Embedding an Intelligent Tutor into existing Business Software to provide On-the-Job Training

摘要

Traditional on-the-job training for software typically involves human tu-toring, training videos, manuals and open-ended exploration of the software.This Masters Thesis investigates embedding a constraint-based intelligenttutor into an existing business software system to enhance the training op-tions for new users. Chreos Tutor, developed using the ASPIRE authoringsystem, is embedded into Chreos business software. It provides opportunitiesfor users to practice two di erent types of data input tasks in the actual soft-ware environment. The student interface in Chreos Tutor is the combinationof a new tutoring screen and existing data input screens. The main learningtool of a constraint-based tutor is the feedback on performance errors. Inorder to evaluate the e ectiveness of the feedback provided by Chreos Tutor,the experimental group were given feedback pertaining to individual errorsin the submitted solution, whereas the control group received no feedbackwhile working on a task. Analysis of pre-test and post-test results indicatedthat participants who received feedback while working on a task achieveda higher learning gain than participants who were presented with the idealsolution after submitting their solution. This suggests that Chreos Tutor ise ective at teaching data input tasks to new users.