Inspiration
Myers Briggs Type Indicator(a.k.a MBTI) is a viral yet controversial indicator of one’s personality. Hence, we have been curious on how one’s linguistic expressions in SNS portrays their personality in terms of MBTI. Many people have faced personality conflicts while messaging their family/friends/colleagues/teachers etc. Moreover, there are increasing cases of individuals becoming socially misfit. Therefore, by analyzing each individual’s MBTI and the relationship compatibility, it can allow users to deal with such conflicts at much ease.
What it does
By identifying the personality type of the person one is talking to, one can get a general sense of how well themselves and the opposing person are congenial to each other in terms of personalities. Hence, they can avoid possible conflicts with those who don’t match themselves well, which can be helpful in several human relationships that bolsters through online network such as business relationships, friend relationships, and family relationships etc. In the end, the repeated use of our machine learning solution can lead to the decrease in social misfit problems evident in our daily society.
How we built it
Refer to Slide attached
Challenges we ran into
The Datasets that we used were quite biased. Low accuracy rate in CNN model
Accomplishments that we're proud of
Refer to Slide attached
What we learned
How NLP is used in Machine Learning
What's next for MBTI Prediction
We plan to build a better model that can take in conversations from Telegram and make a more accurate guess and build a website for this project.
Log in or sign up for Devpost to join the conversation.