AI for Database Education
Echelon: An AI Tool for Clustering Student-Written SQL Queries

My ultimate goal is to use AI to improve teaching and learning. To that end, we developed AI-based system (Echelon) that help instructors quickly identify trends in students' solutions.  Echelon is capable of extracting features that instructors deem significant from students' SQL queries and using them to generate clusters that capture the critical approaches taken. The system creates a two-dimensional projection (see figure below), which is then linked to a dashboard that instructors can use to rapidly assess class performance. 

Overview-v2 (1)
Echelon User Interface
Echelon's UI: (a) shows the Aggregate Mode with multiple clusters of similar queries (b) shows an excerpt of the plot part of the UI on the same question using the Point Mode

Improving Active Learning with Echelon

AI Tool for Clustering Student-Written
SQL Querie

Instructors can use Echelon to rapidly assess class performance, using clustering algorithms to group student approaches into clean, intuitive categories. They can then address a variety of student approaches, and thus create a more responsive classroom.

Example Use Case

As part of the flipped-classroom model, CS 411 students work on in-class collaborative learning assessments. At the end of an in-class SQL group activity, a database instructor can upload all SQL submissions to a problem and use Echelon to show students clusters of common (correct and incorrect) approaches to  problem.

As shown in the figure below, by clicking on different clusters, echelon provides sample queries from the selected cluster, allowing the instructor to provide essential feedback that can help student avoid common mistakes and over-complicated queries.