Kethan Pabbi

A hardworking and passionate job seeker with strong organizational skills eager to secure an entry-level Data Analyst position. Complex problem-solver with an analytical and driven mindset. Dedicated to achieving demanding development objectives according to tight schedules while producing impeccable code. Detail-oriented and systematic with excellent follow-through and a superior work ethic. Excellent communication, project management and problem-solving abilities.

YouTube Gender Prediction Using Faces (May 2022- September 2022):

• Access a YouTube video through API, implement face detection through RetinaFace and face recognition through DeepFace.

• Implement gender prediction through various methods such as OpenCV, Dlib, and Mtcnn and compare the results obtained.

• Deploy the infrastructure through flask, HTML, CSS and JavaScript.

• Develop an interface to show the stats of the video to the user.

• Provide the user with the MP4 file of the video after face detection and gender prediction.

• Presented the finding at WISPNET Conference and published the paper in IEEE Transactions.

• Summarize and visualize the Amazon Fine Foods dataset using techniques like boxplot, scatterplot, and frameworks like TensorFlow, Pandas and NumPy.

• Implemented Transformer attention model with regularization and SoftMax to classify a review as positive or negative. Used Rouge and Bleau to evaluate the model’s accuracy.

• Developed a machine learning model using a K-Neighbors classifier to predict the weather condition in a city on a given date, based on the data of previous decades.

• Implemented a simple user interface in python to take the date and city as input and display the weather prediction for it.