment module and a question generation module.

[0025] Preferably, the key chess game detection module: identifies and classifies typical chessboard layouts, and selects representative situations as the basis for the questions, specifically:

[0026] The goal of the key chess game detection module is to identify and classify typical chess board layouts in Go games, and select representative situations as the basis for questions. The steps are as follows

[0027] S1.1 Data preprocessing

[0028] Collect a large amount of Go game data, each game contains the board states of multiple steps;

[0029] Represent each chessboard state as a fixed-size two-dimensional array (e.g., a 19x19 chessboard), where each element indicates whether there is a chess piece at that position and its color;

[0030] S1.2 Feature extraction

[0031] For each chessboard state, extract key features. For example, the entire chessboard state can be considered as a feature vector, and the length of the vector is equal to the number of positions on the chessboard (for example, 19x19=361);

[0032] S1.3 Clustering

[0033] Cluster all extracted feature vectors using a clustering algorithm (e.g. KMeans);

[0034] The purpose of clustering is to group similar chessboard layouts into a group, where each cluster represents a typical chessboard layout;

[0035] Set the number of clusters to be clustered, and decide how many typical layouts to identify according to your needs;

[0036] S1.4 Select a representative layout

[0037] For each cluster, calculate the distance from the cluster center to all points in the cluster;

[0038] The points closest to the cluster center are selected as representative layouts. These points represent the most representative board states in each cluster.

[0039] S1.5 Output results

[0040] The selected representative chess games are output as key chess games.

[0041] Preferably, a difficulty assessment module: assesses the difficulty of a question based on the board layout and key moves, and finely grades the difficulty of the question by analyzing factors such as the complexity of the game and the diversity of solutions;

[0042] The specific steps are as follows:

[0043] S2.1 Feature extraction

[0044] Extract key features from the board layout. These features can include: the distribution of black and white pieces on the board, the location and importance of key moves, the existence of complex situations such as ko fights, live pieces and dead pieces;

[0045] S2.2 Complexity analysis, calculate the complexity of the chess game. The complexity can be measured by the following indicators: the total number of pieces on the board, the density of pieces in key areas, the number of ko fights, and the number of live and dead pieces;

[0046] S2.3 Solution diversity analysis: Analyze the diversity of solutions. Diversity can be evaluated by: counting the number of possible solution paths, evaluating the number of steps and complexity of each solution path, and considering the results and impacts of different solution paths;

[0047] S2.4 Difficulty score, a difficulty score is assigned to each chessboard layout based on the complexity and solution diversity. The score can be calculated based on the following formula: Difficulty score = complexity score + solution diversity score;

[0048] S2.5 outputs the result and attaches the evaluated difficulty score to each chessboard layout as the basis for grading the difficulty of the question.

[0049] Preferably, a question generation module generates questions of corresponding difficulty according to the difficulty assessment results and provides detailed problem-solving solutions. This module can automatically generate hints and answers to help users better understand and solve the questions.

[0050] The goal of the question generation module is to generate Go questions of corresponding difficulty based on the difficulty assessment results and provide detailed solutions. This module needs to automatically generate hints and answers to help users better understand and solve the questions. The specific steps are as follows:

[0051] S3.1 input the difficulty evaluation result, and receive the chessboard layout after difficulty evaluation and its corresponding difficulty score;

[0052] S3.2 generating questions, for each selected chessboard layout, generating a corresponding question according to its difficulty score;

[0053] Primary questions: provide clear win or loss results (e.g. Black goes first, White loses);

[0054] Intermediate questions: describe complex situations and require answers to key steps (e.g., find the robbing point and win);

[0055] Advanced questions: A complete situation is given and players are required to analyze and develop a strategy (e.g., find the best move to gain an advantage).

