You are on page 1of 16

PROJECT NAME: PERSONALITY

PREDICTION USING MBTI

 Presented By  Supervised By

Name: Umme kulsum Partha Chakrabarty


Id : 11508009
Name: Kulsum Akter Assistant Professor
nisha Dept. of CSE
Id : 11508038 Cumilla University
MOTIVATION

 Now-a-days personality recognition has attracted of researchers for


developing automatic personality recognition systems.

 Social networking advantage has produced a wide way of


researching sentiment among people and predicting personality.

 In social network site, Twitter becomes one of the most popular


social micro blogging sites, which allows the users to read and post
tweets up to 140 characters.

 Twitter is a tool that enables interested parties to follow individual


users’ thoughts and commentary on events in their lives in almost
real-time .
INTRODUCTION

Personality classification is the task of detecting a


personality by different category of measurement.

The increasing use of social networking sites such as Twitter


and Facebook has prompted the online platform to share
ideas, feelings, opinions and emotions with each other,
reflecting their attitudes, behavior and personality.
CONTRIBUTION

 From the beginning of our work, we tried to learn and find ways to
collect data . The publicly accessible standard data set is collected
from Kaggle.

 Machine learning algorithm is used in the beginning of our research


example, SVM classifier (linear SVC), Naive bias (MultinomialNB)
and XGBoost Classifier.

 Evaluated the output of the experiments on the basis of accuracy,


weighted average precision results, recall and F1-measure.

 A comparison between proposed work and recent works.


METHODOLOGY

 Fig.1.
Proposed
system model
METHODOLOGY (CONT.)

Data Collection

 In dataset, Each user’s has at


most 50 tweets and it consists of
8675 rows.

 The following four groups are


shown:

1) Introversion (I)/ Extroversion (E)


2) Intuition (N) / Sensing (S)
3) Thinking (T) / Feeling (F)
4) Judging (J) / Perceiving (P)
METHODOLOGY (CONT.)

Data Cleaning

 The dataset was cleaned by removing,

1. all URLs, hash tags (e.g. topic), targets(@username),


2. all punctuations ,symbols, numbers,
3. Non-English Tweets,
4. Unnecessary spaces.
METHODOLOGY (CONT.)

Data processing

 Fig.2.
Step of data
processing
METHODOLOGY (CONT.)

Classifier

 Here some machine learning algorithm is used our


research example,
1. SVM classifier (linear SVC),
2. Naive bias (MultinomialNB) and
3. XGBoost Classifier.
RESULT ANALYSIS

 Using Naive bayes, got accuracy for I/E =


76%,N/S=85%,T/F=80%,J/P=73%.

 Using Linear SVC, got accuracy for I/E =


80%,N/S=86%,T/F=80%,J/P=72%.

 Using XGBoost, got accuracy for I/E =


86%,N/S=90%,T/F=84%,J/P=80%.
COMPARISON

Result Comparison with recent works in personality prediction


CONCLUSION

We have achieved satisfactory results with unbalanced


data.
In the case of support vector machines, several kernels
and parameters have been used in early experiments. In
our experiment, it gives a quite good result.

The results using naive Bayes reflect on its ability to


classify text, but its accuracy is not good enough.

Finally by the XGBoost Classifier gives us satisfactory


result.
FUTURE WORK

Gender can be a key factor in the differentiation of the


form of personality based on choice of word. Word-
choice that could help to improve the performance of
any classifier.

Data available in the format of images and videos on


social networking sites can be experimented for the task
of personality traits identification
REFERENCES

 S. bharadwaj, s. sridhar, r. choudhary and r. srinath, ”persona traits


identification based on myers-briggs type indicator(mbti) - a text classification
approach,” 2018 international conference on advances in computing,
communications and informatics (icacci), bangalore, 2018, pp. 1076-1082.
 B. y. pratama and r. sarno, ”personality classification based on twitter text
using naive bayes, knn and svm,” 2015 international conference on data and
software engineering (icodse), yogyakarta, 2015, pp. 170-174.
 S. chaudhary, r. sing, s. t. hasan and i. kaur, “a comparative study of different
classifiers for myers-brigg personality prediction model,” irjet, vol.05,
pp.1410-1413, 2018.
 B. plank, and d. hovy, “personality traits on twitter—or—how to get 1,500
personality tests in a week.” in proceedings of the 6th workshop on
computational approaches to subjectivity, sentiment and social media analysis,
pp. 92-98, 2015. .
REFERENCES(CONT.)

 J. golbeck, c. robles, m. edmondson and k. turner, ”predicting personality from


twitter,” 2011 ieee third international conference on privacy, security, risk and
trust and 2011 ieee third international conference on social computing, boston,
ma, 2011, pp. 149-156.
 F. celli and l. rossi, “the role of emotional stability in twitter conversations,” in
proceedings of the workshop on semantic analysis in social media, association
for computational linguistics, pp. 10-17, 2012.
 Preslav nakov, alan ritter, sara rosenthal, fabrizio sebastiani, and
veselinstoyanov. semeval-2016task4:sentiment analysis in twitter. in
proceedings of the 10th international workshop on semantic evaluation
(semeval-2016), pages 1–18, 2016.
 Pablo gamallo, marcos garcia, “citius: A naive-bayes strategyfor sentiment
analysis on english tweets”, 8th internationalworkshop on semantic evaluation
(semeval 2014), dublin, ireland,aug 23-24 2014, pp 171-175.
THANKS TO ALL

You might also like