Siddharth Kashinath

Complex problem-solver with analytical and driven mindset. Dedicated to achieving demanding development objectives according to tight schedules with determination.

1) Breast Cancer Classifier – Classification.

• Develop a model to predict whether the cancer is benign or malignant using K Nearest Neighbors with an peak validation accuracy of 94-96%.
• Used scikit-learn to build a trained model which compares over various factors over a Python sklearn library dataset to predict with high accuracy the malignancy of the tumor using k = 100 Neighbors for this setup.

2) Smart Parking System with License Detection - Sensor activation, Image Processing.

• Developed a web based cloud parking system that predicts empty slots and can detect license plates.
• Restricted parking slots can be arranged for an office or home setting, which only allows a certain preexisting license plate number.
• Built on a Arduino framework and uses a Cloud IO for website of portal; Android and IOS app could be developed in the future.
• Uses still image processing for License Detection.

Yelp Rating Prediction - Real world dataset, Regression

• Developed a model that predicts how the rating of the customer varies on the basis of several factor sthat were madeavailable inthe dataset.
• Using scikit learn, the model is built and trained and a finally a plot of multiple linear as well as logistical regressionis presented on the basis of different categories of customer demographics.

Customer Segmentation – Clustering.

• Clustered the data of various stores based on annual spending amounts of different product categories.
• Used different clustering algorithms like K-means, Agglomerative to find optimum number of clusters.