06.03.2013 Views

Artificial Intelligence and Soft Computing: Behavioral ... - Arteimi.info

Artificial Intelligence and Soft Computing: Behavioral ... - Arteimi.info

Artificial Intelligence and Soft Computing: Behavioral ... - Arteimi.info

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Procedure create-tree (facts, fuzzy-belief);<br />

Begin<br />

create root of the data-tree;<br />

open the datafile, i.e., the file containing the database;<br />

Repeat<br />

i) Read a fact from the datafile; //after reading, the file pointer increases//<br />

ii) If the corresponding predicate is found at the second level of<br />

the data-tree, Then mark the node Else create a new node at the<br />

second level to represent the predicate <strong>and</strong> mark it;<br />

iii) Search the fact among the children of the marked predicate;<br />

iv) If the fact is not found, Then include it at the third level<br />

as a child of the marked predicate<br />

Until end-of-datafile is reached;<br />

Close the datafile;<br />

End.<br />

Root<br />

sustains-injury (X)<br />

abbreviated si(X)<br />

{si(ram), 0.9} {si(shyam), 0.8} {m(ram,jadu),0.8} {m(lata,asha), 0.6}{m(sita,ram), 0.6} {ha(ram), 0.6}<br />

Fig. 23.8: The data-tree representation of the database.<br />

23.5.3 The Knowledge Base<br />

Murdered (X,Y)<br />

abbreviated m(X,Y)<br />

Has alibi (X)<br />

abbreviated ha(X)<br />

The KB of the proposed ES comprises of a set of default rules (DR) [1], a<br />

set of production rules (PR) <strong>and</strong> two working memory (WM) modules. DR<br />

are used to guess about the suspects. The typical structure [1], [4] of the DR<br />

is as follows:

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!