PROJECT TITLE :

Solving a Complex Language Game by Using Knowledge-Based Word Associations Discovery

ABSTRACT:

“The Guillotine” may be a language game whose goal is to predict the unique word that's linked in some approach to five words given as clues, typically unrelated to every alternative. The flexibility of the human player to seek out the solution depends on the richness of her cultural background. We have a tendency to designed a synthetic player for that game, based mostly on a massive data repository designed by exploiting many sources on the market on the net, like Wikipedia, that give the system with the cultural and linguistic background required to perceive clues. The “brain” of the system is a spreading activation algorithm that starts processing clues, finds associations between them and words among the data repository, and computes a listing of candidate solutions. During this paper we tend to target the problem of finding the most promising candidate answer to be provided as the final answer. We tend to improved the spreading algorithm by means that of two strategies for locating associations also between candidate solutions and clues. Those strategies allow bidirectional reasoning and choose the candidate answer which is the most connected with the clues. Experiments show that the performance of the system is cherish that of average human players.