Michael Klosiewski's CSCI460 project weblog

Today I completed the first version of the part of my code that removes certain suffixes from words, counting the root word as it would any other. I ran a couple of tests on it, and thus far it does not work very well. Luckily, I have it creating a file of all of the changes to words that it is making so that I can do analysis to make it better.

After further testing and improving this part of the program, I will either start trying to implement a solution that generates multiple reports and combines them. I should have another report of my activity in the next couple of days.

Today I worked on converting my program to work off of small reports. After a little while of debugging, I've gotten it to create the smaller reports, which seem accurate by inspection, but the final result isn't quite making sense yet, and I have to look more at how I'm preparing it.

On the other hand, and have almost given up on the suffix-remover/word-combiner. The English language is not lending itself to following patterns that I can use, and unless I can figure out a way to make it work, I don't think it's really worth it. This is especially true when I consider that I may not even want to reduce words to their root words in all cases anyway. As an example, in the phrase "committed relationship", the word "committed" is being used more like an adjective than a past tense verb. Do I really want to reduce it to "commit"?

All in all, I think things are still progressing nicely. If I can get this part worked out tomorrow or the next day I should have enough time to run some runtime stats on the program and document for the walk-through the week after next. At that point it should be producing results of the quality I desire in the final product, though perhaps not very efficiently.

One thing that I overlooked before going on break was the fact that I had actually not done anything in terms of recording sentence data. I have worked on it a little bit over break, but that part will probably take the next week to complete.

One thing that I did get further on is planning how the different parts of the program are going to interact, and I have prepared a couple of documents detailing this. However, in order to do it the way I would like to, ideally I should be able to make each part of the project a dynamically linked library, and will need to do some studying on how to set up projects that way.

I should also begin my work with the flash environment as soon as I can gather sentence data, as I believe we now have a couple of work stations equipped with it. This should all happen within the week to get back on schedule, as I am now slightly behind.

This past week I have not been able to get much done, simply for lack of time. I have been able to work some on the Sentence Gatherer, but it is not yet completed. For now, that is what I must do.