I think what are asking for would be some book on AI. Maybe you would decide whether to investigate rule-based systems (see Wikipedia entry on expert systems) or trainable systems (eg neural nets). When you decide on the rule based system, you will decide on the kind of logic to use, whether to have rule chaining, backwards or forwards.
Then you will usually decide that the easiest thing for people to.use is a rule-based system without chaining, and thence usually to schematron.
Hi Folks,
Would you recommend an algorithm’s book please? Specifically, a book containing algorithms about how to determine if an input (that conforms to a grammar) has property P.
Allow me to explain.
Suppose I have a grammar for a Book. I could use XSD, or RNG, or Backus-Naur, or EBNF, or something else to express the grammar. Suppose the grammar expresses this:
Book must contain a Title, one or
more Authors, a Date, an ISBN, and
a Publisher.
Many instances are then created. The instances could be expressed in XML, or JSON, or many other ways. Here’s an instance, expressed in XML:
<Book>
<Title>Introduction to Graph Theory</Title>
<Author>Richard J. Trudeau</Author>
<Date>1993</Date>
<ISBN>0-486-67870-9</ISBN>
<Publisher>Dover</Publisher>
</Book>
Next, I feed into an algorithm the grammar, an instance, and a property. Here is an example of a property: Does the instance have more than one Author?
Here is a graphic which summarizes the situation:
Are there books which describe algorithms for determining if an input (that conforms to a grammar) has a specified property?
/Roger