Garbage-In Garbage Out (GIGO) is the most widely used maxim in machine learning, but how does one assess the quality of each step in an analysis pipeline? This course teaches students how to understand their data, models, and pipelines using visualization and computational skepticism, empowering them to create robust, trustworthy AI systems.
Throughout the semester, students move from high-level philosophy to building agentic AI tools that implement principles of computational skepticism to validate data, models, and analytical pipelines.
- AI Fluency & Botspeak: Master the Nine Pillars of AI Fluency (Strategic Delegation, Critical Evaluation, Stochastic Reasoning, etc.) to optimize human-AI collaboration.
- Computational Skepticism: Develop systematic doubt in AI workflows using adversarial validation and critical thinking frameworks.
- Visual Understanding & Design: Learn to understand statistical properties visually, choose the right chart for the right question, and apply advanced visual design principles like rhythm, hierarchy, and unity.
- Generative AI for Data: Explore the creation and evaluation of high-quality synthetic data across text, numeric, image, audio, and video modalities.
- Causal Inference: Go beyond correlation by using Evidence Knowledge Graphs (EKG), Directed Acyclic Graphs (DAGs), and counterfactual frameworks to identify true causal relationships.
Syllabus/- Full course syllabus, schedule, and grading policies.Assignments/- Weekly agentic AI tool builds and project submissions.Exams/- Exam materials and study guides.
Botspeak/- Implementations of the Nine Pillars of AI Fluency.Data_Visualization/- Visual design principles and exploratory data analysis.Causal_Inference/- DAGs, counterfactuals, and observational study methodologies.Forecasting/- Time series analysis and predictive modeling.
GIGO_A_Crash_Course_in_Data/- Fundamentals of data cleaning, imputation (KNN/MICE), and dimensionality reduction.Bear/- Data Science AI Cheatsheet covering bias-variance tradeoffs and ML metrics.Notebooks/- Comprehensive Jupyter notebooks organized by course modules (Generative AI, Statistical Methods, Preprocessing).
- Clone this repository:
git clone https://github.com/nikbearbrown/INFO_7390_Art_and_Science_of_Data.git
cd INFO_7390_Art_and_Science_of_Data
- Switch to the organized branch:
git checkout organized_branch
- Course Instructor: Nik Bear Brown (ni.brown@neu.edu)
- YouTube Lessons: Bite-sized lectures available on the nikbearbrown channel.
Students are expected to follow Northeastern University's academic integrity policies. All AI usage (ChatGPT, Claude, etc.) must be cited, specifying which portions of the work were AI-assisted and describing how the tool was used.