ADATPM: Analysis and Design of Author’s Trait Processing Module for textual Data
Gaurav Kansal1, Maneesha2
1Gaurav Kansal, E&R, Infosys.
2Maneesha, E&R, Infosys
Manuscript received on September 01, 2012. | Revised Manuscript received on September 02, 2012. | Manuscript published on September 05, 2012. | PP: 150-158 | Volume-2 Issue-4, September 2012. | Retrieval Number: D0918072412/2012©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Trait theory is a major approach to the study of human personality. Personality is the branch of psychology which is concerned with providing a systematic account of the ways by which we can differentiate one-another. Individuals differ from one another in a variety of ways: their anatomical and physiognomic characteristics, their personal appearance, grooming, manner of dress, their social backgrounds, roles and other demographic characteristics, their effect on others or social stimulus value and their temporary states, moods, attitudes and activities at any given moment in time.. In this paper we have designed a system that takes text input and returns the author’s trait accordingly. Since human tendencies are largely dependent on environmental and situational consistencies, we have considered five different traits in our identification. These are High Extrovert, Low Extrovert, High Introvert, Low Introvert and Ambivert. Our algorithm refines the author’s text under eight different properties. The text undergoes POS tagger where each word is assigned a tag. After analyzing the tag we generate Feature Vector matrix (FVM), we use this FVM for our analysis as well as for the classification. We have applied our proposed algorithm on different 280 files. These files are also annotated by human. We compare the result got from human annotation and proposed algorithm and we found that the accuracy of our algorithm is 84.26%.
Keywords: FVM, POS, SVM, Trait Theory.