Hello, I am Deepika Velapure (she/her)
I am a Master's student in the Computer Science department.
Previously I worked as a Java Developer for 8 years at Cognizant Technology Solutions, India. During my work experience,
I got a lot of opportunities to work on several backend and UI-based projects. I also got the opportunity to work on
Glassfish server upgrades and several test automation tools like Katalon and JGiven.
However, being a student from an Electronics background, I did not get much chance to explore the software technologies
in depth. Therefore, I decided to pursue Master's in Computer Science at Portland State University. I plan to complete
the AI/ML track and learn more about the latest technologies in the computer science field.
The project was implemented using a two-layer neural network consisting of one hidden-layer with a test accuracy of 94%. The input to this network was the MNIST Dataset containing gray scale values of hand-written digits. The output of this network identifies the class or the digit that the input pixels represent.
Click here to view the project details on GitHub!
Diabetes prediction model was implemented using Multilayer Perceptron (MLP) and Gaussian Naïve Bayes classifier. The dataset available at https://www.kaggle.com/datasets/houcembenmansour/predict-diabetes-based-on-diagnostic-measures was used for training and testing the model. The model implemented using MLP has a test accuracy of 87%, while the performance of the model implemented using Gaussian Naïve Bayes classifier was improved after removing the redundant features to a test accuracy of 91%.
Click here to view the project details on GitHub!
This is a one stop API for an individual’s daily schedule, latest news, weather information, date and time, along with some fun facts and activities including creative painting, reading, basic dance and cooking videos. It also periodically reminds to drink water and suggests some desk exercises to relax one’s mind and body.
Click here to view the project details on GitHub!