Monday, November 10, 2025 - 2:00pm to 3:30pm ET
The short course focused on a virtual short course on Data-Driven Problem Solving and Neural Computing, featuring presentations and workshops led by Padmanabhan from George Mason University. This short course covered essential data science concepts, including the three Vs of data, neural networks, and various data analysis tools like Excel, Python, and no-code platforms. Throughout the sessions, Padmanabhan emphasized practical applications of data science and AI in education, demonstrated various data analysis techniques, and highlighted the importance of developing students' data competencies and creative problem-solving skills.
Data Science Workforce Trends
The meeting began with NISS welcoming participants to a virtual short course on Data-Driven Problem Solving and Neural Computing, featuring Padmanabhan from George Mason University. Padmanabhan discussed the growing importance of data-related jobs, highlighting that Big Data Specialists were reported as the fastest-growing job in 2025, followed by FinTech and AI experts. He emphasized the need for a deeper understanding of data science and statistics, noting that the workforce values skills such as creative thinking, resilience, and analytical abilities, while also urging educators to focus on practical applications of algorithms rather than just theoretical knowledge.
Data Analytics and Neural Networks
Padmanabhan discussed the three Vs of data: velocity, variety, and volume, emphasizing their importance in understanding data flow. He introduced a workshop on using neural networks in Excel and Python, highlighting a paper published in the Bulletin of Mathematical Biology that demonstrates these techniques. Padmanabhan also mentioned a second paper on teaching Python through literate programming, a concept introduced by Donald Knuth. He concluded by introducing a game to engage participants and encourage interaction.
Data Science in Education Evolution
Padmanabhan led an interactive workshop on data science, focusing on the importance of data competencies and the integration of AI in education. He discussed the evolution of data science curricula in Virginia schools, emphasizing the need to bridge gaps between statistics, mathematics, and computer science. Padmanabhan also highlighted the significance of machine learning components, including training, learning, testing, and predicting, while explaining the role of computational thinking in AI development.
Kodak: Data Analysis Tool Introduction
Padmanabhan introduced Kodak (Common Online Data Analysis Platform) as a tool for data analysis, demonstrating how to access and explore datasets, including one on roller coasters. He showed how to visualize data using graphs and discussed methods for analyzing and interpreting data, emphasizing the importance of addressing missing values and understanding data patterns. Padmanabhan also highlighted the tool's interactive features and its suitability for teaching data science to students with diverse interests.
Statistical Tools for Data Visualization
Padmanabhan demonstrated various statistical tools and data visualization techniques, including measures of central tendency, spread, and outliers, using a dataset of roller coaster statistics. He showed how to create scatterplots to explore correlations between variables like top speed and maximum height, and introduced the concept of linear regression through an interactive demonstration. Padmanabhan emphasized the importance of engaging students through discovery-based learning rather than traditional lecture methods, and offered to provide workshops on teaching with technology.
Data Analysis Tools Demo
Padmanabhan demonstrated various data analysis tools, including CodeApp for no-code data visualization and Google Sheets for low-code analysis. He showed how to use CodeApp to analyze roller coaster data, create simulations, and fetch real-time weather data from NOAA. Padmanabhan also highlighted Google Sheets' capabilities, such as collaborative features, data validation, and integration with Google Translate. He mentioned that CodeApp supports machine learning and classification tools, while Google Sheets offers k-means clustering but requires more manual setup.
KNIME, Excel, and Python Overview
Padmanabhan introduced KNIME, a no-code machine learning tool, and demonstrated its capabilities for data analysis and visualization. He also covered basic Excel techniques such as pivot tables and conditional formatting. The session concluded with an introduction to Python programming for data analysis, specifically focusing on the Pandas library for working with datasets.
Python for Data Analysis and ML
Padmanabhan demonstrated how to perform exploratory data analysis (EDA) using Python, including calculating means, standard deviations, and correlations. He showed how to import and visualize data using libraries like Pandas and Matplotlib, and explained the concept of Monte Carlo simulations. Padmanabhan also introduced the basics of neural networks and physics-informed neural networks, demonstrating how to implement them using Excel and Python. He emphasized the importance of teaching creative problem-solving and statistical thinking in education.
Machine Learning Implementation Overview
Padmanabhan provided an overview of machine learning and neural network solutions, explaining how to implement them using Python and various datasets. He demonstrated regression techniques and physics-informed neural networks, emphasizing the importance of incorporating physics into the loss function. Padmanabhan also introduced no-code platforms like KNIME and encouraged participants to explore different problem-solving frameworks. He concluded by sharing his experience with applying machine learning to real-world problems, such as monitoring drug addiction and preventing suicide.
Thanks & Acknowledgements
The National Institute of Statistical Sciences (NISS) extends its sincere thanks to Dr. Padmanabhan of George Mason University for leading an engaging and insightful virtual short course on Data-Driven Problem Solving and Neural Computing. We also gratefully acknowledge all participants for their active involvement, thoughtful questions, and contributions to the interactive discussions and workshops. Special appreciation goes to the NISS staff and organizers for their coordination and support in delivering a seamless virtual learning experience. Together, these efforts made the short course a valuable opportunity to explore practical applications of data science, machine learning, and AI in education and beyond.