Projects
Sentiment-analysis-of-amazon-review-dataset
This project involves building a sentiment analysis web app using Streamlit and the pre-trained DistilBERT model. The app allows users to input text and receive sentiment analysis results, indicating whether the sentiment is positive or negative, with a confidence score. The app features a simple interface with a text area for input and a button to trigger the analysis. Upon clicking the button, the app processes the input and displays the sentiment along with an emoji representing the sentiment. The application utilizes the Hugging Face transformers library for sentiment classification.
Malaria-Cells-Classifier
In this project, a Convolutional Neural Network (CNN) was developed to classify cell images as either "Parasitized" or "Uninfected" using TensorFlow and Keras. The dataset, consisting of 22,048 images, was split into training and test sets, and a CNN model was trained with data augmentation techniques. The model achieved a validation accuracy of approximately 94.3% and a test accuracy of 93.2%. After training, the model was saved and used to predict the class of new images, achieving an accuracy of 96.8% on a test sample.