AboutSaurabh Kudesia Expertise I can answer your queries related to AI, Fuzzy logic, algorithms, VR and simulation theories.
Experience With an electronics background, I was rated as one of the Top 5 expert of Yahoo!Expert Group for three years in the category AI/Robotics. I have had the honour of evaluating books of international repute on AI/Robotics as Subject Matter Expert. I have in my credit more than 50 papers/articles published in different National and International Magazines/websites covering various aspects of AI, Robotics, VR and Mind theories. I am also contributing my skills and expertise to help movements like Singularity, Transhumanity and Cryonics.
I am looking for some smart AI software components that would read XML, and then create categories, intelligently and accurately from data in the XML.
For example:
Basketball would be under the category 'Sport'.
The categories are in a database and ideally, new categories could evolve from the reading of the XML.
Am i talking about some natural language processing combined with AI to achieve this? Do you know of any off the shelf software that could help me here?
Any help kindly appreciated.
Thanks,
Codi
Answer Hi Codi,
Thanks for your mail. If I am correct, you need to develop a class of data objects that can partially describes the behavior of program that process them. In this case, you can very well use the AIML (Artificial Intelligence Markup Language) to achieve your goal. See http://www.alicebot.org/TR/2001/WD-aiml/
AIML is based on XML, describes the class of data objects as AIML objects and partially describes the behaviour of computer programs that process them. In your case, the behavior of data should be known before the program categorize them. For instance, Basketball can be placed under both "Sport","Material", or "Sport-Equipment" category. Before creating the new categories, the program should be capable enough to evaluate the likelihood of the data falling under various categories and pick the best category. AIML objects are made up of units called topics and categories, which contain either parsed or unparsed data. The most important part is that the AIML elements encapsulate the stimulus-response knowledge contained in the document. Character data within these elements is sometimes parsed by an AIML interpreter, and sometimes left unparsed for later processing by a Responder.
I haven't tried AIML myself and so I am not sure how it can be applied to your case but I believe it is closest to what you are looking for.
