Nikita Kumari

Nikita Kumari

Intern

Followers of Nikita Kumari568 followers
location of Nikita KumariDelhi, India

Connect with Nikita Kumari to Send Message

Connect

Connect with Nikita Kumari to Send Message

Connect
  • Timeline

  • About me

    Data analyst intern @adani digital lab | Ex-intern @Ministry of statistics| SQL | PowerBI | ML and Python Skills | Advance Ms Excel | Tableau | Hackathons | Traveller

  • Education

    • Bhai parmanand DSEU shakarpur campus -II

      2021 - 2024
      Bachelor's degree Data analytics
    • School of Excellence

      2020 - 2021
      12th PCMB 87%
  • Experience

    • Ministry of Statistics and Programme Implementation

      Jul 2023 - Aug 2023
      Intern

      Data Analyst Intern• Data Preprocessing: Successfully tackled the challenge of preprocessing large and scattered 5 different government datasets (IIP, NAS, PLFS, CPI) spanning multiple years, ensuring data consistency for analysis.• Visualization Selection: Effectively determined the optimal 5 visualization types (bar charts, pie charts, line charts, tree maps, sunburst charts) to represent complex data relationships and trends for each dataset.• Created 300+ insightful visualizations, providing valuable insights Show less

    • Adani Digital Labs

      Feb 2024 - now
      Data analyst intern

      Sales Data Analysis and Frequent Item set Matrix Creation• Analyzed sales data from 12 Lakh transactions using Python and Excel to identify frequently bought together products.• Developed a Python-based solution to build an n*n matrix of product co-occurrences of Category based on count and sales of 12L transaction and 3k unique items.• Optimized data extraction process by implementing efficient join and Functions building strategies on large datasets in Python, reducing processing time by 20%.Recommendation System with Collaborative Filtering and Price Constraints• Designed and implemented a recommendation system in Python using Pandas and collaborative filtering to suggest the top 5 complementary items based on purchase history of 6L unique customers.• Leveraged frequent item set analysis results to ensure recommended 5 products belong to 12 different categories, enhancing purchase diversity.• Incorporated price constraints into the recommendation algorithm using Python functions, ensuring suggested products meet minimum value thresholds for targeted upselling.• Optimized the recommendation system by replacing 10 loops with vectorized Pandas operations, significantly improving performance of model by 30%. Show less

  • Licenses & Certifications