Pineapple

SENTHILKUMAR SEKAR

Data Scientist
senthilkumar2ai@gmail.com
Mob: 4379892028

DATA SCIENCE PROJECTS PORTFOLIO

A Display of the Machine Learning Models developed as part of my Data Scientist Journey.
Click on the Title of the project to view teh complete code, analysis, model and findings.

Similar Products Search Engine

NLP Use Case solved with BOW Vector, Tf-Idf and Cosine Similarity. ML Based Search engine that fetches Similar products from the 180K product inventory data scraped from Amazon.

Digital Ad Click Predictor

A Click Predictor Model built on 4 different Classification algorithms and the accuracies compared. The click predictor is an important ML Feature for Digital Ad Exchanges to spend the budget wisely.

Kyphosis Predictor

NLP Use Case solved with BOW Vector, Tf-Idf and Cosine Similarity. ML Based Search engine that fetches Similar products from the 180K product inventory data scraped from Amazon.

Clustering Of Drivers Data

A K Means Clustering Model developed to cluster the Drivers' data into segments based on input features. Elbow Method implemented to decided optimum segments number.

Titanic Survivor Predictor

A Predictor Model developed to point the prediction probability of a traveller on the Titanic Ship. Data Cleaning of Null Values implemented using an identified Pattern.

Spam SMS Detector

A Natural Language Processing Use Case where Naive Bayes CLassifier Model is used to classify the Spam SMS from teh Real ones. Tf-Idf Word Embedding is implemented.

House Price Estimator

A Supervised Regression Model that calculates the price of Housing units based on Real Estate Data including many input features and the price of the sample housing units.

Data Science Skill Set

Programming languages : Python, R
Libraries : Python - Sklearn, seaborn, numpy, pandas. R - ggplot2, caTools

Algorithms & Statistics : K Means Clustering, KNN, Decision Trees, Random Forests, Linear Regression, Logistic Regression, Naive Bayes, Cosine similarity, Pairwise Distances, CountVector(BOW), Tf-Idf

Data Science Solutions developed for : Clustering, Classification, Forecastig, Text Classification, Text Similarity Score, Image Similarity, Chat Bot, Voice To Text Bot Assistant, OCR, Social Media Analytics kind of problem statements