Key to crossword solving : NLP
A. R. Jobin, Anand G Menon, Ashwin Sekhar, Vinay Damodaran · 2017
The struggle of technology to understand natural language has been one of the greatest hurdles of humankind. Various techniques and algorithms have contributed to significant advancements in the field of NLP. One of the most primary challenges that NLP faces is being able to determine the meaning and essence of a sentence which may have multiple variations of syntax and semantics. Crossword solving involves utilizing the knowledge one has over the language and applying it with the various constraints that a clue will provide. As opposed to generic brute force crossword solvers, this work aims to implement automated solving of a crossword using knowledge-based concepts of NLP. The solver provides the user with a potential solution set which contains a collection of words out of which the user may choose the answer which he or she feels is most appropriate. Thus, the entire crossword can be solved by using the knowledge obtained from natural language processing, and this method stays true to the essence of crossword solving.