e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:04/Issue:04/April-2022 Impact Factor- 6.752 www.irjmets.com
PERSONALITY PREDICTION USING MACHINE LEARNING
Devesh Agarwal*1, Mr. M. Karthikeyan*2
*1Student, Computer Science And Engineering, SRM Institute Of Science And Technology,
Jaipur, India.
*2Assistant Professor, Computer Science and Technology, SRM Institute of Science and
Technology, Chennai, India.
ABSTRACT
Data Science and AI are revolutionizing the planet through technical transformations. We can observe many
machine learning applications in our day-to-day lives, but one of the greatest applications of machine learning
is to classify individuals based on their personality traits. Every person on the planet is unique and carries a
unique personality type. The availability of a high- dimensional data has paved the way for increasing
marketing campaigns' effectiveness by targeting specific groups of people. Such personality-based
communications are highly effective in increasing the recognition and attractiveness of products and services. e
developed a system for personality prediction using Big five personality traits in this project. Daily
numerous students are writing competitive exams that mainly focuses on his/her Personality. The main
motive of these exams is to check the student’s personality and skills. This project helps to write the personality
test and check the personality of the person. From the personality classification, the person can view the type of
personality and can improve the personality based upon the results.
Keywords: Big Five Personality Model, Feature Analysis, Predicting Personality, Personality Traits.
I. INTRODUCTION
The big five personality traits, often mentioned as OCEAN, and sometimes CANOE, are Openness,
Conscientiousness, Extroversion, Agreeableness, and Neuroticism. These five personality traits represent
broad domains of human behavior and account for differences in both personality and decision making. Today,
the model is employed by HR practitioners to measure potential employees and marketers to
understand the audiences of their products We are deploying OCEAN Model in this project for developing the
algorithm.
Openness to experience: Also called as intellect or imagination, this personality trait represents the
willingness to try new things and think out of the box. This trait includes insightfulness, originality, and
curiosity.
Conscientiousness: The desire to be careful, diligent and regulate immediate gratification with self-
discipline. This trait includes ambition, discipline, consistency and reliability.
Extroversion: A state where an individual draw energy from others and seek social connections or
interaction, as opposed to being alone. This trait includes being outgoing, energetic, and confident.
Agreeableness: The measure of how an individual interacts with other individuals, characterized by the
degree of compassion and cooperation. This trait includes tactfulness, kindness, and loyalty.
Neuroticism: A tendency for negative personality traits, emotional instability, and self-destructive thinking.
This trait includes pessimism, anxiety, insecurity, and fearfulness.
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[212]
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:04/Issue:04/April-2022 Impact Factor- 6.752 www.irjmets.com
II. LITERATURE SURVEY
“Text-based Personality Prediction using multiple Social Media data sources”
This proposed a multi-modal deep learning architecture with a pre-trained language model BERT, RoBERTa,
and XLNet; alongside additional NLP Features (sentiment analysis, TF-IGM, NRC emotion lexicon database) as
features extraction method for personality prediction system. This method performs better when compared to
the previous method in predicting personality traits.
“A Neural Network Approach to Personality Prediction based on the Big-Five Model”
This model was proposed by Mayuri Pundlik Kalghatgi. For the model analytics, parallelism is examined
between the personality traits of an individual's linguistic information. Personality traits are identified by
means of linguistic information which is enabled by the OCEAN model. This shows the personality traits that
can be applied to a number of fields, such as business intelligence, marketing and psychology.
“The General Factor of Personality”
Dimitri van der Linden has published the paper. A criterion-related validity review and a meta-analysis of Big
Five intercorrelations. To assess the presence of a GFP, this study looked at the interrelationship between the
five personality traits i.e., Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. The
meta-analysis provides evidence for GFP at the highest operational level and this paper concluded that the GFP
features a major factor because it's linked to supervisor-rated job performance.
III. PROPOSED WORK
One of the major challenges for the project will be the collection of input datasets for the algorithm. For
conducting the test, we are using K – Mean Clustering Algorithm. The dataset for testing the algorithm is
collected from the participant. This is done by giving a questionnaire on personality classification. Then, the
collected information is fed to the personality classification algorithm i.e., K-Mean Clustering Algorithm. Finally,
the algorithm evaluates the data on the basis of the big five personality traits and displays the result. To give
you an idea, here is a diagrammatical representation of the whole process we will follow for the conduction of
the project.
