This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
|
hekate:xtt [2008/04/30 13:49] gjn |
hekate:xtt [2008/04/30 13:55] gjn |
||
|---|---|---|---|
| Line 1: | Line 1: | ||
| - | ====== The XTT Knowledge Representation ====== | ||
| - | |||
| - | ===== Introduction ===== | ||
| - | |||
| - | XTT is a knowledge representation formalism for rules. | ||
| - | |||
| - | XTT allows for structuralization of the rules base, by introducing: | ||
| - | * tables, used, to group rules. | ||
| - | * intertable links, used to provide inference control between tables. | ||
| - | |||
| - | Rules use an expressive attribute language. | ||
| - | |||
| - | So going deeper, general issues are as follows: | ||
| - | * the [[xtt#Attributive Language]] on which the rule is built, | ||
| - | * the [[xtt#Rule syntax]], and semantics, what it means to fire a //single// XTT rule, | ||
| - | * the table structure, how we encode a table using XTT rules | ||
| - | * the inference control, how do we interpret a table, and a set of tables | ||
| - | |||
| - | Now, considering a system, containing knowledge expressed in XTT, some other issues are: | ||
| - | * environment interaction: how do we exchange information with the env? | ||
| - | * interpreter model: when, how, do we run/query the XTT rulebase | ||
| - | |||
| - | Some issues include: | ||
| - | * XML serialization of the XTT model | ||
| - | * Prolog representation | ||
| - | |||
| - | ===== Attributive language ===== | ||
| - | |||
| - | ===== Rule syntax ===== | ||
| - | ===== Table structure ===== | ||
| - | ===== Inference control ===== | ||
| - | ===== Environmental interaction ===== | ||
| - | ===== Interpreter model ===== | ||
| - | ===== XML serialization ===== | ||
| - | ===== Prolog representation ===== | ||
