Turner falls, Devis

It was a beautiful sunny day in Norman of our spring break. I woke up early in the morning to take my friend to the hospital. It was little bit gloomy early in the morning but the sun started shining after ten or eleven. I did not feel like working so decided not to go the school and planning a short trip. Until three I tried my best to convince not to go anywhere because lipa's exam in knocking at the door. Finally I could not and call Adnan and Zafar! and we end up going Turner falls.

It is my second time in Turner falls but this time I was planning for taking lot of pictures! Well, I took a little bit and it was quite nice place, but I guess it is even better for camping not just a short trip. So we end up planning a camping in Turner falls sometimes in coming weeks.

A beautiful sunny day of our spring break!

I was going through National Geographic website today and realized probably I can add a little bit of story with each of my photo post! Why? Well I guess to make it more interesting. I don't think my pictures tell any story, I am not that good in photography yet! At least I am trying to be there. So until then I guess I have to add my story with my photo posts; that might make the entire thing a little more interesting.

Higher dimension!

I was trying to extend my outlier detection technique for higher dimensions. I kept the basic same and try to distinguish outliers in higher dimension. My original technique was based on nearest neighbor and I figure out my approach is not working for higher dimension very well :( I went deep into the program and trying to figure out 'why?' Eventually I realized the average density within certain radius is not very different than the average density of an outlier! In brief the distance is not a discriminative feature of an outlier!

I look into the dataset and it looks okay, then I look into the literature and found something equally interesting and horrifying :( nearest neighbor for higher dimension is not very meaningful; for high dimensional data the ratio of the distance to the nearest neighbor to the distance to the furthest neighbor is 1 and therefore the nearest neighbor based discrimination is not fruitful at all! The situation may improve if we change the distance metric from euclidean to manhattan but not full proof! The good thing about this result is all distance based outlier detection is use less for higher dimension but the bad things is mine is also use less! I have to look for something very different than nearest neighbor.

It is not very intuitive to visualize higher dimensional space and therefore is not easy to imagine the notion of outliers in higher dimension! I guess I need to figure out about what's work on higher dimension. That would not be a easy journey at all.

iOS 4.3

I updated my phone into iOS 4.3! Cool! interestingly there is nothing new for old devices! you can only find the new features (if any!!!) in new devices! Cool! I know Steve Jobs is a fanatic supporter of the idea hardware software decoupling does not work well; wow! even the new updates for same OS do not work for old devices!  Perfect! I wish I had the money to try reverse, I believe that would not work either. Apple does it all time! I wonder why apple does it every time, is it because apple is so stupid that it does not know how to make it decoupled! or what? I find it very irritating! Apple should make dump devices for dump people instead of smart devices. People want more diversity or customization from smart devices. 

New life

What's new in this new life? Definitely first thing is I have to constantly think about a second person. It is quite a difficult task for people like me who are mostly self centered. Most of the time I missed time to spend for myself. I always knew no matter whom I marry she is not gonna understand me; and it is true. Very few people understand me very well and certainly my is not one of them.

The One I Love

This one goes out to the one I love
This one goes out to the one I've left behind
A simple prop to occupy my time
This one goes out to the one I love
Fire. Fire.
This one goes out to the one I love
This one goes out to the one I've left behind
A simple prop to occupy my time
This one goes out to the one I love
Fire. (she's comin' down on her own, now)
Fire. (she's comin' down on her own, now)
This one goes out to the one I love
This one goes out to the one I've left behind
Another prop has occupied my time
This one goes out to the one I love
Fire. (she's comin' down on her own, now)
Fire. (she's comin' down on her own, now)
Fire. (she's comin' down on her own, now)
Fire. (she's comin' down on her own, now)

[R.E.M]

Fast Algorithms for Mining Association Rules - Review

Authors - Rakesh Agrawal, Ramakrishnan Srikant
Summary - The paper presents two association rules mining algorithm for large datasets. Association rule mining is an interesting area of data mining which discovers the relations among the items in transactions. The new algorithms execute at least three times faster than existing algorithms. The first algorithm is called Apriori which incrementally build the large itemsets (an itemset is a set of items, the support of an itemset is calculated by the count the set appear in the database and a large itemset has support greater than minimum support) until no new bigger large itemset can be added, but this algorithm requires many passes over the database to discover all large itemsets. To tackle this problem this paper also presents another algorithm called AprioriTid, which does not require many passes over the database. AprioriTid stores the list of transactions that requires passing again to compute bigger large itemsets in the subsequent steps. However, the list of transactions that requires passing again may grow larger than the original size of the database which deteriorates the performance instead of improving it. This problem is severe for greater number of large itemsets but drastically disappear for smaller number of large itemset. Hence this paper proposes the use of a hybrid algorithm composed of Apriori and AprioriTid called AprioriHybrid. AprioriHybrid uses Apriori at the beginning when the number of large itemsets is high and switched to AprioriTid when the number of large itemsets became small. An empirical study confirms their claim at the end of the paper.
Further thoughts - (1) In order to obtain meaningful rules how does a user choose minimum support and minimum confidence? (2) Attribute data are fundamentally different from transactional data, how does the association rule mining work for attribute data?

