Name
Abstract | ||
|---|---|---|
In order for artificial intelligence (AI) to develop the same way as any other engineering discipline, it is necessary to have a methodology which starts from a definite set of specifications providing a detailed description of the process to achieve the corresponding implementation. In fact, those specifications are knowledge that the final implementation must display. Thereafter, a suitable verification and validation process must assess compliance with specifications, that are knowledge in this case. This is the way to ensure no disagreement between analysis and synthesis. However, this is not what happens in the current state of the art. Based on a study of causality in the whole computational process, and taking into account the environment where computation takes place, the knowledge disagreement between analysis and synthesis results is discussed. There is knowledge attributed to the implementations that is neither modelled nor implemented computers are usually assisted by human beings who understand the computer outputs and provide the computer inputs interpreting the whole computational process. Examples are illustrated by means of SCHEMA, a knowledge edition interface, and the consideration of causality providing the key that opens the way for computers to really hold knowledge without any human assistance or misinterpretation is discussed. |
Year | DOI | Venue |
|---|---|---|
2000 | 10.1049/ip-sen:20000900 | IEE Proceedings - Software |
Keywords | Field | DocType |
verification,schema,artificial intelligence | Causality,Domain knowledge,Software engineering,Verification and validation,Computer science,Knowledge-based systems,Implementation,Knowledge extraction,Artificial intelligence,Schema (psychology),Computation | Journal |
Volume | Issue | Citations |
147 | 5 | 3 |
PageRank | References | Authors |
0.58 | 5 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Juan Carlos Herrero | 1 | 17 | 3.42 |
| José Mira | 2 | 543 | 71.44 |