CLAS : an approach for full use of application software
Kenichi Matsumoto, Shuuji Morisaki, Akito Monden, Koji Torii · Institutional Repositories DataBase (IRDB) · 2001
Many of the latest application software provide users with a large number of functions; however, the full set of functions is too large for diverse users. Most of users usually use only 10% of the full set of functions provided by application software. This paper describes an approach “Community-based Learning for application software (CLAS)” to let users learn useful functions at low cost for the purpose of improving the users' productivity in using application software. In the CLAS, community members mutually compare their own histories of software function executions, which can be collected automatically, in order to identify a set of candidate functions that should be learned. It is based on the assumption that among the functions that each member of a community uses, some differences in the software usage experiences of each member are observed and the function belonging to the difference is useful for the member to perform his/her current task. The experimental results with a prototype system show both novice and experienced members can more easily obtain usage knowledge of the new functions by the CLAS than a conventional online help system. More than half of functions learned by using the CLAS are new functions whose existences were unknown to the members and cannot be found by the conventional online help system. In the CLAS, users can share the knowledge of software usage in a systematic and effective way.