ChatGPT Sentiment Analysis

[JIGYASAKARAKOTI]

Pull Request for DL-Simplified 💡

Issue Title : ChatGPT Sentiment Analysis

• Info about the related issue (Aim of the project) : The aim of this project is to analyze the sentiments of the tweets made on/against ChatGPT.
• Name: Jigyasa Karakoti
• GitHub ID: JIGYASAKARAKOTI
• Email ID: jigyasakarakoti@gmail.com
• **Idenitfy yourself: SWOCS4

Closes: #411

Describe the add-ons or changes you've made 📃

The code implements sentiment analysis on Twitter data using various machine learning models, including Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Logistic Regression, Naive Bayes, and Random Forest. The dataset is loaded, explored, cleaned, and preprocessed before being used to train and evaluate the models.

Key points:

-> Data Exploration and Cleaning:
The initial exploration revealed the distribution of sentiment labels in the Twitter dataset.
Data cleaning involved removing duplicates, links, special characters, and stopwords.
Balancing the data was performed to address class imbalance.

-> Text Data Preprocessing: Tokenization and lemmatization were applied to convert text data into a suitable format for machine learning models.
Word clouds were used to visualize the most frequent words before and after cleaning.

-> Model Training and Evaluation:
The code implemented models such as CNN, LSTM, Logistic Regression, Naive Bayes, and Random Forest for sentiment analysis.
Training and testing sets were created, and models were trained on the training set and evaluated on the testing set.
Various metrics, including accuracy, recall, precision, and F1 score, were calculated to assess model performance.

-> Visualization:
Visualizations, such as bar charts, pie charts, word clouds, and ROC curves, were used to gain insights into data distribution, model performance, and the impact of data preprocessing.

-> User Interaction:
The code includes an interactive function allowing users to input sentences for sentiment prediction, demonstrating the practical use of the trained models.

-> Model Comparison:
By training and evaluating multiple models, the code provides a comparative analysis of their performance on sentiment analysis tasks.

Type of change ☑️

What sort of change have you made:

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Code style update (formatting, local variables)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

How Has This Been Tested? ⚙️

Describe how it has been tested
Describe how have you verified the changes made

Checklist: ☑️

• My code follows the guidelines of this project.
• I have performed a self-review of my own code.
• I have commented my code, particularly wherever it was hard to understand.
• I have made corresponding changes to the documentation.
• My changes generate no new warnings.
• I have added things that prove my fix is effective or that my feature works.
• Any dependent changes have been merged and published in downstream modules.

[github-actions]

Our team will soon review your PR. Thanks @JIGYASAKARAKOTI :)

[abhisheks008]

1. CNN and LSTM looks good to me, but replace the other machine learning models with deep learning models such as MobileNet, VGG and so on.
2. In the Dataset folder, update the file name from Dataset Link to README.md and put the dataset link there.
3. Add the EDA results/outputs in the README.md file along with the accuracy metrices generated from the project.

@JIGYASAKARAKOTI

[JIGYASAKARAKOTI]

can i use VADER Model and Twitter-roBERTa-base?

[abhisheks008]

can i use VADER Model and Twitter-roBERTa-base?

Yes, you can.

[abhisheks008]

Approved.
@JIGYASAKARAKOTI