Author Identification, Verification And Profiling Using Stylometry

Zahra Ahmad
5 min readDec 20, 2021

Stylometry is an analytical and statistical study of written text based on the assumption that we follow specific patterns that uniquely identify us.

Photo by Aaron Burden on Unsplash

Introduction

Between October 1787 and April 1788, Alexander Hamilton, James Madison, and John Jay published The Federalist Papers, a series of 85 documents, originally disseminated anonymously, supporting the ratification of the United States Constitution.

Two days prior to his death in 1804, Hamilton compiled a list of the Federalist Papers that he claimed to have written, and provided the list to his lawyer.

The trouble was that this list directly contradicted a list that Madison would publish in 1817, in which Madison claimed to have written twelve of the papers that Hamilton had asserted were his own.

These twelve disputed Federalist Papers became an early notable instance of an authorship attribution problem, for which scholars could not conclusively agree on which of the disputed papers were written by Hamilton, and which were written by Madison.

In the mid-20th century, Mosteller and Wallace (1963) approached the problem of attributing authorship to the twelve disputed papers by means of statistical inference, by studying the distribution of a…

--

--

Zahra Ahmad

MSc in Data Science, I love to extract the hell out of any raw data, sexy plots and figures are my coffee