Gender detection and identifying one's handwriting with handwriting analysis
No Thumbnail Available
Date
2017
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Pergamon-Elsevier Science Ltd
Access Rights
info:eu-repo/semantics/closedAccess
Abstract
Today, graphology is seen as an experimental field of science that is dedicated to suggest ideas about diseases, profession choices, mood and characteristics of a person by investigating his/her handwriting. Graphology is in cooperation with medicine, psychology, sociology or other disciplines that are based on observation. Graphology is used for staff recruitment in business, diagnoses in medicine, identification of criminals in forensics, choosing a profession in education, guidance and counseling and other practices at every level of social structure. It is quite interesting that the number of the scientific studies on graphology is limited around the world, that there are no specific institutions providing education of graphology and that institutions except for a few international corporations do not benefit from graphology at all. In terms of demographic properties, many statistical and mathematical analyses investigate similar and different variables. Especially, differences regarding gender have become subject to research. Therefore, detecting gender through handwriting can give pace to research in other disciplines. Moreover, the research can be useful in any field where gender detection is needed. This study fulfills two objectives. The first one is to find out whether a writer can identify his/her own handwriting. The second objective is to detect the gender of a writer of a text with the help of graphology and computer sciences. The impact of the study is reflected in the fact that findings can be used in fields where gender detection is needed, and that the detection is done with the help of expert and intelligent systems. At the end of the study, gender detection was performed for the individuals by making use of 133 attributes. Then, a decision tree and lists of rules were created with some algorithms. The purpose was to detect the gender of the person by making a character analysis of the handwriting with the help of decision tree formation methods in data mining. The analysis showed that it is possible to detect the gender of a person with the use of the specified attributes. The study reached a success level of 93.75% with ID3 algorithm. (C) 2017 Elsevier Ltd. All rights reserved.
Description
Keywords
Gender Detection, Graphology, Data Mining, Decision Tree, J48, ID3
Journal or Series
Expert Systems With Applications
WoS Q Value
Q1
Scopus Q Value
Q1
Volume
79
