Nowadays, most of the decisions are taken in various organizations/sectors by analyzing stakeholder’s data. This is true for the education sector also. Therefore, minimal knowledge of data analysis is mandatory at all levels in the education sector, to take proactive decisions in improving the system. Education and training are progressively taking place in digital environments. As a result, these environments are generating both structured and unstructured amount of interaction and behavioral data that can be used to design better learning and teaching models for teaching, learning and assessment. The main objective of this course is to use different kinds of methods from data analytics to identify unique patterns from educational data. In particular, the learners will learn about methods and models that are being used in data analytics, students' behavior modeling, and personalized learning material recommendations. The module will be covered both at the theoretical level as well as the practical level where software tools will be used to analyze the data.
Overview
Syllabus
Week 01: Data Analytics – An Overview (2.5 Hours)
Definition of Data Analytics and its relevance; Types of Data – Structure vs Unstructured and Quantitative vs Qualitative; Data Analytics workflow – Collection, Data Cleansing & Transformation, Data Modelling, Data Visualization; Types of Data Analytics; Data Security; Case studies.
Week 02: Clustering and Classification Techniques (2.5 Hours)
Introduction to Data Science & Methodology, Various Methods of Data Science (Clustering and Classification), Descriptive and Predictive Analytics. A Case Study of use of clustering and classification methods on educational data.
Week 03: Machine Learning for Data Science (2.5 Hours)
Introduction to Machine Learning, Neural Network and Deep Learning; A black box approach to Regression Analysis; Popular Data Analytic Tools. Case studies on educational data.
Week 04: Social Network Analysis (2.5 Hours)
Social Network Analysis in Education, A Simple Case Study of analysing Twitter/Facebook data.
Week 05: Educational Data Analytics (2.5 Hours)
Learning Associations – Classification – Regression – role of educational data analytics - Behaviour Detection - Data Synchronization - Feature Engineering - Feature Generation and Feature Selection for behaviour detection.
Week 06: Performance Factors Analysis (2.5 Hours)
Latent Knowledge Estimation - Bayesian Knowledge Tracing - Performance Factors Analysis - Relationship Mining - Correlation Mining -Students' Interaction Network Analysis.
Week 07: Data Visualization (2.5 Hours)
Visualization - Educational Visualization and Learning Curves- Heat Maps, Parameter Space Maps, State-space Network - Structure Discovery.
Week 08: Learning from Multiple Representations (2.5 Hours)
Applications of Clustering in EDA, Factor Analysis, Knowledge Inference (Qmatrix and Learning Factor Analysis) - Personalized Recommendation - Topic-based Content Recommendation - Course Recommendation. Case studies on data analytics practices by Google, Amazon, Healthcare, Government etc.
Taught by
Prof. Chandan Chakraborty
