Towards Benchmarking Conversational Intelligence
- Topics:
- Benchmarking
- Tags:
- Agent,
- Benchmarking,
- Business Operations,
- Games,
- Personal Technology,
- Real Estate,
- University Of Technology Sydney
- Source:
- University of Technology Sydney
FREE Registration is required
Overview: This paper presents the motivation and impetus to construct an alternative benchmark to the Turing Test to evaluate the performance of intelligent conversational agents. The criteria that a game problem or test must satisfy are discussed and the Diplomatic Test (DT) is proposed as a suitable candidate worthy of consideration. The implicit requirements of the DT software framework envisioned (currently under development) are outlined. The core tenet of the DT is that it is a more realistic option in providing a standard medium-term research target for constructing intelligent conversational agents, and that it addresses the identified gaps in the Turing Test.
(Is this item miscategorized? Does it need more tags? Let us know.)
Format: PDF | Size: 354KB | Date: Dec 2004 | Pages: 4
People who downloaded this item also downloaded
 |
Stand Tall: A Benchmarking Analysis Can Help You Increase Your Company's Value |
|
Boosting Business Performance Through Benchmarking |
|
Warming Up to Performance Dashboards |
|
When Benchmarks Don't Work |