[0056] Compared with the prior art, the present invention provides a method for generating Go problems, which has the following beneficial effects:

[0057] Improve efficiency

[0058] Automatic generation: Go questions are automatically generated through algorithms, greatly reducing the time and labor cost of manually writing questions;

[0059] Fast iteration: The difficulty and type of questions can be adjusted quickly based on user feedback and data analysis results, improving the adaptability and flexibility of question generation;

[0060] Personalized Learning

[0061] Difficulty grading: Provide customized questions based on the player's rank and skill level, helping players learn and improve at a difficulty level that suits them;

[0062] Diversified questions: Generate various types of questions, including life and death, tricks, layout, etc., to enrich the learning content and meet the needs of different players;

[0063] Improve learning outcomes

[0064] Accurate analysis: Using data analysis technology to deeply explore the characteristics of chess games and generate more challenging and educational questions;

[0065] Instant feedback: Combined with teaching software, it provides instant feedback and suggestions on problem solving, helping players correct mistakes in time and deepen their understanding;

[0066] Expand your audience

[0067] Popularize Go education: Through online question banks and teaching software, more people can learn Go and promote Go culture;

[0068] Lower the threshold: Provide easy-to-understand questions and tutorials for beginners to reduce the difficulty of learning Go;

[0069] Promoting research and development

[0070] Data accumulation: A large amount of Go problem and solution data can be used for Go AI research and training, promoting the development of Go AI technology;

[0071] Algorithm optimization: By continuously iterating the question generation algorithm, we can explore more efficient question generation methods and improve the intelligence level of the algorithm;

[0072] Economic Benefits

[0073] Business model innovation: Develop paid Go question services to provide high-quality learning materials for professional players and Go enthusiasts;

[0074] Increase revenue sources: By cooperating with Go teaching institutions or selling question generation software, new revenue channels can be created for the company;

[0075] Fine grading: Through a scientific difficulty assessment and grading system, the difficulty of questions is graded finely and objectively, which can accurately match the training needs of players of different levels and improve the training effect;

[0076] In summary, the automated Go question generation method can not only improve the efficiency and effectiveness of Go learning, but also promote the dissemination of Go culture and the development of the Go industry, and has significant social and economic value. DETAILED DESCRIPTION

[0077] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0078] The present invention provides a technical solution:

[0079] A method for generating Go questions, the method is as follows:

[0080] Step 1: Data Collection

[0081] Collect a large amount of Go game data from the Internet or Go competition database as the basis for question generation. The data includes but is not limited to the ranking information of both players, game records, and win and loss results;

[0082] Step 2: Data Analysis

[0083] Use advanced chess game analysis algorithms to process the collected game data and extract key features. The analysis includes:

[0084] The rank of both players: understand the level of the players so as to facilitate the difficulty classification of the questions;

[0085] Win rate and point difference of chess board situations: Evaluate the situational advantage to determine the key points of the question;

[0086] Key moves on the chessboard: Identify the moves that have a decisive impact on the game;

[0087] Winning or losing result: Determine the final result of the game;

[0088] Step 3: Question generation algorithm

[0089] Design and implement a Go problem generation algorithm, and automatically generate Go problems of different difficulty levels based on data analysis results;

[0090] Step 4: Output the title

[0091] Output the generated questions to the database or Go teaching software for users to use. The output forms may include:

[0092] Online question bank: Through Go teaching software or websites, users can access and practice generated questions;

[0093] Question bank file: export questions into standard format files for offline use and printing;

[0094] Application integration: Integrate with existing Go teaching and training applications to achieve seamless connection.

[0095] The algorithm consists of the following main parts: key chess game detection module, difficulty assessment module and question generation module.

