Abstract
Electronic text stylometry is concerned with analyzing the writing styles of input electronic texts to extract information about their authors. For example, such extracted data could be the authors’ identity or other aspects, such as their gender and age group. This survey paper presents the following contributions: 1) A description of all stylometry problems in probability terms, under a unified notation. 2) A survey of data representation (or feature extraction) methods. 3) A comprehensive evaluation of 23, 760 feature extraction methods followed by a thorough discussion of the results. This extensive evaluation is critical since the known data representation methods are often not evaluated under the same unified testbed.
Highlights
Improving solvers of stylometry problems is essential for enhancing various application domains, such as forensics, privacy, active-authentication [1]–[3], the detection of compromised accounts [4], recommender systems [5], deception detection, market analysis, and medical diagnosis [6], [7]
FEATURES EVALUATION RESULTS This evaluation aims to identify properties of the feature extraction functions that correspond to the increase in classification accuracy. Since this evaluation tests many feature extraction functions that are special cases of the at least l-frequent dir-directed k-skipped n-grams, the properties that we evaluate their effects on the classification accuracy are l, dir, k, n, and grams
This paper introduced electronic text stylometry problems under a unified notation in probability terms, their importance in enhancing various upper-layer applications, the key challenges currently faced in this field, the critical limitations of stylometry problem solvers, and suggestions for future directions to solve them
Summary
Improving solvers of stylometry problems is essential for enhancing various application domains, such as forensics, privacy (or anti-forensics), active-authentication [1]–[3], the detection of compromised accounts [4], recommender systems [5], deception detection, market analysis, and medical diagnosis [6], [7]. Author identification can be accurately performed on program source codes [8], [9] as well as compiled binaries [10] Enhancing such application domains is growing increasingly more interesting thanks to the availability of large amounts of textual data via the Internet. Electronic text stylometry problems aim at inferring information about authors of input electronic texts. Such inferred information could be the identity of the authors, their genders, age groups, personality types, or even the diagnosis of specific illnesses [6], [7], [11]–[15]. A common taxonomy of electronic text stylometry problem solvers that is often followed by the literature is as follows:
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
