Navanshu Khare

Driven Data Science Intern ready to thrive in demanding digital intelligence processing environments. Well-informed on latest machine learning advancements with experience in industry . Ready to combine tireless hunger for new skills with desire to exploit cutting-edge data science technology.

MATS : Market Basket Analysis for Targeted Sales

• Performed different association analysis and clustering techniques using observed insights to construct a personalized and efficient model for Instacart for targeted sales using Python.

Digit classification using Convolution Neural Networks

• Designed an Digit classification system using single convolution layer with a small filter size (3,3) and a modest number of filters followed by max pooling layer with Relu activation function and achieved accuracy of 87.6%.

Credit Card Fraud Detection using machine learning models

• Developed an Random forest model which detect fraudulent behavior in credit card transactions and Data Balancing was used to eliminate over fitting.
• PUBLISHED in IJPAM (SCOPUS), Presented at International Conference on Research Development in Engineering Science & Technology-2018 ISSN 1314-339

Terrorism in India: Integrated Approach Using Machine Learning models And Data Mining Analysis

• Developed a predictive model which will determine terrorist group responsible for an attack and an eclectic analysis of data to discover a pattern among attacks | Accuracy : 94.6%
• Performed Ensemble approach with algorithms like random forests, neural network,
• PUBLISHED in International Journal of Scientific Research in Engineering and Management (IJSREM) ISSN: 2582-3930

Sentiment Analysis of Twitter

· Analyzed and Visualized the tweets based on positive or negative expression of thought by using the Lexicon Based approach in R language.