Day XXVI:

Today I started planning out my presentation. I not sure what the scope of it should be so I probably will have to cut some stuff out or change my focus. As of right now my topic is "The Application of Neural Networks to the prediction of Coronal Mass Ejections". I don't think that I will be able to both explain how Neural Networks work, and how their attributes can help in the prediction of the onset of a Coronal Mass Ejection. I may have to only talk about Neural Networks or the prediction of CMEs.

I also played some more Call of Duty today for the Visual Perception lab's experiment. The eye tracker apparently was not working but they let me play anyway. It is a fun game.

Another CME has been projected to hit the earth tonight. There have been a series of events beginning earlier this month that have all made an impact here on Earth. The most surprising part about this is that we are not necessarily being hit by the biggest bursts. For instance, the sun may have one large explosion followed by a much smaller one. As common knowledge would have it, the larger CME would be the one hitting the Earth but I guess that is not the case. Another part of this project now is to determine not only when a CME is occurring but also if a CME will actually make contact with the Earth. I still don't understand how we are supposed to determine the trajectory of a CME just from the two dimensional images provided by the LASCO C2 sensor.

Day XXV:

I was very capricious in my work today. From 9 to 12 I did some more work looking at X-Rays spikes and their relation to CME events. After lunch, Piyush took me, Cicely, and Kevin over to the Computer Engineering and Computer Science departments just to take a look around and explain how things happen at RIT. It was pretty interesting hearing all of the various fields of Computer Science and Computer Engineering.

Afterward I went to the Visual Perception to partake in their experiment. I basically got to play Call Of Duty for an hour while they tracked my eye movements. It was fun.

After that I saw Bob and Dan in the hall moving tables and equipment. I decided to help. Apparently they are converting a computer lab into a workshop in order to make room for more graduate students. I didn't exactly understand why they didn't just turn the computer lab into offices and leave the original workshop (the one that we were moving all the equipment from) alone. We had to transport a lot of heavy equipment from the first floor to the second. This required a lot of brute force. I'm probably going to help them again tomorrow.

Day XXIII:

I was at the computer almost all day today, but I got through a lot of data. From what I have found there seems to be a pretty decent correlation between X-Ray spikes and CME events. While the idea is not perfect and still needs some more support, the X-ray flux spike and CME almost happen simultaneously, with the X-ray spike happening just before the actual event. So far there are many exceptions to the rule, especially during times of increased CME activity, but it could possibly be a good forecast that a CME is occurring. I took a brief look at other types of data and there are similar trends but they need to be explored further.

I also continued working on the GUI for my model Neural Network today. I decided to break it up into different windows as opposed to everything being in the same window, creating clutter. I hope I can actually get it to work.

Day XXII:

Today I spent more time working on trying to find correlations in the recent CME data and past events. It is very up and down. At one time there might appear to be a trend in the data, such as a spike or a pattern, but it will not match up with data from other times. Caitlin, Becky, and I met with Dr. Dube to discuss our findings. Some of the data we need is very specific and hard to find because certain sensors or satellites did not exist at the time of older events. Nonetheless I found a database that contains a lot of X-Ray and other sun data.

When looking at the data, it is just as important (in theory) to find the things that don't match up with a CME event because that way erroneous data need not be entered into the Neural Network, making the Network run faster.

I listened to another Ph.D dissertation today and it was just as over my head as the first one I listened to. There were a lot of references to certain software and techniques that I had never heard of before, so my understanding of the topic beyond the general goals was very limited.

Day XXI:

Today Becky, Caitlin, and I went to be stand-ins for a photo shoot for someone. We were there for two hours and now my face hurts. Afterward we went to Dr. Dube's office where he told us about a CME that occured on august 1 and has been predicted to hit the earth on August 4th. We now are trying to take a look at different possible indicators of the CME such as X-ray flux or sunspot density. It is hard to determine if there is a correlation. For instance, the X-ray flux before the CME happens to have a large spike in intensity. But when looking at former events the X-ray flux does not always point to a CME event.

On a side note, the fire alarms went off today in the building.

Starting Day + 19:

Today I did some more coding and debugging of my neural network and started designing a user interface for it, mostly for my own benefit so I can better see what errors are going on in my program. It will be fun and interesting to create something that is not pure text.

I also went to the Scube today (I'm not sure how it is spelled). It was not in working order but it was cool nonetheless. The idea is that a person can become completely immersed in a computer generated environment. Four projectors cast images onto each of the four vertical sides of the cube, creating interactions that are three dimensional. There were some technical difficulties but at least I got to understand the ultimate goal.

Day XIX:

Today we had another Space Weather group meeting. Dr. Dube explained his work on trying to see if a neural network can make a correlation between spikes in separate data streams happening at different times. He explained the x-correlation technique and how every point in one set is multiplied by every point in the other set and is then normalized. This produces a huge spike in the output data. Dr. Glenn showed his work on image processing and how each quadrant of an image is in a particular state is is compared to a baseline or "quiet state". When a CME occurs, each quadrant show a spike in the different between the quiet state and the actual state. The problem is to see if a neural network can take in the data from the image processing technique and find a pattern. This would ultimately turn everything on its head and result in the prediction of, as opposed to reaction to, Coronal Mass Ejections.