Screenshot of the preliminary prototype visualizer.
This is the first blog post about a project that has been on my mind for a while. Essentially, I have always looked for technically challenging areas in Comp.Sci, so for things like low-level, hard-core coding, as well as something that is more artistic, something that is graphical, interactive, intuitive to everyone the moment they see it. In other words, I looked for a project that is very tricky inside, but looks nice and simple on the outside.
I think I might have found an idea that just might satisfy these two requirements.
While taking a Data Mining course at the University of Manitoba, I have found it peculiar and weird as how people deal with the results of data mining algorithms. No doubt, throughout that course, I learned how the data can be mined and what one can discover, but, at least to me, the missing link was the ability for everyone to see that, in fact, after performing all of these calculations, we can observe X or Y or Z, etc. By everyone I mean everyone - even my mom, as she has no idea in computers, should be able to understand what is going on. With these thoughts in mind, I thought I would give it a try as a short course project.
Goal of the project is to visualize frequent patterns in 3D.
Eventually, I realized that this idea might have more potential. Not only the visualization itself, but the challenges it contains 'under the hood' - simply processing 4+ million records is an interesting challenge.
Here is a short list of what has been achieved so far:
- A prototype system has been developed. It allowed to identify potential problems, explore tricky areas and clarify the direction of the projects, its scope and the amount of work required to build a truly good visualization.
- Further project directions have been given some thought. A list of potential solutions to the existing problems in the prototype has been proposed, although, it is still quite possible the entire project will be rebuild in Unity instead of a conventional OpenGL framework. That said, however, there is nothing wrong with the framework that has been used.
- Since this research will be a part of a course project at the University of Manitoba, a project proposal has been worked on, some minor literature review has been done on the topic.
Next steps:
- I need to narrow down the list of things required to be done at the initial stages. It is always hard to find focus of the project at the initial stages, however, that is always a critical part.
- Decide on the platform and/or framework. That is also tricky, but much easier than the first point.
- Get all the formal stuff out of the way. Major step here was the proposal, with that done, it is time to narrow the focus and just go.
- Just enjoy it. This step is easy - can't wait to get down to coding & seeing this thing come to life.