Automatic compact vocabulary crossword generator using multi-objective evolutionary optimization

One of the best tools for learning or mastering a language by students is to use vocabulary crossword puzzles. It is desired to put as many words as possible on a small-sized puzzle. However, designing a compact crossword puzzle using a limited vocabulary is a complex task that usually requires manual participation. In this study, we report the successful application of multi-objective evolutionary optimization algorithms to generate compact crosswords. The algorithm tries to find the best possible words from the predefined vocabulary such that the required space for the puzzle is minimized, i.e., more words can be placed in a compact puzzle. The whole process is automatic and at the end of the algorithm, the set of puzzle keys and clues on the Pareto front of the algorithm solution are generated. The effectiveness of the proposed method is proved through a real-word experiment to generate crossword puzzles from a vocabulary in the field of computer science.

کلیدواژه ها:

automated puzzle design ، crossword puzzle ، vocabulary puzzles ، multi-objective optimization algorithm.