GMap: Supporting Free-Form Text Entry of Game Titles in Games User Research

The input of free-form text is frequently utilized in surveys for game-related research. While this provides flexibility, it also presents the challenge of dirty data, which includes spelling errors, missing series titles, and unofficial yet popular abbreviations inputted by the user. The manual resolution of these anomalies is impractical and resource-intensive. To address this issue, a fuzzing string machine-based game mapping system was designed and evaluated using 1,096 game titles input by users. GMap-R, a real-time autocomplete game title system to aid runtime user entry, was also created and evaluated using 150 game titles provided by 30 participants, each of whom listed their five favorite games twice. With GMap-R, the correct mapping percentage increased to 98.67%. These preliminary evaluations indicate that the proposed strategy can significantly enhance the cleansing and input of game titles’ free-form text. In turn, this helps to conserve resources when obtaining unsupervised data through online studies.