Please use this identifier to cite or link to this item: http://hdl.handle.net/10316/100143
Title: Providing a Model for Detecting Personality of People in Social Networks
Authors: Asghari, Omid
Orientador: Hosseinzadeh, Mehdi
Sheikholharam Mashhadi, Peyman
Keywords: Social networks; MBTI; Deep Learning; Supervised learning; Machine learning
Issue Date: 2018
Place of publication or event: Islamic Azad University
Abstract: The exponential growth of the use of social networks in cyberspace has led individuals to share a lot of information, including image, voice and text. Analyzing social Networks data provides details information about individual personality. The complexity and large volume of extracted data is that such that it requires to apply machine learning algorithms. In this paper the author has analyzed the behavior patterns using writing. we first need to know the standard personality models. one of the most reliable models is MBTI model. the goal of this thesis is to find a supervised Learning model that can determine personality factors by people writing in social networks. due to the fact that experience has shown for complex problems with many parameters, deep learning methods can be more effective, we used deep learning model and two personality factors are considered an introversion - extroversion and intuition - sensing. The obtained results show a good accuracy which based on we were able to distinguish an introversion - extroversion personality factor with precision of 62 % accuracy and intuition – sensing factor with precision of 58 %
Description: Documentos apresentados no âmbito do reconhecimento de graus e diplomas estrangeiros
URI: http://hdl.handle.net/10316/100143
Rights: openAccess
Appears in Collections:UC - Reconhecimento de graus e diplomas estrangeiros

Files in This Item:
File Description SizeFormat
Thesis.pdfDissertação2.29 MBAdobe PDFView/Open
Show full item record

Page view(s)

32
checked on Aug 5, 2022

Download(s)

16
checked on Aug 5, 2022

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.