Summary
Overview
Work History
Education
Skills
RESEARCH WORK/ PUBLICATIONS
Timeline
Generic
Divya Joshi

Divya Joshi

Data Science Consultant
New Delhi,Delhi

Summary

Analytics professional with almost 6 years of engagement in Data science and Analytics. Experienced in solving typical business problems using Data science and statistical techniques using Python and PySpark.

Specialization: Data Science, Data Analytics, Retail Data Analytics, Customer Analytics, Marketing Analytics, Rewards Analytics, Machine Learning

Overview

6
6
years of professional experience

Work History

Consultant

Fractal Analytics
07.2021 - Current

Domain: Customer Rewards Analytics(Google)

  • Target setting process: Identifying and assigning goals to each Google Partner to receive a reward on completion
  • Defining priority growth areas (PGAs): Identifying PGAs for partners in order the boost the performance
  • Incrementality testing: Analysis of Impact of each PGAs(goals) towards the overall revenue in each quarter
  • Winner/Non-Winner Analysis: Analysis of winner/non-winner Google partners in Rewards Program
  • Forecasting analysis and simulating win rate for effective Budget Utilization
  • Promotion (Kicker Reward) Impact Analysis: Difference-in-Difference Methodology
  • Customer Benchmarking: To display performance of similar companies on partners portal.
  • Challenge setup pipelines: Setting up scripts in a workflow using Google Pipelines to pull out daily performance data which feeds directly to the dashboards.

Client Handling Experience: End to end client handling experience from requirement gathering till project delivery.

  • Providing regular updates on projects
  • Sharing QA results every week on backend scripts/workflows’ health
  • Collaborated with multiple teams for end-to-end delivery of quarterly process
  • Created a good client satisfaction with quick turnaround on urgent queries and ad hoc analysis.

Applied Data Scientist

Dunnhumby
08.2019 - 04.2021

Domain: Retail Analytics

  • Target audience selection Analysis: Created PySpark script which calculates the target audiences for a campaign
  • Feasibility Analysis: Created SAS/Python scripts and have used modelling
  • Test and Control data calculation to target customers for a campaign
  • Evaluation Analysis: Evaluating campaign data and Calculating Uplift & other KPIs for retail client
  • Segmentation: Identifying Customer segmentation during Christmas/Easter/Summer
  • Churn Analysis: Analyzing customer buying behavior to identify lapsed customers
  • Covid19 Impact Analysis: Analyzed impact of COVID19 in various categories across the market
  • Derived insights at category and other segment (Life stage, Lifestyle) level
  • Campaign long term Impact analysis: Analyzed impact of a campaign over a longer period after the campaign period has end using t-test
  • Recommender system: Developed a machine learning model which provides recommendations for relevant products when primary product is out of stock or not available

Associate Consultant

Liquidhub Analytics Pvt.Ltd, A Capgemini Company
10.2017 - 08.2019
  • MS Product license Categorization: Classified Microsoft license data into their respective hierarchy
  • Involved in designing, developing and deploying ETL (SSIS) in MS SQL Server environment using in
  • Business Intelligence Development Studio (BIDS)
  • Created ETL packages to fuzzy match licensing data from Sql server
  • Techniques: SSIS, PL/SQL, Text Mining, Regex, Fuzzy Matching (Soundex, FuzzyWuzzy)

Education

M.Tech - Computer Science and Engineering

Amity University

B.Tech - Computer Science and Engineering

Uttarakhand Technical University

Skills

  • Python, PySpark
  • Techniques : scikit-learn, Pandas, Numpy, Scipy, Matplotlib, Seaborn, Statsmodel
  • Machine Learning skills: Linear/logistic regression, T-Test, Hypothesis Testing, KNN, Decision Tree, K-means/X-
  • Mean clustering, Random Forest, ANOVA/ANCOVA, Market Basket Analysis, A/B Testing
  • Google analytical tools(PLX, Pipelines, AutoTFX ,Google Data Studio), ETL, Excel
  • Databases : SQL server, Oracle, Hive
  • Google Analytical Tools : Google PLX, Pipelines, Colab, Datasite/Data studio

RESEARCH WORK/ PUBLICATIONS

  • Developed an air pollution prediction model using time series and clustering. Air pollution data is taken from Central Pollution Control Board of India, Delhi. Developed a Clustered model using K-Means and K-Nearest Algorithms to cluster the highly polluted areas in NCR and a time series prediction model to predict air pollution data based for the next year. Prediction accuracy ~ 62%.
  • Research paper titled “Challenges and Data mining model for IoT” published in International Journal of Engineering Applied Sciences and Technology, 2016.
  • Research Paper titled “Air pollution data analysis using time series clustering for IoT” published in International Journal of Control Theory and Applications, Nov 2016.

Timeline

Consultant

Fractal Analytics
07.2021 - Current

Applied Data Scientist

Dunnhumby
08.2019 - 04.2021

Associate Consultant

Liquidhub Analytics Pvt.Ltd, A Capgemini Company
10.2017 - 08.2019

M.Tech - Computer Science and Engineering

Amity University

B.Tech - Computer Science and Engineering

Uttarakhand Technical University
Divya JoshiData Science Consultant