#benfordslaw — Public Fediverse posts

🧐 Oh, look! Another tech bro trying to impress us with a "mind-blowing" #revelation about numbers that magically predict #fraud, because clearly no one knew about Benford's Law until now. 🤯 Spoiler alert: it's just #math, not Hogwarts. 🙄🔢
https://vatsalbakshi.com/blog/benfords-law/ #techbro #BenfordsLaw #skepticism #HackerNews #ngated

An interactive explorer for Benford's Law across real datasets

https://vatsalbakshi.com/blog/benfords-law/

Comments: https://news.ycombinator.com/item?id=48825816

#HackerNews #BenfordsLaw #dataScience #interactiveExplorer #datasets #analytics

Just discovered #BenfordsLaw and it’s blowing my tiny brain and the kids fresh big ones

New! A Look At Benford's Law

Benford's Law is an interesting heuristic in data analysis. It states that in any large collection of numbers that are created naturally, you should expect to see numbers starting with the number 1 about 30% of the time.

In this article we will look at how to calculate Benford's Law and then applying the law against a series of data sets to see if we can spot any issues.

https://www.hashbangcode.com/article/look-benfords-law
#benfordsLaw #php #hashbangcode

Hot off the press (well, Substack) - Benford’s Law in Python

https://codedrome.substack.com/p/benfords-law-in-python

Benford's Law describes the distribution of the first digits of most sets of numeric data and in this article I explain the principles and implement a demonstration in Python.

#statistics #benfordslaw #datascience #programming #python #pythonprogram

New study: "[Impact factor stats for] #OpenAccess journals adhere to #BenfordsLaw more closely than subscribed journals."
https://ieeexplore.ieee.org/abstract/document/10753145
(#paywalled)

Benford's law is weird & fascinating.
https://en.wikipedia.org/wiki/Benford%27s_law

I have no idea what to make of its appearance here.

The authors' take: "The rate of increase in [non-OA] journals is much higher than that of OA journals…[&] violations of Benford's Law are more frequent in [non-OA] journals compared to open [OA] journals.

#JIF

https://en.wikipedia.org/wiki/Benford%27s_law=

Still questioning #electionIntegrity around #election2020 or even #election2022? #Antiscience Hooey.

To all at #FoxNews and all the other Trumpian #socialmedia acolytes still using doubt and fear for political or financial advantage, while pushing us toward civil war: show us the #dataforensics #electionforensics' #datascience or stfu.

Here a blurb on #electiondata from wikipedia on #BenfordsLaw to tell you where you need to look.

Benford's Law turns out to be a poor test for election fraud

"Why do Biden's votes not follow Benford's Law?"

https://youtube.com/watch?v=etx0k1nLn78

The distribution of values fails to meet the multiple orders of magnitude requirement for applying Benford's Law.

"Benford's Law and the Detection of Election Fraud"

Abstract: The proliferation of elections in even those states that are arguably anything but democratic has given rise to a focused interest on developing methods for detecting fraud in the official statistics of a state's election returns. Among these efforts are those that employ Benford's Law, with the most common application being an attempt to proclaim some election or another fraud free or replete with fraud. This essay, however, argues that, despite its apparent utility in looking at other phenomena, Benford's Law is problematical at best as a forensic tool when applied to elections....

https://www.cambridge.org/core/journals/political-analysis/article/benfords-law-and-the-detection-of-election-fraud/3B1D64E822371C461AF3C61CE91AAF6D

#ForensicAccounting #FraudDetection #BenfordsLaw #elections

For a simple #BenfordsLaw like analysis of election data, the logarithmic distribution of the total vote counts must be wide, and the folded distribution nearly uniform. The #Chicago per-ward data in the #USPresidentialElection has a very narrow log10 distribution: 50% from 400 to 620 votes. The mean is 10^2.7 = 490. Biden should typically get about 0.82*490 = 400 votes. A first-digit peak of 3s and 4s for Biden is statistically reasonable.

https://chicagoelections.gov/en/election-results-specifics.asp

For a simple #BenfordsLaw like analysis of election data, the logarithmic distribution of the total vote counts must be wide; and the folded distribution nearly uniform. The #Milwaukee per-ward data #USPresidentialElection has a narrow log10 distbn: 50% from 570 to 1200 votes. Mean: 10^2.9 = 800. Biden should get about 0.7*800 = 560 votes. A first-digit peak of 5s for Biden is statistically reasonable.

http://web.archive.org/web/20201108052246/https://county.milwaukee.gov/EN/County-Clerk/Off-Nav/Election-Results/Election-Results-Fall-2020

Biden: https://archive.today/2qWYe
Trump: https://archive.today/hkWci