Searcharxiv⌕ Search

arXiv subjects

M. V. Simkin

Publications and source records attributed to M. V. Simkin.

At least 19 recordsLinked to original sources
arXiv2021

Estimating the number of serial killers that were never caught

Many serial killers commit tens of murders. At the same time inter-murder intervals can be decades long. This suggests that some serial killers can die of an accident or a disease, having been never caught. We use the distribution of the killers by the number of murders, the distribution of the length of inter-murder intervals and USA life tables to estimate the number of the uncaught killers. The result is that in 20th century there were about seven of such killers. The most prolific of them likely committed over sixty murders.

physics.soc-ph↗
arXiv2021

The correlation coefficient between citation metrics and winning a Nobel or Abel Prize

Computing such correlation coefficient would be straightforward had we had available the rankings given by the prize committee to all scientists in the pool. In reality we only have citation rankings for all scientists. This means, however, that we have the ordinal rankings of the prize winners with regard to citation metrics. I use maximum likelihood method to infer the most probable correlation coefficient to produce the observed pattern of ordinal ranks of the prize winners. I get the correlation coefficients of 0.47 and 0.59 between the composite citation indicator and getting Abel Prize and Fields Medal, respectively. The correlation coefficient between getting a Nobel Prize and the Q-factor is 0.65. These coefficients are of the same magnitude as the correlation coefficient between Elo ratings of the chess players and their popularity measured as numbers of webpages mentioning the players.

cs.DL↗
arXiv2021

Stochastic modeling of scientific impact

Recent research has found that select scientists have a disproportional share of highly cited papers. Researchers reasoned that this could not have happened if success in science was random and introduced a hidden parameter Q, or talent, to explain this finding. So, the talented high-Q scientists have many high impact papers. Here I show that an upgrade of an old random citation copying model could also explain this finding. In the new model the probability of citation copying is not the same for all papers but is proportional to the logarithm of the total number of citations to all papers of its author. Numerical simulations of the model give results similar to the empirical findings of the Q-factor article.

physics.soc-ph↗
arXiv2021

Statistical study of time intervals between murders for serial killers

We study the distribution of 2,837 inter-murder intervals (cooling off periods) for 1,012 American serial killers. The distribution is smooth, following a power law in the region of 10-10,000 days. The power law cuts off where inter-murder intervals become comparable with the length of human life. Otherwise there is no other characteristic scale in the distribution. In particular, we do not see any characteristic spree-killer interval or serial-killer interval, but only a monotonous smooth distribution lacking any features. This suggests that there is only a quantitative difference between serial killers and spree-killers, representing different samples generated by the same underlying phenomenon. The over decade long inter-murder intervals are not anomalies, but rare events described by the same power-law distribution and therefore should not necessarily be looked upon with suspicion, as has been done in a recent case involving a serial killer dubbed as the "Grim Sleeper." This large-scale study supports the conclusions of a previous study, involving three prolific serial killers, and the associated neural net model, which can explain the observed power law distribution.

physics.soc-ph↗
arXiv2020

Success in creative careers depends little on product quality

In the recent article Janosov, Battiston, & Sinatra report that they separated the inputs of talent and luck in creative careers. They build on the previous work of Sinatra et al which introduced the Q-model. Under the model the popularity of different elements of culture is a product of two factors: a random factor and a Qfactor, or talent. The latter is fixed for an individual but randomly distributed among different people. This way they explain how some individuals can consistently produce high-impact work. They extract the Q-factors for different scientists, writers, and movie makers from statistical data on popularity of their work. However, in their article they reluctantly state that there is little correlation between popularity and quality ratings of of books and movies (correlation coefficients 0.022 and 0.15). I analyzed the data of the original Q-factor article and obtained a correlation between the citation-based Q-factor and Nobel Prize winning of merely 0.19. I also briefly review few other experiments that found a meager, sometimes even negative, correlation between popularity and quality of cultural products. I conclude that, if there is an ability associated with a high Q-factor it should be more of a marketing ability than an ability to produce a higher quality product. Janosov,

cs.DL↗
arXiv2015

Chess players' fame versus their merit

We investigate a pool of international chess title holders born between 1901 and 1943. Using Elo ratings we compute for every player his expected score in a game with a randomly selected player from the pool. We use this figure as player's merit. We measure players' fame as the number of Google hits. The correlation between fame and merit is 0.38. At the same time the correlation between the logarithm of fame and merit is 0.61. This suggests that fame grows exponentially with merit.

physics.soc-ph↗
arXiv2014

Why does attention to web articles fall with time?

