View on Snap!Con
Presented By: Yasin Silva, Heather Flynn, Humberto Luiz Razente, Michael Barden
Abstract:
The ability to retrieve data from a data store and perform core operations such as filtering, merging, and aggregating tasks, is becoming a critical skill in a data driven world where Data Science is becoming a fundamental interdisciplinary field.
In this presentation we will describe and show the key features of DBSnap, a web application to build database queries (based on relational algebra operators) by snapping blocks. An important characteristic of DBSnap, which was built adapting existing Snap! modules, is that it uses a tree-based structure to represent a database query. This query structure has been extensively used by educators and in many textbooks as an intuitive way to describe the operators and datasets that are used in a given query and the way the operators interact with each other. DBSnap is also a highly dynamic tool that shows intermediate results of a query as the user adds more data and operator blocks. It also enables the inspection of the intermediate results associated with any node in the query tree. In this presentation we will (1) present the core components of DBSnap to create a query, (2) demonstrate the construction of queries using common operators such as selection, projection, join and grouping, and (3) highlight some recently implemented features such as the support of views and the ability to save and load queries. DBSnap, developed at Arizona State University, is a publicly available tool and aims to have the same transformational effect on database learning as other block-based systems had on traditional programming learning.
Additional information about DBSnap can be found at: DBSnap
Presenters: Yasin Silva, Humberto Razente, Michael Barden, Heather Flynn