Retrieval of similar chess positions

Abstract

We address the problem of retrieving chess game positions similar to a given query position from a collection of archived chess games. We investigate this problem from an information retrieval (IR) perspective. The advantage of our proposed IR-based approach is that it allows using the standard inverted organization of stored chess positions, leading to an efficient retrieval. Moreover, in contrast to retrieving exactly identical board positions, the IR-based approach is able to provide approximate search functionality. In order to define the similarity between two chess board positions, we encode each game state with a textual representation. This textual encoding is designed to represent the position, reachability and the connectivity between chess pieces. Due to the absence of a standard IR dataset that can be used for this search task, a new evaluation benchmark dataset was constructed comprising of documents (chess positions) from a freely available chess game archive. Experiments conducted on this dataset demonstrate that our proposed method of similarity computation, which takes into account a combination of the mobility and the connectivities between the chess pieces, performs well on the search task, achieving MAP and nDCG values of 0.4233 and 0.6922 respectively.

Chess, visual memory and geometric transformations

“…These questions highlight the importance of visualization that can interfere with issues such as geometric transformations when using chess as the main tool, because chess has capabilities such as the development of visual memory, spatiality, logic, among others (Blanco, 2020). Likewise, the works on chess focus mainly on the mathematical modeling of the game (Ganguly, Leveling & Jones, 2014), but another large number diverts the gaze of how this game helps to improve the cognitive abilities of the players, on topics such as the identification of regular figures (octagons or other polygons) through the movement of the pieces and their connectivity with others (Subia, 2020).…”

“…The chess player not only generally remembers his games taking into account similar positions (Ganguly, Leveling & Jones, 2014), but he also pays attention to particular details that help to classify his information, such as endings, sacrifices, material advantage, the positionality. In addition to the lines of action or active zones, which are the set of accessible squares where a piece can be moved from its particular position.…”

“…Therefore, SGCS are sensitive to conceive modifications between the pieces. These connections are made up of lines of action or active zones (Ganguly, Leveling & Jones, 2014).…”

“…As the chess player has his mental database thanks to his experience with the game, he has formed generalizations in his head (Ganguly, Leveling & Jones, 2014), so in certain http://journals.ums.ac.id/index.php/jramathedu positions he will resort to this information. Thus, a chess player shows a great memory capacity (Schneider & Shiffrin, 1977).…”

Domain Specific Named Entity Recognition Referring to the Real World by Deep Neural Networks

“…Similar to bio-medical NEs (Settles, 2004;Tateisi et al, 2002), these NEs are useful for applications in the game domain. Our method could be used to improve automatic game commentary systems (Kameko et al, 2015b;Chen et al, 2010) or to build a state search method that uses natural language queries instead of state notations (Ganguly et al, 2014). In addition to these interesting applications, game states have another advantage for NLP research.…”