[0096] Key chess game detection module: Identify and classify typical chess board layouts, and select representative situations as the basis for questions. Specifically:

[0097] The goal of the key chess game detection module is to identify and classify typical chess board layouts in Go games, and select representative situations as the basis for questions. The steps are as follows

[0098] S1.1 Data preprocessing

[0099] Collect a large amount of Go game data, each game contains the board states of multiple steps;

[0100] Represent each chessboard state as a fixed-size two-dimensional array (e.g., a 19x19 chessboard), where each element indicates whether there is a chess piece at that position and its color;

[0101] S1.2 Feature extraction

[0102] For each chessboard state, extract key features. For example, the entire chessboard state can be considered as a feature vector, and the length of the vector is equal to the number of positions on the chessboard (for example, 19x19=361);

[0103] S1.3 Clustering

[0104] Cluster all extracted feature vectors using a clustering algorithm (e.g. KMeans);

[0105] The purpose of clustering is to group similar chessboard layouts into a group, where each cluster represents a typical chessboard layout;

[0106] Set the number of clusters to be clustered, and decide how many typical layouts to identify according to your needs;

[0107] S1.4 Select a representative layout

[0108] For each cluster, calculate the distance from the cluster center to all points in the cluster;

[0109] The points closest to the cluster center are selected as representative layouts. These points represent the most representative board states in each cluster.

[0110] S1.5 Output results

[0111] The selected representative chess games are output as key chess games.

[0112] Difficulty Assessment Module: Assess the difficulty of questions based on the board layout and key moves, and finely grade the difficulty of questions by analyzing factors such as the complexity of the game and the diversity of solutions;

[0113] The specific steps are as follows:

[0114] S2.1 Feature extraction

[0115] Extract key features from the board layout. These features can include: the distribution of black and white pieces on the board, the location and importance of key moves, the existence of complex situations such as ko fights, live pieces and dead pieces;

[0116] S2.2 Complexity analysis, calculate the complexity of the chess game. The complexity can be measured by the following indicators: the total number of pieces on the board, the density of pieces in key areas, the number of ko fights, and the number of live and dead pieces;

[0117] S2.3 Solution diversity analysis: Analyze the diversity of solutions. Diversity can be evaluated by: counting the number of possible solution paths, evaluating the number of steps and complexity of each solution path, and considering the results and impacts of different solution paths;

[0118] S2.4 Difficulty score, a difficulty score is assigned to each chessboard layout based on the complexity and solution diversity. The score can be calculated based on the following formula: Difficulty score = complexity score + solution diversity score;

[0119] S2.5 outputs the result and attaches the evaluated difficulty score to each chessboard layout as the basis for grading the difficulty of the question.

[0120] Question generation module: Generate questions of corresponding difficulty based on the difficulty assessment results and provide detailed problem-solving solutions. This module can automatically generate hints and answers to help users better understand and solve questions;

[0121] The goal of the question generation module is to generate Go questions of corresponding difficulty based on the difficulty assessment results and provide detailed solution plans. This module needs to automatically generate hints and answers to help users better understand and solve the questions. The specific steps are as follows:

[0122] S3.1 input the difficulty evaluation result, and receive the chessboard layout after difficulty evaluation and its corresponding difficulty score;

[0123] S3.2 generating questions, for each selected chessboard layout, generating a corresponding question according to its difficulty score;

[0124] Primary questions: provide clear win or loss results (e.g. Black goes first, White loses);

[0125] Intermediate questions: describe complex situations and require answers to key steps (e.g., find the robbing point and win);

[0126] Advanced questions: A complete situation is given and players are required to analyze and develop a strategy (e.g., find the best move to gain an advantage).

[0127] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for producing a Go problem, characterized in that: Here’s how: Step 1: Data Collection Collect a large amount of Go game data from the Internet or Go competition database as the basis for question generation. The data includes but is not limited to the ranking information of both players, game records, and win and loss results; Step 2: Data Analysis Use advanced chess game analysis algorithms to process the collected game data and extract key features. The analysis includes: The rank of both players: understand the level of the players so as to facilitate the difficulty classification of the questions; Win rate and point difference of chess board situations: Evaluate the situational advantage to determine the key points of the question; Key moves on the chessboard: Identify the moves that have a decisive impact on the game; Winning or losing result: Determine the final result of the game; Step 3: Question generation algorithm Design and implement a Go problem generation algorithm, and automatically generate Go problems of different difficulty levels based on data analysis results; Step 4: Output the title Output the generated questions to the database or Go teaching software for users to use. The output forms may include: Online question bank: Through Go teaching software or websites, users can access and practice generated questions; Question bank file: export questions into standard format files for offline use and printing; Application integration: Integrate with existing Go teaching and training applications to achieve seamless connection.

