Building expert systems
MonitoringKnowledge Acquisition Instead of Evaluating Knowledge Bases
EKAW '00 Proceedings of the 12th European Workshop on Knowledge Acquisition, Modeling and Management
| Hi-index | 0.00 |
In the mid-80's when I joined the Texas Instruments Defense Systems Artificial Intelligence Lab, expert systems were a solution looking for a problem. My first assignment as a knowledge engineer was knowledge acquisition for a small expert system supporting an internal TI project. Preparing for the task, I reviewed texts [1][2][3] and journal articles [4][5] in the lab's small but growing technical library. Most texts agreed that effective knowledge acquisition was essential and that the process could be a bottleneck in system development. But the how-to, what-to, and how-fix were often missing. In an effort to supplement the limited material, the lab made available to its fledgling knowledge engineers internal guidelines to aid in beginning knowledge acquisition [6][7][8].