Chandra Prakash Jagrat
Download
Recommend
Messages
Share
C

Chandra Prakash Jagrat

About

Detail

Data Scientist at Newton School
Bengaluru, Karnataka, India
Contact Chandra regarding: 
work
Full-time jobs

Timeline

work
Job
school
Education

Résumé

Jobs verified_user 0% verified
  • Newton school
    Data Scientist
    Newton school
    Sep 2023 - Current (2 years 4 months)
    • Improved prediction accuracy of the User360 project by implementing Random-Forest algorithms, resulting in an 85% accuracy rate. Utilized SHAP and LIME for interpretability, enabling the model to serve as a valuable second opinion for the Admissions team, thereby enhancing decision-making processes by 25%. • Developed and deployed multiple dashboards for the Academics team using Google Looker Studio and Metabase API, facilitating comprehensive monitoring of key metrics such as Batch Health, NPS, TA and Mentor Sessions, Placements, Project Submission, resulting in a 20% increase in data accessibility and decision-making efficiency. • Enhanced the quality of the Data Science course by integrating real-world projects and use cases of Deep
Education verified_user 0% verified
  • M S Ramaiah Institute of Technology
    Bachelor of Engineering
    M S Ramaiah Institute of Technology
    Jan 2017 - Jan 2021 (4 years 1 month)
  • A
    Advanced
Projects (professional or personal) verified_user 0% verified
  • M
    Mall Customers Segmentation
    • Conducted customer segmentation using Python and machine learning techniques, specifically K-means clustering. Gathered relevant data on customer demographics, purchasing behavior, and visit frequency. • Employed data visualization tools to visually represent the identified customer segments, making complex data more interpretable and actionable. • The customer segmentation analysis led to improved marketing strategies, resulting in enhanced customer engagement and increased sales for the mall.
  • C
    Crowd Counting
    • Developed an Edge-AI application for real-time crowd counting and analysis, focusing on public space safety and efficient crowd management. • Implemented using MATLAB, Python, Google Colab, and NVIDIA Jetson Nano; involved Convolutional Neural Networks, including Faster RCNN and PedNet. • Attained an accuracy of 90% in crowd estimation using a six-feature model, as validated on the Mall Dataset.
  • A
    Amazon Product Sentiment Analysis
    • Developed a sentiment analyzer using Python and Natural Language Processing (NLP) techniques. Additionally, created a dashboard to visualize the insights derived from the sentiment analysis. • Employed techniques such as text preprocessing, sentiment lexicons, and machine learning algorithms to achieve a high accuracy rate of 91%. • Integrated data visualization tools to effectively communicate sentiment trends and patterns to stakeholders.
Publications verified_user 0% verified
  • "
    "A Survey of Blockchain Based Government Infrastructure Information"