Aditya Sharma — Data Scientist

Growing up, I enjoyed Maths and Science and participated in various olympiads in middle school. I came across Data Science during my college years, which sparked my curiosity. So I started doing some online courses, and predictive analytics became my interest. In my last semester of college, I got an internship at Playpower. I was fortunate enough to dive deep into real-world data. I learned a lot from our CEO, Chief Data Scientist, and clients during my internship and then as a Data Scientist about framing research questions, practicing different statistical modeling techniques, designing visualizations for storytelling, and working on different projects and datasets. Over the past 6 years, I’ve worked as a Data Science consultant for one of the biggest publishing companies, helping them understand student and teacher usage behaviors, understanding what improves student outcomes, and sharing insights to improve product designs and make data-driven decisions. I also got the opportunity to work on different types of data across projects over the years like event data; student question responses; skill and standard data; school and district demographic data; web analytics data; educational games data; process data; knowledge structure data; support calls data; order and licensing data; roster data; server load data; psychometric data. Research and Awards: - I have been part of 7 peer-reviewed research items - Wrote an R package called seqClustR to perform sequence clustering on sequential data - Awarded Best Short Paper for ‘"Curriculum Pacing: A New Approach to Discover Instructional Practices in Classrooms" at the 14th International Conference on Intelligent Tutoring Systems - 2nd Place Winner in “NAEP (National Assessment of Educational Progress) Educational Data Mining Competition 2019” - Winner of “The 2022 EM:IP Cover Graphic/Data Visualization Competition” and the visualization will feature on the cover of one of the upcoming issues of 'Educational Measurement: Issues and Practice'. My interest areas in data science and learning analytics include statistical modeling, user behavior analysis, storytelling through visualizations, equity in education, machine learning, and feature engineering. You can contact me directly at adityasharma9352@gmail.com

Stackforce AI infers this person is a Data Science expert in Educational Technology with strong analytical skills.

Location: Kota, Rajasthan, India

Experience: 9 yrs

Skills

Data Science
Data Visualization
Predictive Analytics
Statistical Modeling
Learning Analytics

Career Highlights

Developed award-winning data visualizations.
Achieved high accuracy in predictive modeling.
Contributed to multiple peer-reviewed research items.

Work Experience

Playpower Labs

Senior Data Scientist - Learning / Product Analytics (6 yrs 9 mos)

Data Scientist - Learning / Product Analytics (1 yr 11 mos)

Data Science Intern - Learning / Product Analytics (4 mos)

Education

Post-Graduate Diploma at Indian Statistical Institute, Kolkata

Bachelor of Technology - BTech at Nirma University

at Bakhshi's Springdales Sr Sec School

at Saint Paul's Sr. Sec. School

Aditya Sharma

Data Scientist

Kota, Rajasthan, India9 yrs experience

Most Likely To SwitchAI Enabled

Key Highlights

Developed award-winning data visualizations.
Achieved high accuracy in predictive modeling.
Contributed to multiple peer-reviewed research items.

Stackforce AI infers this person is a Data Science expert in Educational Technology with strong analytical skills.

Contact

adityasharma9352@gmail.com LinkedIn

Skills

Core Skills

Data ScienceData VisualizationPredictive AnalyticsStatistical ModelingLearning Analytics

Other Skills

AnalyticsBusiness AnalyticsCC++CSSData AnalyticsData MiningData StorytellingEducational TechnologyExploratory Data AnalysisHTMLMachine LearningMicrosoft ExcelMySQLOptimization

About

Experience

Playpower labs

3 roles

Senior Data Scientist - Learning / Product Analytics

Promoted

Jun 2019 – Present · 6 yrs 9 mos

Created an automated reporting system to generate PDF reports in R (Programming Language) using child markdowns for a cognitive tests product with psychometric data. Coordinated with developers and product designers and trained a data scientist
Performed data quality checks, cleaned the data for the reporting system, and created a data logging guideline for the developers
Designed and coded multiple data dashboards in R shiny (dashboard tool) and SQL at the platform, product, and district levels for product teams, leadership, and district admins
Designed ‘EdOptimize’ an open-source learning analytics platform
Designed and coded 2 email campaigns to share custom reports with the districts through emails
Optimized the email campaigns to create reports for 500+ districts in less than 20 minutes
Won 2nd place in the NAEP Educational Data Mining Competition 2019 out of 89 teams, predicting if a student exhibits ineffective test-taking behavior
Developed a framework to automate parameterized reports at a large scale in an R&D environment for flexibility using R (Programming Language), SQL, Google Sheets, googlesheets4 package, and officer package
Directed interns and fellow Data Scientists to answer research questions
Designed and coded dashboards in R shiny (dashboard tool) for internal products to support data-driven product design changes
Collaborated with different teams and aggregated 5 data sources to prepare a regression model to predict maximum server load during COVID where the outcome was within 95% Confidence Interval of the predictions
Wrote an R package ‘seqClustR’ to perform sequence clustering on process data
Won “The 2022 EM:IP Cover Graphic/Data Visualization Competition”. The visualization will feature on the cover of 'Educational Measurement: Issues and Practice'

Data AnalyticsProject ManagementData StorytellingSQLProblem SolvingData Mining+4

Data Scientist - Learning / Product Analytics

Jun 2017 – May 2019 · 1 yr 11 mos

Conducted exploratory & explanatory analysis using R (Programming Language) and SQL, & shared the insights with districts and product teams using reports (EDA in R using packages dplyr, magrittr, ggplot2, lubridate, etc)
Supported districts to improve student outcome, implementation, and improve curriculum design
Shared actionable data-driven product insights with the product teams and the leadership
Performed statistical modeling to infer which product components affect student’s performance on high stake tests
Performed multi-level regression modeling to identify class effects on student outcomes on high stake tests
Designed and coded data dashboards using R shiny and SQL for different products, districts, and internal research work
Linked user accounts between Google Classroom and an LMS, based on their parent account’s information and their personal information using fuzzyjoin package
Performed data audit of a product using exploratory and explanatory quantitative analysis, which showed the product’s inefficacy
Mapped orders and licenses in an old system by aggregating multiple data sources to prevent revenue leakage
Used factor analysis and topic modeling to analyze support call data to categorize support call issues
Added school and district demographic data to the usage data by combining multiple data sources
Performed data regression analysis to identify the relationship between school poverty and students’ math performance
Designed award-winning informational new visualizations in R using ggplot2
Generated metadata for products to provide detailed data insights
Won best short paper award for the paper ‘Curriculum Pacing: A New Approach to Discover Instructional Practices in Classrooms’
Prepared a content management tool as a POC using R shiny for the product team to understand how effective the content is with the help of content-level stats, skill-level stats, item-level stats, and knowledge graphs

Data AnalyticsSQLData MiningExploratory Data AnalysisRData Visualization+4

Data Science Intern - Learning / Product Analytics

Jan 2017 – May 2017 · 4 mos

Learned to extract data using SQL, perform exploratory data analysis in R (Programming Language), apply different statistical models on data, and communicate the insights discovered by creating a story using visualizations.
Worked on a class churn machine learning prediction model to predict if a class would drop out after 2 weeks of initial usage on an online learning platform. The accuracy of the model was 78%.