Lyrics

Yesterday, I was planning start writing a short review of every paper I would read. Well I am hoping I would have a repository of review of papers at the end of the day. Keeping that in mind I came to my blog this morning I found I have posted almost seventy distinct lyrics! Wow, I know all those seventy songs were my favorite and I have a repository of my favorite songs! Is not it cool; I was pretty excited when I realized I have seventy lyrics. Every lyrics remind me a time, a time of struggle, a time of love or may be something else. This is what my blog is all about; it is keeping the past to share with the future. I know I would love my blog even more in coming days, I can feel I am becoming more lonely. :(

Anyway, I am gonna start writing short review about the papers I would read from now on; probably I should start putting all the reviews I have already and the new reviews would be added as  I read new papers. But my question is how I am gonna put the reviews? If it requires too much effort I am gonna stop that after while; so it has to be incredibly easy while effective. lets start it if it doesn't work then I'll stop it like I did for many things in this blog!

iPad

I was not very convinced when apple released iPad last year. I was not quite sure why would I use it? They released the second version this year and I am still not quite convinced about why would I use it? It looks like some people are gonna use apple product no matter what it is! Apple has a tremendously loyal customer base and it makes product for them!

Interesting enough, many people even try to copy iPad and now the market is full of this kind of tablet, funny huh! I would not be wondered if I buy one tomorrow and start thinking this is cool gadget! There is no doubt it is a cool gadget, but my point is what's for?

I did a little research today, I am amazed that Steve jobs is not the one who invented tablet pc, it is again Microsoft, Microsoft introduced tablet pc back in 2000, at that time it was not popular. It was more like other Microsoft product, a set of buttons and crappy look and so; the true difference from today's iPad is it was not neat! People did not like it at that time, the sad part about Microsoft they have 90% market share for operating system but zero fan where as apple got near 5% market share and even more fan! They are just fanatic, they don't care what apple is making, they know got to have one! I know apple products are beautiful for they are not the best! I personally find no reason to an apple product! Well I bought an iPhone but I would love to trade it for an android phone. The reason I bought an iPhone is it was the cheapest deal at that time, and I am still not in love with it. Certainly I am in love with smart phone but as long as other smart phone provide the same facility I am happy.

I am thinking it would be even more fun to use a android phone. I could make my own application. So there is no seamless interface for app building for iPhone apps in windows! Pity!

BlogSpace

Looks like I got an app for iPhone blogging. It passed my initial test, it has portrait and landscape mode and it can do tagging! Two most important feature I expect from blogger application.

I think it support picture and video for blogspot. That's cool, I think this is the best free blogging app for blogspot. I hope it is usable, in my experience most iPhone apps are not usable; they crash every so often. One more thing don't go back to see local draft while bloging, because it does not save every word you type.

I would say this is the best free blogging app, I don't know about the paid apps because I didn't use any of them. The only irritating part is ads; but that's okay, they have to earn money anyway.

The initial screen freezes very often and you have to tap more than once to something and the behavior is kind of random. At least it can save after putting in sleep.

Meeting today

I have a meeting with Dr. Gruenwald today, huh one more meeting, lots of promises, no results and so and so. I really don't like to meet with her. I am wondering what we would discuss? I tried to run the experiments and it is not completed yet. We are planning to submit a journal version paper. This paper would include our new experiments for multi dimensional data. I have been doing the same things for two semester now! That's pretty bad. I believe I can do better than what I am doing right now but somehow I am becoming demotivated day-by-day.

Okay, finally the meeting is over. I did not promise anything today but she asked me to do something. She is too good to me, I didn't present much today and still she did not tell me anything. I am wondering why? Either she starts trusting me or she has something in her mind. I personally prefer to go with the first one. Yes, she trusts me and that made me stay here after my MS. I know that was a wrong decision and I don't know how to fix it! I don't want a bitter end, I want a smooth transition but I now know I am scared of uncertainty and I am not going to stand it forever.