Google Track

Friday, June 20, 2014

Must read before attending any data science interview

Here are fundamental resources that you should check out before your next data science interview. Read these documents thoroughly to get prepared, impress your interviewer, boost your chances to be hired, and get bigger paycheck if hired:
Resources to read 2-3 days before your job interview:
Related articles

Sunday, March 23, 2014

MH370 flight mystery may have an answer


Data Science perspective: MH370 flight data are crucial

It has been a while and still no answer for the missing airplane of flight MH370. This case may give many possibilities, so guessing by trying each of them is painful and may lead to wrong directions, time consuming and frustrations. So simulating both physical and mathematical model maybe an answer

Mathematical Model

Data Science is newly established profession to solve problems with data insights in high volume of data, so in this case like airplane information systems, weather condition, pattern recognition, path selections etc.
Based in the data above, Data Scientists like me can build specific models to adapt exact or similar scenarios of the missing plane in the line MH370
This may give as good picture what may happen to flight, where may landed, what weather condition we had on that time, what other circumstances occurred and what impact they had over the plane.



Picture: Possible routes

If Malaysian Airport can provide with specific technical information about MH370 airplane, historical data of all MH370 flights, we may find important patterns to solve this mystery. Period between departure and lost signal may give us the distance compared to average distance of all MH370 flights on the same period. Adding weather condition may help us to segments the other MH370 flights that had same weather condition to seek specific scenarios under same weather condition. Route lines are similar to all flights in the same airline, but segmenting them to specific conditions of the missing flight like weather, airplane type, technical conditions, air pressure etc… may lead us to better targets. Technical check before departure may give us information of what have been checked or not and if there is room for possible technical failure. If there is room for failures, we can use data from all technical failure plan crashes to predict the time occurred the failure by that distance also segmented to adapt most the missing flight model.
Big Data technology provides us with the power to analyze big and complicated data sets, and there are plenty of professionals to do so.

Physical model- Simulation

I am not the expert of the field but it can be very smart to build a flight simulation of MH370 based on the mathematical model that we provided here and including other external data that have impact on the fight itself. Sometimes visualization may bring in table other factors that may be decisive in solving the mystery.
These represent alternative approaches to what may help involved institutions to solve this case and I hope they may consider these.

Wednesday, January 29, 2014

Tableau explain why FC Barcelona is still the best team in Spain (and the whole World)


Introduction


Futbol Club Barcelona, also known as Barça is a professional football club, based in Barcelona, Catalonia, Spain. Founded in 1899 by a group of Swiss, English and Catalan footballers led by Joan Gamper, the club has become a symbol of Catalan culture and Catalanism, hence the motto "Més que un club" (More than a club). Unlike many other football clubs, the supporters own and operate Barcelona. It is the world's second-richest football club in terms of revenue, with an annual turnover of $613 million and the third most valuable sports team, worth $2.6 billion.The official Barcelona anthem is the "Cant del Barça" and it is knows to all 480 000 000 fans around the World.

Team Philosophy

Johan Cruyff and Charly Rexach returned to Barcelona in 1988 and began to install a philosophy that would change the way people play and view football forever. Both Cruyff and Rexach admit that it wasn't a completely new philosophy. Admitting that it was one adapted from ideas given to them by Michels, and one which many believe had been given to Europe by the Hungarian side of the early 1950s. The foundation of this philosophy was, and still is, built upon the basic template of touch, technique, maintaining possession, and stretching the pitch with continuous circulation of the ball (Tiqui-Taka). Elements that, at the time, were not valued by many Barcelona supporters (Hunter, 2012)



The chart down (OptaPro) shows the power of Barcelona over other La Liga teams, its play philosophy of short passes and ball possession.


La Liga


Team Discipline and attractivness (support by fans)


Why FC Barcelona is better than any team, this season also?!
Conclusion
With this in mind, it seems that a lot of clubs steal ideas from Barcelona that they see on the surface, such as tactics, and implement them in the short term, but fail to intertwine them into their own specifically moulded model. It is one thing to use Barcelona as inspiration, but it must be remembered that Barcelona's philosophy is tailored to THEIR own needs, no one else's, and the coaching and playing staff have grown together surrounded by it. They have lived by it and through it. Due to Barcelona's on field success, many are using various aspects of the Barcelona model to shape their own coaching methods, training programmes, and playing style etc. Often missing the point that a club philosophy needs to be self defined and fully committed to by all. There are many different styles of playing football, Barcelona have their own unique style of playing, but it should be remembered that this playing style is born out of wider and deeper beliefs in the cultural values of their personal, independent philosophy.

Friday, January 24, 2014

Big Data and Data Science Books - A Baker's Dozen

Here are 13 informative and inspirational books on Big Data and Data Science.  This is definitely not intended to be a comprehensive list (since a complete list of such readings would itself be a form of "Big Data", and consequently the number of possibilities is a nearly uncountable number!NOTE definition of "uncountable" = an infinite set that contains too many elements to be countable.)
  1. Big Data: A Revolution That Will Transform How We Live, Work, and T..., by Viktor Mayer-Schonberger and Kenneth Cukier
  2. The Signal and the Noise: Why So Many Predictions Fail-but Some Don't, by Nate Silver
  3. Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie..., by Eric Siegel 
  4. The Human Face of Big Data, by Rick Smolan and Jennifer Erwitt
  5. The Black Swan: The Impact of the Highly Improbable, by Nassim Nicholas Taleb
  6. Competing on Analytics: The New Science of Winning, by Thomas H. Davenport and Jeanne G. Harris
  7. Super Crunchers: Why Thinking-by-Numbers is the New Way to Be Smart, by Ian Ayres
  8. Big Data Marketing: Engage Your Customers More Effectively and Driv..., by Lisa Arthur
  9. Journeys to Data Mining: Experiences from 15 Renowned Researchers, by Mohamed Medhat Gaber (editor)
  10. The Fourth Paradigm: Data-Intensive Scientific Discovery, by T.Hey, S.Tansley, and K.Tolle (editors)
  11. Seven Databases in Seven Weeks: A Guide to Modern Databases and the..., by Eric Redmond and Jim Wilson
  12. Data Mining And Predictive Analysis: Intelligence Gathering And Cri..., by Colleen McCue
And here are two more, as a bonus:
 14. A Statistical Guide for the Ethically Perplexed, by Lawrence Hubert and Howard Wainer