Loading...
Please wait, while we are loading the content...
Similar Documents
Solving Sudoku Puzzle by Evolutionary Algorithm
| Content Provider | Semantic Scholar |
|---|---|
| Author | Das, Kedar Nath Bhatia, Sumit Puri, Shubin Deep, Kusum |
| Copyright Year | 2016 |
| Abstract | Abstract: ‘Sudoku’ means ‘Single number’. ‘Sudoku Puzzle’ is an interesting and popular Japanese game, where the non-givens need to be filled by a single number (from 1 to 9) provided no repetition occur in the corresponding rows, columns and sub-squares. Solving a Sudoku puzzle is challenging due to its easy rules and difficult inherent phenomenon. Although a number of approaches exist for solving a given Sudoku puzzle, it becomes a challenge among the researchers to solve it by using evolutionary algorithms. In this paper, a Retrievable Genetic Algorithm is proposed to solve a given Sudoku puzzle. A new fitness function is designed with puzzle-character-dependent constraints. The Genetic Algorithm is made “Retrievable”, since the population is reinitialized after a certain number of generations in order to escape from the premature convergence or from being trapped in the local minima. A set of 9 sample puzzles of different difficulty label have been considered for comparison. The superiority of Ret-GA is ensured from the comparative results and discussions. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://atcm.mathandtech.org/EP2016/contributed/4052016_21261.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
