Arvind Ravichandran
I.  Artificial Learning through Pattern Recognition 

II. The goal of this project is to simulate the process of learning by pattern recognition.

III.    The purpose of this project is to use a computer language to simulate 
learning thorugh pattern recognition. If succesful, the project will provide a 
unifying force to Artifical Intelligence - thus greatly expanding the capabilites of this field. 

IV. Scope of Study. The success of the project will be judged by its ability to
learn language - specifically english - through the process of pattern recognition.
The largest challenge will be creating a universal system of pattern-recognition.
The program will not be designed only to learn a language; thus the language-recognition
will only be a reflection of its ability to translate data from any source into readable form. 

V.  Background and review of current literature/research in this area. 
        The overall field of AI is relatively new and so is pattern recognition.
Some limited and restricted experiments in the field have been undertaken: for 
example the rocks-paper-scissors model here: http://www.newscientist.com/article.ns?id=dn6914. 
Other adventures into machine-learning have been made, especially in neural networks. 
However, on the large scale, they have come up limited and very specific in their 
ability to learn, similar to most other aspects of AI.

VI.   Procedure and Methodology.
  The approach used will be completely console based. This is to limit the number 
of inputs to a manageable level, although the goal is to keep the program abstract 
enough to include more advanced inputs later. The primary programming language 
will be java, due to its ease of use and object-oriented heirarchal approach, 
which in many ways is similar to the nature of the program. The program will be 
sub-divided into three parts. The "brain" will be the center for reasoning along
with the cross roads for data from the other two parts - its primary job is to 
process data. The memory will be divided into two groups, short and long; all 
the data will be accessed and stored by the brain. Finally, will be the i/o core
- for this project there will only be one (the console) however, the approach 
will be abstract enough so that anyone may add additional inputs/outputs based 
on their needs - it is here where the enormous power and flexibility of the system 
will arise. While these are the main systems, additional systems may be added 
as necessary - possible systems include a message system (if it needs to be 
independent) along with splitting the input/output core.
                            
VII.    Expected Results & Value to Others
     The program, when succesful, should demonstrate a command of the english 
knowledge equivalent to that of an intelligent five-year old. While not having 
an enormous vocabulary or amazing articulation skills, the program should be able
to compile mostly correct (grammatically) sentences that expresses some basic ideas.