We analyze access statistics of a hundred and fifty blog entries and news articles, for periods of up to three years. Access rate falls as an inverse power of time passed since publication. The power law holds for periods of up to thousand days. The exponents are different for different blogs and are distributed between 0.6 and 3.2. We argue that the decay of attention to a web article is caused by the link to it first dropping down the list of links on the website's front page, and then disappearing from the front page and its subsequent movement further into background. The other proposed explanations that use a decaying with time novelty factor, or some intricate theory of human dynamics cannot explain all of the experimental observations.

cs.IR↗
arXiv2014

Stochastic modeling of a serial killer

We analyze the time pattern of the activity of a serial killer, who during twelve years had murdered 53 people. The plot of the cumulative number of murders as a function of time is of "Devil's staircase" type. The distribution of the intervals between murders (step length) follows a power law with the exponent of 1.4. We propose a model according to which the serial killer commits murders when neuronal excitation in his brain exceeds certain threshold. We model this neural activity as a branching process, which in turn is approximated by a random walk. As the distribution of the random walk return times is a power law with the exponent 1.5, the distribution of the inter-murder intervals is thus explained. We illustrate analytical results by numerical simulation. Time pattern activity data from two other serial killers further substantiate our analysis.

physics.soc-ph↗
arXiv2013

Statistics against irritations: a response to Dickens's apologists

In a recent article (arXiv:0909.2479) I reported the results of the test, where the takers had to tell the prose of Charles Dickens from the prose of Edward Bulwer-Lytton. The former is a required reading in school, and the latter has a bad writing contest named after him. Nevertheless, the test-takers performed on the level of random guessing. This research has met much criticism, which I refute the in the present article.

physics.soc-ph↗
arXiv2013

Scientific evaluation of Charles Dickens

I report the results of the test, where the takers had to tell the prose of Charles Dickens from that of Edward Bulwer-Lytton, who is considered by many to be the worst writer in history of letters. The average score is about 50%, which is on the level of random guessing. This suggests that the quality of Dickens' prose is the same as of that of Bulwer-Lytton.

physics.soc-ph↗
arXiv2013

A mathematical theory of fame

We study empirically how the fame of WWI fighter-pilot aces, measured in numbers of web pages mentioning them, is related to their achievement, measured in numbers of opponent aircraft destroyed. We find that on the average fame grows exponentially with achievement; the correlation coefficient between achievement and the logarithm of fame is 0.72. The number of people with a particular level of achievement decreases exponentially with the level, leading to a power-law distribution of fame. We propose a stochastic model that can explain the exponential growth of fame with achievement. Next, we hypothesize that the same functional relation between achievement and fame that we found for the aces holds for other professions. This allows us to estimate achievement for professions where an unquestionable and universally accepted measure of achievement does not exist. We apply the method to Nobel Prize winners in Physics. For example, we obtain that Paul Dirac, who is a hundred times less famous than Einstein contributed to physics only two times less. We compare our results with Landau's ranking.

physics.soc-ph↗
arXiv2012

Stochastic modeling of Congress

We analyze the dynamics of growth of the number of congressmen supporting the resolution HR1207 to audit the Federal Reserve. The plot of the total number of co-sponsors as a function of time is of "Devil's staircase" type. The distribution of the numbers of new co-sponsors joining during a particular day (step height) follows a power law. The distribution of the length of intervals between additions of new co-sponsors (step length) also follows a power law. We use a modification of Bak-Tang-Wiesenfeld sandpile model to simulate the dynamics of Congress and obtain a good agreement with the data.

physics.soc-ph↗
arXiv2011

Monkeys get a silver in Abstract Art Olympics

Experiment shows that art students prefer abstract art to monkey art in about two-third of the cases. Since the number is above 50%, some argue that abstract art is different and better than animal art. I compare this result with figure skating competitions, where on average 73% of judges prefer gold medalist to silver medalist. This means that the difference between abstract artists and animal artists is less than the difference between gold and silver medalists.

physics.soc-ph↗
arXiv2011

Abstract art grandmasters score like class D amateurs

Hawley-Dolan and Winner had asked the art students to compare paintings by abstract artists with paintings made by a child or by an animal. In 67% of the cases, art students said that the painting by a renowned artist is better. I compare this with the winning probability of the chessplayers of different ratings. I conclude that the great artists score on the level of class D amateurs.

physics.pop-ph↗
arXiv2011

Berezovsky number

Berezovsky number is defined analogously to Erdos number. Berezovsky network is investigated.

physics.gen-ph↗
arXiv2011

Theory of citing

We present empirical data on misprints in citations to twelve high-profile papers. The great majority of misprints are identical to misprints in articles that earlier cited the same paper. The distribution of the numbers of misprint repetitions follows a power law. We develop a stochastic model of the citation process, which explains these findings and shows that about 70-90% of scientific citations are copied from the lists of references used in other papers. Citation copying can explain not only why some misprints become popular, but also why some papers become highly cited. We show that a model where a scientist picks few random papers, cites them, and copies a fraction of their references accounts quantitatively for empirically observed distribution of citations.

physics.soc-ph↗
arXiv2011

Scientific comparison of Mozart and Salieri

I report the results of the internet quiz, where the takers had to tell the music of Mozart from that of Salieri. The average score earned by over eleven thousand quiz-takers is 61%. This suggests that the music of Mozart is of about the same quality as the music of Salieri.

physics.soc-ph↗