Forgot your password?
typodupeerror

Comment: Re:Lots of patterns (Score 1) 466

by crunchyeyeball (#35659384) Attached to: FBI Wants You To Solve Encrypted Notes From Murder

Also, virtually every "word" ends with an "E" - this suggests that perhaps the writer had a number of coding schemes which could be carried out in his head on a word-by-word basis, and he's mainly using coding scheme "E" in this case.

The smallest word I can see is "SE" which appears at least twice - could this be "a" or "I" perhaps?

Comment: Re:33 = 3*11, 11*3 (Score 1) 486

by crunchyeyeball (#33787820) Attached to: The Binary Code In Canada's Gov-Gen Coat of Arms

You are, of course, correct. I meant to say that 33 has only two prime factors. This was meant to be in reference to Carl Sagan's Arecibo Message*, which used a similar encoding scheme, but with 1679 bits forming a 2d image if plotted as rows & columns based on its prime factors (23*73).

* http://en.wikipedia.org/wiki/Arecibo_message

Comment: Re:Immature and Gun Happy (Score 2, Informative) 1141

by crunchyeyeball (#33648326) Attached to: Hunters Shot Down Google Fiber

Could you clarify where you got that statistic from? According to my research, the relative murder rates* are:

US: 0.042802 per 1,000 people
UK: 0.014063 per 1,000 people

i.e. you are more than 3 times as likely to be murdered in the US.

*Source http://www.nationmaster.com/graph/cri_mur_percap-crime-murders-per-capita

Comment: Re:The Objective (Score 5, Informative) 104

by crunchyeyeball (#29501597) Attached to: BellKor Wins Netflix $1 Million By 20 Minutes

Basically, you were asked to predict how a number of users would rate a number of movies, based on their previous ratings of other movies.

You were supplied with 100 million previous ratings (UserID, MovieID, Rating, DateOfRating), with the rating being a number beween 1 and 5 (5=best), and asked to make predictions for a seperate ("hidden") set comprising roughly 10% of the original data. You could then post a set of predictions to their website which would be automatically scored, and you'd receive a RMSE (Root Mean Squared Error) by email.

To avoid the possibility of tuning your predictions based on the RMSE, you could only post one submission per day, and the final competition-winning results would be scored against a seperate hidden set, independent of the daily scoring set.

It really was a fantastic competition, and anyone with a little coding knowledge (or SQL knowledge) could have a decent go at it. Personally, I scored an RMSE of 0.8969, or a 5.73% improvement over Netflix's benchmark Cinematch algorithm, having learnt a huge amount based on the published papers and forum postings of others in the contest, and my own incoherent theories.

In a way, everyone wins. Netflix gets a truly world-class prediction system based on the work of tens of thousands of researchers around the world hammering away for years at a time. Machine learning research moves a big step forward. BellKor et al get a big juicy cheque, and enthusiastic amateurs like myself get access to a huge amount of real-world research and data.

We all like praise, but a hike in our pay is the best kind of ways.

Working...