How to become a superhero

from Amazing Fantasy by Stan Lee and Steve Ditko

We analyze a collaboration network based on the Marvel Universe comic books. First, we consider the system as a binary network, where two characters are connected if they appear in the same publication. The analysis of degree correlations reveals that, in contrast to most real social networks, the Marvel Universe presents a disassortative mixing on the degree. Then, we use a weight measure to study the system as a weighted network. This allows us to find and characterize well defined communities. Through the analysis of the community structure and the clustering as a function of the degree we show that the network presents a hierarchical structure. Finally, we comment on possible mechanisms responsible for the particular motifs observed.

Gleiser, P. (2007). How to become a superhero Journal of Statistical Mechanics: Theory and Experiment, 2007 (09) DOI: 10.1088/1742-5468/2007/09/P09020 (arXiv)

Alan Turing's declassified papers

@ulaulaman via @MathisintheAir about #AlanTuring on #arXiv

Recently Ian Taylor has uploaded on arXiv a couple of declassified papers by Alan Turing about statistics, probability and cryptography:

---

1. The Statistics of Repetitions

In order to be able to obtain reliable estimates of the value of given repeats we need to have information about repetition in plain language. Suppose for example that we have placed two messages together and that we find repetitions consisting of a tetragramme, two bigrammes, and fifteen single letters, and that the total overlap was 105, i.e. that the maximum possible number of repetitions which could be obtained by altering letters of the messages is 105; suppose also that the lengths of the messages are 200 and 250; in such a case what is the probability of the fit being right, no other information about the day's traffic being taken into consideration, but information about the character of the enciphered text being available in considerable quantity?

2. The Applications of Probability to Cryptography

The theory of probability may be used in cryptography with most effect when the type of cipher used is already fully understood, and it only remains to find the actual keys. It is of rather less value when one is trying to diagnose the type of cipher, but if definite rival theories about the type of cipher are suggested it may be used to decide between them.

The Championships' Final

by @ulaulaman via @Airi_Talk about #WorldCup2014 #Brazil2014 predictions: #ESP-#GER

Jürgen Gerhards, Michael Mutz and Gert Wagner developed an economic model in order to predict the results of the football team during the international cups. The researchers evaluated the market value of every player and described every team with an economic price: in this way they predict the winners in 2006 (World Cup: Italy), 2008 (Euro Cup: Spain), 2010 (WC: Spain), 2012 (EC: Spain). Starting from the previous results, the three researchers realized the board of the challenges from the eighth:

The predicted final will be between Spain and Germany, with Spain favorite, but there is a little hope for the other teams, first of all Brazil: in 2012 the model didn't predict the other team in the final game of the Euro Cup: Italy, that should not have reached even into the semi-finals:

Marketization and globalization have changed professional soccer and the composition of soccer teams fundamentally. Against the background of these shifting conditions this paper investigates the extent to which the success of soccer teams in their national leagues is determined by (a) the monetary value of the team expressed in its market value, (b) inequality within the team, (c) the cultural diversity of the team, and (d) the degree of turnover among team members. The empirical analyses refer to the soccer season 2012/13 and include the twelve most important European soccer leagues. The findings demonstrate that success in a national soccer championship is highly predictable; nearly all of our hypotheses are confirmed. The market value of the team is, in today's world, by far the most important single predictor of athletic success in professional soccer.

Jürgen Gerhards, Michael Mutz, Gert Wagner (2014). Die Berechnung des Siegers: Marktwert, Ungleichheit, Diversität und Routine als Einflussfaktoren auf die Leistung professioneller Fußballteams. Zeitschrift für Soziologie, Jg. 43, Heft 3, 231-250

---

via University of Berlin, galileonet.it, airicerca.org

Persi Diaconis explains randomness

#PerciDiaconis #MartinGardner #statistics #mathematics #video

Perci Diaconis is a mathematician and a student of Martin Gardner. He was born on the 31st January, 1945, so, with a bit of delay, this is a way to say (after Rudi Mathematici): Happy birthday!

Stanford Statistics and Mathematics professor Persi Diaconis talks about how our notion of "randomness" can vary depending how much information you know. With perfect knowledge an event is not random, but we use randomness to account for all the factors or uncertainty that make an event unpredictable. From Program 15 of the Against All Odds series,

Martin Gardner about Persi Diaconis:

I think he was a late teenager. He was a professional card shark, or a card mechanic, as they call it in the trade. He worked ships between New York and South America. Of course, nobody suspected him of being skillful with cards because he was just a teenager. He was a student at City University of New York, and he paid his way through the university with the money that he got from poker games on ships. At that time Persi was very anxious to get into Harvard. The head of statistics department at Harvard was Frederick Mosteller, who is a magic buff. He was very active in magic, and his picture has been on the cover of magic magazines. I knew Mosteller slightly, so I wrote him a letter and said, "This young student is one of the best card mechanics in the country. He does a fantastic second deal and bottom deal." (Those are terms for fake deals. When you are dealing from a deck, there is a way to deal the second card instead of the top card, and there is a way to deal the bottom card instead of the top card.) I got back a letter right away from Mosteller, which said, "If he’s willing to major in statistics, I can get him into Harvard." So I asked Persi if he was willing to major in statistics, and he said, "Of course!" So he got in, got his Ph.D. in statistics, and is now at Stanford.

(from Notices of AMS vol.52, n.6 - pdf)

What is 5 sigma?

Before the announce of the discover of a new boson in LHC by the two experiments ATLAS and CMS, I try to explain the concepts of sigma, $\sigma$. Today I would propose you a video from Fermilab about the statistics used in physics and in particular in the high energy field:

In alternative you can read 5 Sigma - What's that? by Evelyn Lamb.