Frank Linton
The MITRE Corporation
linton@mitre.org
This dataset of ca. 70,000 rows was acquired by monitoring day to day usage of Microsoft Word by more than 20 individuals at the MITRE corporation during the calendar years of 1997-1998. MITRE is a federally funded not-for-profit corporation performing research in the public interest. The users consisted of artificial intelligence engineers, technical staff, and support staff. The users had been employed at MITRE from one to thirty years with a median of eight years. The users worked on Apple Macintosh platforms.
The data format is text, tab delimited. The data are identified and described in the following table:
| Column Label | Description |
| user ID | Unique identifer for each user |
| version of Word | Version of Word in use at the time of logging |
| file size | The size of the file, in bytes, when logging was initiated |
| file date | The date logging was initiated for that file |
| operating system and version | The operating system and version in use at the time of logging |
| command name | The name of the command the user entered. The command name is preceeded by the type of command, i.e., EditSelectAll is the text editing command SelectAll nominally appearing on the Edit menu. |
| command date | The date the user entered the command |
| command time | The time the user entered the command |
Notes:
The data should be treated as an arbitrary sample from a non-randomly selected set of users. No user was logged for the full time period. Users were added as the logger became more robust and were dropped when they changed from Macs to PCs. For more information about the users and the monitoring and logging process, including its limitations, consult the references.
Linton, F., Charron, A., & Joy, D. (1998). OWL: A Recommender System for Organization-Wide Learning. (PDF)
We describe the use of a recommender system to enable continuous knowledge acquisition and individualized tutoring of application software across an organization. Installing such systems will result in the capture of evolving expertise and in organization-wide learning (OWL). We present the results of a year-long naturalistic inquiry into an applications usage patterns, based on logging users actions. We analyze the data to develop user models, individualized expert models, confidence intervals, and instructional indicators. We show how this information could be used to tutor users.
Linton, F., Joy, D., & Schaefer, H. (1999). Building User and Expert Models by Long-Term Observation of Application Usage. (Manuscript submitted for publication)
We describe a new new kind of user model and a new kind of expert model and show how these models can be used to individualize the selection of instructional topics. The new user model is based on observing the individuals behavior in a natural environment over a long period of time, while the new expert model is based on pooling the knowledge of numerous individuals. Individualized instructional topics are selected by comparing an individuals knowledge to the pooled knowledge of her peers.