Power law distributions in information science: Making the case for logarithmic binning
S. Milojević. Journal of the American Society for Information Science and Technology, 61 (12):
2417--2425(2010)
DOI: 10.1002/asi.21426
Abstract
We suggest partial logarithmic binning as the method of choice for uncovering the nature of many distributions encountered in information science (IS). Logarithmic binning retrieves information and trends “not visible” in noisy power law tails. We also argue that obtaining the exponent from logarithmically binned data using a simple least square method is in some cases warranted in addition to methods such as the maximum likelihood. We also show why often-used cumulative distributions can make it difficult to distinguish noise from genuine features and to obtain an accurate power law exponent of the underlying distribution. The treatment is nontechnical, aimed at IS researchers with little or no background in mathematics.
Description
Power law distributions in information science: Making the case for logarithmic binning - Milojević - 2010 - Journal of the American Society for Information Science and Technology - Wiley Online Library
%0 Journal Article
%1 milojevi2010power
%A Milojević, Staša
%D 2010
%I Wiley Subscription Services, Inc., A Wiley Company
%J Journal of the American Society for Information Science and Technology
%K binning distribution fitting law power
%N 12
%P 2417--2425
%R 10.1002/asi.21426
%T Power law distributions in information science: Making the case for logarithmic binning
%U http://dx.doi.org/10.1002/asi.21426
%V 61
%X We suggest partial logarithmic binning as the method of choice for uncovering the nature of many distributions encountered in information science (IS). Logarithmic binning retrieves information and trends “not visible” in noisy power law tails. We also argue that obtaining the exponent from logarithmically binned data using a simple least square method is in some cases warranted in addition to methods such as the maximum likelihood. We also show why often-used cumulative distributions can make it difficult to distinguish noise from genuine features and to obtain an accurate power law exponent of the underlying distribution. The treatment is nontechnical, aimed at IS researchers with little or no background in mathematics.
@article{milojevi2010power,
abstract = {We suggest partial logarithmic binning as the method of choice for uncovering the nature of many distributions encountered in information science (IS). Logarithmic binning retrieves information and trends “not visible” in noisy power law tails. We also argue that obtaining the exponent from logarithmically binned data using a simple least square method is in some cases warranted in addition to methods such as the maximum likelihood. We also show why often-used cumulative distributions can make it difficult to distinguish noise from genuine features and to obtain an accurate power law exponent of the underlying distribution. The treatment is nontechnical, aimed at IS researchers with little or no background in mathematics.},
added-at = {2012-07-05T00:15:52.000+0200},
author = {Milojević, Staša},
biburl = {https://www.bibsonomy.org/bibtex/2987d980c24ef23908f2ccdb820d41ea2/folke},
description = {Power law distributions in information science: Making the case for logarithmic binning - Milojević - 2010 - Journal of the American Society for Information Science and Technology - Wiley Online Library},
doi = {10.1002/asi.21426},
interhash = {a13ef5af5ecf26b7e0f20a471fdc3090},
intrahash = {987d980c24ef23908f2ccdb820d41ea2},
issn = {1532-2890},
journal = {Journal of the American Society for Information Science and Technology},
keywords = {binning distribution fitting law power},
number = 12,
pages = {2417--2425},
publisher = {Wiley Subscription Services, Inc., A Wiley Company},
timestamp = {2012-07-05T00:15:52.000+0200},
title = {Power law distributions in information science: Making the case for logarithmic binning},
url = {http://dx.doi.org/10.1002/asi.21426},
volume = 61,
year = 2010
}