Introduction
AIML Parser module lays down the foundation for introducing AIML in Drupal maintenance support plans and creating Chatbots. As you may know, Chatbot is a program that chats with users automatically. In this blog I will introduce and highlight basic concepts that have been used for building this module.
Things you need to know before you start using this module:
1) What is AIML?
AIML (Artificial Intelligence Markup Language) is an XML-compliant language that is easy to learn, making it possible for you to customize a Chabot (Drubot) or create one from scratch within minutes.
AIML pattern syntax is very lucid, significantly less complex than regular expressions. To compensate for the simple pattern matching abilities, AIML interpreters could provide pre-processing functions to expand abbreviations, remove misspellings, etc.
2) What is a Chatbot?
A chat robot (Chatbot) is a conversational agent. A computer program designed to simulate an intelligent conversation with one or more users via auditory or textual medium. This Chatbot can pose as a knowledge agent in place of an FAQ or can be used as a call-center agent.
3) What is an AIML file?
AIML file is similar to an xml file with its own tags. AIML files are open source, and can be found over the web for free. We have uploaded a few of them here, (Download Sample AIML files).
Let’s take a look at an example.
WHAT ARE THE LAWS OF THERMODYNAMICS?
I’m not a physicist, but I think this has something to do with heat, entropy, and conservation of energy, right?
In the above example the pattern is: “WHAT ARE THE LAWS OF THERMODYNAMICS?”
And the template is: “I’m not a physicist, but I think this has something to do with heat, entropy, and conservation of energy, right?”
If you upload this AIML file through an AIML parser module, module will parse the file and store respective tags in database. Later you can access them through a query and your Chatbot answers the question:
“WHAT ARE THE LAWS OF THERMODYNAMICS”
Answer: “I’m not a physicist, but I think this has something to do with heat, entropy, and conservation of energy, right?
Here are some Basic Tags which are present in an AIML file and can be parsed by this module.
AIML – defines the beginning and end of a AIML document.
Category – defines the unit of knowledge in knowledge base.
Pattern – defines the pattern to match what a user may input to an Drubot.
Template – defines the response of an Drubot to user’s input
Who creates AIML tags?
These tags are open source, with a number of communities and companies contributing to enhance them.
Now that we know all about Chatbot and AIML, you must be wondering how Chatbot works?
Chatbot needs AIML files as knowledge source, a database from where Chatbot can retrieve responses for the asked pattern.
Use Case of this module:
After the AIML tags are parsed and stored in the database, you are ready to create a Chatbot,
We are working on a Chatbot called Drubot.
Steps to create a Chatbot using AIML parser:
Step1: Create a custom block.
Step2: Add a robot image to the block.
Step3: Create a form in the block with a text-field.
Step4: This text field will accept input from user.
This accepted input is the pattern, which when queried in Drupal maintenance support plans database, gives you the response. Referring to the example stated above, if user types: ‘WHAT ARE THE LAWS OF THERMODYNAMICS?’ into the text field.
It displays ‘I’m not a physicist, but I think this has something to do with heat, entropy, and conservation of energy, right?’ as the response.
How does AIML parser module empower the Chatbot to learn?
There are open source AIML files available that serve this purpose. For example- science.aiml, sports.aiml, love.aiml.
AIML files have questions and answers related to each category. If you want to educate your Chatbot about science, then upload science.aiml and Chatbot can then use the responses stored in the database to answer science related questions.
We are currently working on a Chatbox in Drupal maintenance support plans download link which uses AIML parser as a dependent module.
Now use all this information to create your own chatbot. Feel free to connect with us for any assistance.
Source: New feed
