Skip to content

The current project provides a Machine Learning trained model that is able to classify the trait with maximum value of the Big Five Personality Test, given the answers of this one.

Notifications You must be signed in to change notification settings

UNSpecializationAI/BigFivePersonality

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

34 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Big Five Personality Test

Table of contents

  • Pipeline
  • Data
  • Modelling
  • References The following project developed a solution to grade Big Five Personality Tests. The project was developed according to CRISP-DM phases, as follows:

CRISPDM

Figure 1. CRISP-DM Methodology.
Source: [1-2].

The target variable corresponds to the trait with max value of the test. Further information about the traits can be found below.

Trait Class Description
Extroversion E High scores: People who tend to be social
Low scores: People who prefer to work alone in projects
Agreeableness A High scores: Politically correct people
Low Scores: Direct people
Conscientiousness C High Scores: People who tend to follow the rules and prefer order
Low Scores: People who tend to be disorganized
Neuroticism N How emotional the person can be
Openness to Experience O High Scores: People who tend to “dream with their eyes open”
Low Scores: People who tend to “have their feet on the ground”

Pipeline

The following pipeline summarizes the steps and procedures which have been taken into account during training / validation and testing.

Pipeline

Figure 2. Pipeline.

Data

The exploratory Data Analysis allowed us to conclude that there were a imbalance in target variable. Then two approaches to go over this situation were followed:

  • UnderSampling
  • Oversampling

Data can be downloaded here.

Modelling

The project at hand corresponds to a problem of multiclass classification. The following classifiers were trained and validated:

  • Logistic Regression Classifier
  • Decision Tree Classifier
  • Random Forest Classifier
  • Naive Bayes Classifier

References

[1]	M. F. Hornick, E. Marcadé, and S. Venkayala, “Chapter 3 - Data Mining Process,” in The Morgan Kaufmann Series in Data Management Systems, M. F. Hornick, E. Marcadé, and S. B. T.-J. D. M. Venkayala, Eds. Burlington: Morgan Kaufmann, 2007, pp. 51–83.
[2]	“File:CRISP-DM Process Diagram.png - Wikimedia Commons.” .
[3]	B. Tunguz, L. Petar, and M. Akdag, “Big Five Personality Test | 1M Answers to 50 personality items, and technical information.” [Online]. Available: https://www.kaggle.com/tunguz/big-five-personality-test. [Accessed: 20-Jun-2020].

About

The current project provides a Machine Learning trained model that is able to classify the trait with maximum value of the Big Five Personality Test, given the answers of this one.

Topics

Resources

Stars

Watchers

Forks