Hello
I’m Nehal
Varshney

Language Translator:  
An application designed to automatically translate text inscribed from image to english language to another. The project typically utilized natural language processing and machine learning techniques to analyze and understand the input text, and then generate a translation in the target language with an accuracy of 87%

Sentiment Analysis on Instagram and YouTube Comments Web app:
Created to take in comments from Instagram posts and YouTube videos and specify the sentiment of comments, to help a user to analyze how their content is being perceived by the audience. The accuracy of this model was 81%.

Optimized Object Detection:
Using Multiprocessing Optimized object detection on images and videos using multiple processors, accelerating CPU time from 18 mins to 19 milliseconds.

Cab Pooling:
Motivation to build this project was to solve the issue faced by students at my own university. It is a web app that helps you find a cab pooling partner according to the destination entered by the user.

E-FIR:
Under the assistance of IIT Kanpur, I created an android app. With the help of this app the user can register his/her complaint under the various categories according to the type of crime incurred. This app helps in saving time and one additional feature is that in emergency the user can send the present location to the nearest police station.

Safety Kleen (Clean Harbor) Capstone:
Spearheaded a data-driven pricing optimization for Safety-Kleen, using customer sales data for price elasticity, product trends, geography, and customer churn insights, elevating pricing and marketing strategies. 

Boston Crime Analysis:
Harnessed Python’s machine learning and deep learning prowess, and achieved an outstanding 92% accuracy in Boston crime detection, driven by meticulous exploratory data analysis (EDA) for enhanced public safety.