2. The method for producing a Go problem according to claim 1, characterized in that: The algorithm consists of the following main parts: key chess game detection module, difficulty assessment module and question generation module.

3. A method for producing Go problems according to claim 2, characterized in that: Key chess game detection module: Identify and classify typical chess board layouts, and select representative situations as the basis for questions. Specifically: The goal of the key chess game detection module is to identify and classify typical chess board layouts in Go games, and select representative situations as the basis for questions. The steps are as follows S1.1 Data preprocessing Collect a large amount of Go game data, each game contains the board states of multiple steps; Represent each chessboard state as a fixed-size two-dimensional array, where each element indicates whether there is a chess piece at that position and its color; S1.2 Feature extraction For each chessboard state, extract key features. The entire chessboard state can be regarded as a feature vector, and the length of the vector is equal to the number of positions on the chessboard. S1.3 Clustering Clustering all extracted feature vectors using a clustering algorithm; The purpose of clustering is to group similar chessboard layouts into a group, where each cluster represents a typical chessboard layout; Set the number of clusters to be clustered, and decide how many typical layouts to identify according to your needs; S1.4 Select a representative layout For each cluster, calculate the distance from the cluster center to all points in the cluster; The points closest to the cluster center are selected as representative layouts. These points represent the most representative board states in each cluster. S1.5 Output results The selected representative chess games are output as key chess games.

4. The method for producing a Go problem according to claim 2, characterized in that: Difficulty Assessment Module: Assess the difficulty of questions based on the board layout and key moves, and finely grade the difficulty of questions by analyzing factors such as the complexity of the game and the diversity of solutions; The specific steps are as follows: S2.1 Feature extraction Extract key features from the board layout, including: the distribution of black and white pieces on the board, the location and importance of key moves, the existence of complex situations such as ko, live pieces, and dead pieces; S2.2 Complexity analysis, calculate the complexity of the chess game. The complexity can be measured by the following indicators: the total number of pieces on the board, the density of pieces in key areas, the number of ko fights, and the number of live and dead pieces; S2.3 Solution diversity analysis: Analyze the diversity of solutions. Diversity can be evaluated by: counting the number of possible solution paths, evaluating the number of steps and complexity of each solution path, and considering the results and impacts of different solution paths; S2.4 Difficulty score, a difficulty score is assigned to each chessboard layout based on the complexity and solution diversity. The score can be calculated based on the following formula: Difficulty score = complexity score + solution diversity score; S2.5 outputs the result and attaches the evaluated difficulty score to each chessboard layout as the basis for grading the difficulty of the question.

5. The method for producing a Go problem according to claim 2, characterized in that: Question generation module: Generate questions of corresponding difficulty based on the difficulty assessment results and provide detailed problem-solving solutions. This module can automatically generate hints and answers to help users better understand and solve questions; The goal of the question generation module is to generate Go questions of corresponding difficulty based on the difficulty assessment results and provide detailed solution plans. This module needs to automatically generate hints and answers to help users better understand and solve the questions. The specific steps are as follows: S3.1 inputs the difficulty evaluation result, and receives the chessboard layout after the difficulty evaluation and its corresponding difficulty score; S3.2 generating questions, for each selected chessboard layout, generating a corresponding question according to its difficulty score; Primary questions: provide clear win or loss results; Intermediate questions: describe complex situations and require answers to key steps; Advanced questions: A complete situation is given, requiring players to analyze and develop strategies.