Predicting MBTI types based on Twitter posts
Project description
MBTI Predictor
Predicting someone's MBTI type based on their online posts, using AI.
Approach
I originally attempted to use a simple LSTM, using nltk and Keras. However, this approach led to the model badly overfitting.
I then decided to re-write the program and use a bidirectional LSTM. This was slightly better and, somewhat, prevented overfitting. However, the model trained slowly due to a low learning rate.
I then decided to use a TF BERT Model as a layer within a Keras model. I thought this would improve the performance of the predictor but my GPU (NVIDIA GTX 1650) couldn't run it. I decided to publish this project as a package so that others could attempt to train the BERT model.
Findings
I could not find a significant correlation between a post and the MBTI type of its poster. Feel free to download the source code and try for yourself.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
File details
Details for the file matthewflegg-mbti-0.0.2.tar.gz
.
File metadata
- Download URL: matthewflegg-mbti-0.0.2.tar.gz
- Upload date:
- Size: 6.8 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.8.0 pkginfo/1.8.2 readme-renderer/32.0 requests/2.27.1 requests-toolbelt/0.9.1 urllib3/1.26.8 tqdm/4.62.3 importlib-metadata/4.11.1 keyring/23.5.0 rfc3986/2.0.0 colorama/0.4.4 CPython/3.9.7
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | a3a4e080ddab54d137339d2114f51c6fae9486e48c913e84c0d3f19da53dd29d |
|
MD5 | 78384de023697caa1f21aab354e30ead |
|
BLAKE2b-256 | 4665acae0ca457192acb6e0118183984b2242374bfa01159ecaadb315bf658ff |