Files

• call_of_duty_reviews_50000.csv: The dataset containing 50,000 reviews of the Call of Duty game.
• logistic_regression_model.pkl: The trained Logistic Regression model.
• naive_bayes_model.pkl: The trained Naive Bayes model.
• tfidf_vectorizer.pkl: The trained TF-IDF vectorizer.
• Sklrean_sentiment_analysisis_using_Naive_byes_and_logistic_regression.ipynb: Jupyter notebook for training and evaluating the sentiment analysis models.
• Text_analysis.ipynb: Jupyter notebook for additional text analysis and visualization.

Requirements

• Python 3.x
• Jupyter Notebook
• pandas
• numpy
• scikit-learn
• nltk
• matplotlib
• seaborn
• tqdm
• langdetect
• emoji

Installation

1. Clone the repository:

   git clone <repository-url>
   cd cod_data

2. Install the required packages:

   pip install -r requirements.txt

3. Download NLTK data:

   import nltk
   nltk.download('stopwords')
   nltk.download('punkt')
   nltk.download('vader_lexicon')

Usage

1. Open the Jupyter notebooks:

   jupyter notebook

2. Run the cells in Sklrean_sentiment_analysisis_using_Naive_byes_and_logistic_regression.ipynb to train and evaluate the sentiment analysis models.

3. Run the cells in Text_analysis.ipynb for additional text analysis and visualization.

Example

# Load models and vectorizer
with open('naive_bayes_model.pkl', 'rb') as f:
    loaded_nb_model = pickle.load(f)

with open('logistic_regression_model.pkl', 'rb') as f:
    loaded_lr_model = pickle.load(f)

with open('tfidf_vectorizer.pkl', 'rb') as f:
    loaded_vectorizer = pickle.load(f)

# Example usage of the loaded models
new_review = ["I used to love this game It's not very fantastic! don't ever try this game very bad"]

# Preprocess and transform the new review
new_review_tfidf = loaded_vectorizer.transform([preprocess_text(new_review[0])])

# Predict sentiment using Naive Bayes
predicted_sentiment_nb = loaded_nb_model.predict(new_review_tfidf)
print(f"Predicted Sentiment (Naive Bayes): {predicted_sentiment_nb[0]}")

# Predict sentiment using Logistic Regression
predicted_sentiment_lr = loaded_lr_model.predict(new_review_tfidf)
print(f"Predicted Sentiment (Logistic Regression): {predicted_sentiment_lr[0]}")