Gagandeep Kaur Khera

English–French Language Translation. |NLP, Simple RNN, Bidirectional RNN, and Python.

• Built a neural machine translation model using a transformer-based architecture to translate English text into French. Applied NLP preprocessing, trained on bilingual datasets, and evaluated model performance.

Emotion Detection from Facial Images. Python, CNN, OpenCV, TensorFlow/Keras.

• Developed a CNN-based model to detect and classify human emotions (e.g., happy, sad, angry, surprised) from facial images.
• Trained the model on emotion-labeled datasets with data augmentation to improve generalization.
• Evaluated performance using accuracy across multiple emotion categories.

• Developed supervised ML model using Linear Regression, XGBoost, and Random Forest to predict housing prices based on location, size, and amenities.
  
  Performed extensive EDA with Pandas, Matplotlib, and Seaborn to identify key features and outliers.
  
  Tuned hyperparameters using GridSearchCV and improved model R² score.

E-Commerce Store Review Classification using Deep Learning Techniques with Cloud Deployment (sentiment_analysis)

• Built a text classification model using TF-IDF and Logistic Regression to analyze customer sentiments from e-commerce reviews.
  
  Preprocessed raw text data using NLTK and spaCy (tokenization, stopword removal, lemmatization).
  Achieved over 90% accuracy on test data; visualized insights using word clouds and sentiment polarity plots.
• https://github.com/Gagankhera/Sentiment_Analysis_NLP_Project

• Delivered personalized tutoring in Mathematics, Science, and English to students from diverse academic backgrounds, improving their performance and confidence.
  
  Designed customized learning plans and practice modules based on student needs, curriculum standards, and individual learning styles.
  
  Built strong rapport with students and parents, leading to consistent referrals and repeat engagements.
  
  Conducted regular assessments and progress reviews to identify knowledge gaps and adjust teaching strategies.