This is an automated email from the ASF dual-hosted git repository. aradzinski pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-nlpcraft-website.git
commit cc9b1a2f99b4a3485595fddd55a0fce96c53a603 Author: Aaron Radzinski <[email protected]> AuthorDate: Thu Jun 3 17:57:00 2021 -0700 Update nlpcraft-idl-intent-definition-language.html --- blogs/nlpcraft-idl-intent-definition-language.html | 98 ++++++++++++++++++++-- 1 file changed, 89 insertions(+), 9 deletions(-) diff --git a/blogs/nlpcraft-idl-intent-definition-language.html b/blogs/nlpcraft-idl-intent-definition-language.html index 1373f31..4b1a34d 100644 --- a/blogs/nlpcraft-idl-intent-definition-language.html +++ b/blogs/nlpcraft-idl-intent-definition-language.html @@ -40,34 +40,114 @@ publish_date: June 3, 2021 working with intents in NLP-based dialog and search systems developed using Apache NLPsCraft and at the same time expands the capabilities of them. </p> - <p> - Let's start with examples to demonstrate the general capabilities of the language, provide the necessary explanations, - and then describe the design of the language a bit more formally. - </p> - <div class="bq info"> - Note that at the point of choosing the best matching intent, NLP systems, in general, already have a parsed and processed - user input request, which contains combinations of request’s tokens and other related information. + <div class="bq success"> + <p> + Note that at the point of choosing the best matching intent, NLP systems, in general, already have a parsed and processed + user input request, which contains combinations of request’s tokens and other related information. + </p> </div> </section> <section> <h2 class="section-title">Examples <a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2> + <p> + Let's start with examples to demonstrate the general capabilities of the language, provide the necessary explanations, + and then describe the design of the language a bit more formally. + </p> <pre class="brush: idl"> intent=xa flow="^(?:login)(^:logout)*$" meta={'enabled': true} term(a)={!(tok_id()) != "z"}[1,3] term(b)={ - meta_intent('enabled') == true && - month() == 1 + meta_intent('enabled') == true && // Must be active. + month() == 1 // January. } term(c)~{ + // Variables. @tokId = tok_id() @usrTypes = meta_model('user_types') + // Predicate. (tokId == 'order' || tokId == 'order_cancel') && has_all(@usrTypes, list(1, 2, 3) && abs(meta_tok('order:size')) > 10) </pre> + <p><b>NOTES:</b></p> + <ul> + <li> + The intent name is <code>xa</code>.</li> + <li> + The intent contains three terms. Term is an element that defines a predicate, each of which must + pass for the intent to be selected: + <ul> + <li> + <code>term(a)</code> - the parsed request must contain between one and three tokens with identifiers + other than <code>z</code>, without taking into account data from the dialogue history (term type <code>=</code>). + </li> + <li> + <code>term(b)</code> - the intent must be active - the flag <code>enabled</code> from the model metadata. + In addition, such an intent can only be triggered in January - the <code>month()</code> built-in function. + </li> + <li> + <code>term(c)</code> - the token with identifier <code>order</code> or <code>order_cancel</code> should + be found in the request or in the dialogue history (term type <code>~</code>). There are additional + restrictions in the model metadata values and also the order size absolute value should be more than 10. + In the definition of this term we use IDL variables - and we will talk about them a little bit later. + </li> + </ul> + </li> + <li> + <code>flow=</code> This section defines an additional rule, according to which for the intent to + be triggered, the intent with the <code>login</code> identifier must have been selected at least + once within the current session, and the intent with the identifier <code>logout</code> should have never + been selected in this session. The rule is defined as a regular expression based on the intents IDs + of the previous intents in the user session. Other ways to create similar rules will also be + described below. + </li> + <li> + <code>meta=</code> For the given intent, a certain set of data can be defined - in this case a + configuration - with which you can enable or disable the intent. + </li> + </ul> + <br/> + <pre class="brush: idl"> + intent=xb + flow=/#flowModelMethod/ + ordered=true + term(a)=/org.mypackage.MyClass#termMethod/? + fragment(frag) + </pre> + <p><b>NOTES:</b></p> + <ul> + <li> + The intent name is <code>xb</code>. + </li> + <li> + <code>term(a)</code> The intent can contain one optional term (<code>?</code> quantifier, + detailed explanations will be given below), defined in the code - <code>org.mypackage.MyClass#termMethod</code> method.</li> + <li> + <code>fragment(frag)</code> Fragment with identifier <code>frag</code> extends the list of terms of the intent with + additional terms that are defined elsewhere (in <code>fragment</code> expression). The <code>frag</code> element must + be defined above in the code, or accessible via import statement. + </li> + <li> + <code>flow=</code> Flow contains the condition specified in the model code in the method <code>flowModelMethod</code>. + </li> + </ul> + <p> + You can find more examples <a href="https://nlpcraft.apache.org/intent-matching.html";>here</a> + </p> +</section> +<section> + <h2 class="section-title">Deeper Dive<a href="#"><i class="top-link fas fa-fw fa-angle-double-up"></i></a></h2> + <div class="bq info"> + <p> + Note that this article is only a brief overview of the capabilities of NLPCraft IDL and is not trying to + be a comprehensive manual. Before starting to work with NLPCraft, it is recommended to study a + detailed description of the syntax and all the features review of the language in the corresponding sections + of the <a href="https://nlpcraft.apache.org/intent-matching.html";>documentation</a>. + </p> + </div> </section>
