Psycholinguistic Personality Profiling using Latent Dirichlet Allocation and Vader

An analysis on the psycholinguistics of twitter tweets

Project Description:

This project explores the application of Natural Language Processing (NLP) techniques, specifically Latent Dirichlet Allocation (LDA) and VADER sentiment analysis, for psycholinguistic personality profiling.

Methodology:

1. Data Acquisition:

   • The project utilizes the "MBTI Personality Type Twitter Dataset" available on Kaggle: https://www.kaggle.com/datasets/mazlumi/mbti-personality-type-twitter-dataset/data

2. Data Preprocessing:

   • Text cleaning: Removing URLs, mentions, and special characters.
   • Tokenization: Splitting text into individual words.
   • Stop word removal: Eliminating common words that do not carry significant meaning.
   • Lemmatization: Reducing words to their base form.

3. Topic Modeling with LDA:

   • Applying LDA to identify latent topics within the text data associated with different MBTI personality types.
   • Visualizing topic distributions using pyLDAvis.

4. Sentiment Analysis with VADER:

   • Utilizing VADER to analyze the sentiment expressed in the text data for each MBTI personality type.
   • Comparing average sentiment scores across different personality types.

Ethical Considerations:

• Potential Biases:
  • It is important to acknowledge that the dataset may contain biases due to the search phrases used for data collection and the self-reported nature of the MBTI personality types.

Results and Discussion:

• The project aims to uncover insights into the relationship between language use and personality traits.

Getting Started:

To run this code, you will need to:

1. Upload the kaggle.json file:

   • Obtain your kaggle.json API key from your Kaggle account settings.
   • Upload this file to your Colab environment.

2. Upload the mbti-personality-type-twitter-dataset.zip file:

   • Download the dataset from the Kaggle link mentioned above.
   • Upload the zipped dataset file to your Colab environment.

License:

This project is licensed under the MIT License. See the LICENSE file for details.