Learning and Performing by Exploration: Label Quality Measured by Latent Semantic Analysis

Models of learning and performing by exploration assume that the semantic distance between task descriptions and screen labels controls in part the usersÕ search strategies. Nevertheless, none of the models has an objective way to compute semantic distance. In this study, participants performed twelve tasks by exploration and were tested for recall after a 1-week delay. Latent Semantic Analysis was used to compute the semantic similarity between the task descriptions and the labels in the applicationÕs menu system. When the labels were close in the semantic space to the task descriptions, subjects performed the tasks faster. LSA could be incorporated into any of the current models, and it could be used to automate the evaluation of computer applications for ease of learning and performing by exploration. Keywords Learning by exploration, label-following strategy, cognitive models, semantic similarity, latent semantic analysis

• Type: Conference
• Year: 1999
• Where: CHI
• Authors: Rodolfo Soto
• Keywords: CHI 1999, Human Computer Interaction, Latent Semantic Analysis, Semantic, Task Descriptions