Week of 10.22

Week of 10/22

     This week, we created a survey to investigate correlations between the student population and their personal preferences. We tried to ask some simple yes/no questions to construct a multidimensional plot to find correlations. Most of them were goofy, some of them were kind of legitimately serious. For the sake of data collection, we kept all of our answers in yes/no format, as it could ensure the simplicity of our data as we converted it to -1/1 values. Pushing it out to a large group of people was kind of hard though, since we all had to leave our own bubble of juniors and seniors and also ask the freshmen and sophomores to take the survey as well.

A screenshot of some of the questions in our survey.

Thankfully, Professor Hassibi okay-ed our questions and said that they were pretty good, and whole he made a few suggestions here and there to add to our question list, I think we were doing pretty well. Actually, on the topic of suggestions, Professor Hassibi recommended that we do another survey to figure out the friend groups of people at our school. This was the start to a pretty interesting topic. In terms of the survey, I'm not sure what we really could have done to ensure anonymity, but we're still trying to figure out if we can find other softwares where other people could enter their data in but we on the receiving end wouldn't receive the personal data (their names, their friends names, etc. etc.). I think that it's a bit of a stretch to believe that the anonymity can actually be ensured, but my mind is open to the different possibilities and strategies we can employ to make this happen.

Here's an explanation of the friendship paradox:

The reason why Professor Hassibi (or Dr. Hassibi I guess) wanted us to make this survey was to investigate the friendship paradox. Basically the friendship paradox is a mathematical equation that determines, on average, how many more friends your friends have on average than you do. I think it's kind of sad to find out that my friends will on average have more friends than me, but after some consideration I think it makes sense. The math works out. Obviously, I kind of went back to it for my concept map, but I think that even though I struggled a lot with the understanding of the equation and how the paradox works mathematically, I think that this will lead us to a better direction in terms of working on the karate dojo too. 

Dr. Hassibi and his work of art

We tried to figure out how the equation worked. Aston and I worked in Mr. Lee's room and in Ms. Hiranaka's room and played with the equations a bit to figure out the nuances of it. I think this was pretty interesting- a lot of ideas were thrown around, but at the end of the day they were mostly resolved. These concepts are becoming increasingly fascinating, and I think the process of deriving each equation is pretty fun. I'm looking forward to more knowledge! :)

Here is my hierarchal clustering video. I filmed it on my iPad, and I realized that it was kind of tough for me to film while talking. Oh well, enjoy: