Arvind Ravichandran
Artificial Learning through Pattern Recognition
Period 5
Java - Here's the project's API
Oct. 5: Prototype 1, Class Brain receiving messages. Making an API
Oct. 7 -
Output October 7, 2005:
java.lang.ThreadGroup[name=Peripheral System,maxpri=10]
hello
Console_47
Brain_100
100100000000000011041101110811081111
Description:
The majority of this output is debug code that signifies that the
program is working. The first line shows that the peripheral system (the
message handler and inputs from the console) is in one threadgroup for
easy shutdown and restart.
The next line is user input; a simple hello - we will trace this message
through the program.
The next line shows how the input hello has been package into a type message
class with the id Console_RandomNumber. The first word (before the underscore)
tells the message handler what processes the message. From the console,
the string is packaged and sent to the message handler, from which it is
processed by the Console Message Processor (all message processors derive
from a single message processor class). The resultant data is packaged again
into a message and sent back to the handler from there it goes to the brain
(also deriving from message_processor for ease of use) and the final
message shows the transformed string (from hello) - the series of four
digit numbers represents the string, whereas the numbers in the beginning
indicate statistics about the message (the majority of which are unimplement
currently).
The basic summary to be taken from this success is that the prototype and
shell actually works. None of the code is hard-coded, thus the layer of
abstraction is successful and there is plenty of room to grow and improve
the project, along with room to plug in the actual core of the project
the pattern recognition.
Oct. 14 - Working on 2 things for Brain control. When in a Resting state, when
temporary memory is full, stores memory into general memory.
Thread group, right now there's one thread "Peripheral System" - see output
above. Right now the thread is basic, just is able to shut down the system.
Oct. 19 - Working on the presentation of the API
Dec. 14 - Status Update Report
I refocused the project on a more achievable objective (learning tic-tac-toe).
However, since the first quarter was just constructing the core methods,
no time was lost (good planning eh?). I recently finished a tic-tac-toe
game that will be used as the test. Right now I am adding the tic-tac-toe
processor and the brain algorithim. The tic-tac-toe processor should be
done by today, so I will be beginning the brain algorithim by the time
you read this.