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[213]
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:04/Issue:04/April-2022 Impact Factor- 6.752 www.irjmets.com
Data Collection
Data for testing purposes is procured from real-time participants through the means of Google Forms. They
were used to record the questions and answers, which were then saved in a CSV file for easy data testing and
retrieval. The questions in the form are based upon one of the traits of the OCEAN Model and the user is marked
on a five-point radio button as shown below. Here is a sample Google Form for illustration purposes.
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[214]
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:04/Issue:04/April-2022 Impact Factor- 6.752 www.irjmets.com
IV. DESIGN ALGORITHM
K-Mean Clustering Algorithm
K-Means Clustering is an unsupervised learning technique, which groups the unlabeled dataset into different
clusters. Here K defines the number of pre-defined clusters that need to be created in the process, as if K=2,
there'll be two clusters, and for K=3, there'll be three clusters, and so on. It is a centroid-based technique,
where each cluster is related to a centroid. The motive of the algorithm is to minimize the sum of distances
between the data point and their corresponding clusters. The algorithm takes the unlabeled dataset as input,
divides the dataset into k-number of clusters, and repeats the process until it doesn’t find the simplest cluster.
The value of k is already determined in this algorithm. The k-means clustering algorithm mainly performs two
tasks: Determines the best value for K center points or centroids by an iterative process. Assigns each data
point to the closest k-center. Those data points which are near to the particular center-k, create a cluster.
Ideal K-Mean Clustering Graph
Logistic Regression
Logistic regression is known as the function used at the core of the method, the logistic function. The logistic
function, also called the sigmoid function was developed by statisticians to explain the properties of population
growth in ecology, rising quickly and maxing out at the carrying capacity of the environment. It’s an S-shaped
curve which will take any real-valued number and map it into a value between 0 and 1 but never exactly at
those limits.
1 / (1 + e^-value)
Where e is the base of the natural logarithms (Euler’s number or the EXP() function in your spreadsheet) and
value is the actual numerical value that you simply want to transform. Below is a plot of the numbers between -
5 and 5 transformed into the range 0 and 1 using the logistic function.
Logistic Regression Function
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[215]
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:04/Issue:04/April-2022 Impact Factor- 6.752 www.irjmets.com
V. IMPLEMENTATION
We are dividing the dataset into the training and testing part. The dataset is further scaled using the Standard
sklearn Library. The training part is 70% and the testing part is 30%.
The dataset consists of 972 rows and 8 columns where each column comprises of one of the five personality
trait in OCEAN Model, Age and Gender of the candidate. The rows comprise of participant data.
The dataset for training purposes is taken from https://www.kaggle.com/tunguz/big-five-personality-test.
The data is further converted into array and is fed for Logistic Regression. Then, clustering of each of the
personality trait happens using K-Mean Clustering Algorithm. Thus, the algorithm predicts the personalities of
each participant in the survey and displays the result.
VI. CONCLUSION
This Personality Prediction Model can be used in E-commerce sites, in Competitive exams, Psychometric tests,
Matrimonial sites, Government sectors like Army, Navy, Air force. Thus the personality is automatically
classified by the system after user attempts the survey by the data set provided in the back end. Personality
analysis and prediction is more in recent times so further in future more personality traits can be added.
Further any improvement can be done using the data set and algorithms to improve the accuracy and can be
helpful for career guidance module.
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[216]
e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:04/Issue:04/April-2022 Impact Factor- 6.752 www.irjmets.com
VII. REFERENCES
[1] “A Novel Approach to Predict Personality of a Person”: Prof. Waheeda Dhokley, Randeria Kaiwan
Jehangir, Shaikh Nabeel Rashid, Shaikh Almas Mohd Sarwar.
[2] “Personality Prediction Via CV Analysis using Machine Learning” Atharva Kulkarni, Tanuj Shankarwar ,
Siddharth Thorat.
[3] “A Neural Network Approach to Personality Prediction based on the Big-Five Model”, Mayuri Pundlik
Kalghatgi, Manjula Ramannavar , Dr. Nandini S. Sidnal.
[4] “Predicting personality from patterns of behavior collected with smartphones”, Clemens Stachl, Quey
Au, Ramona Schoedel, Markus Buhner.
[5] Enjoy Algorithm:
[6] https://www.enjoyalgorithms.com/blog/personality-prediction-using-ml
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[217]
