Exploratory sequential data analysis (ESDA) complements data mining and statistical data analysis with visualisation and manipulation techniques of stream data aimed at revealing patterns, similarities and causality relationships, hereby helping researchers to generate hypotheses on the information collected during experiments or out of the lab.

We identify the lack of a generic ESDA application as the reason for which this type of analysis is seldom used, and often invented anew, in current research projects in the computer-human interaction field. That is why we present DASE, an exploratory data analysis framework, that aims to be re-usable in the context of numerous projects. This framework is based on a stream abstraction layer and a very expressive dynamic stream algebra. We use DASE in the context of a simple timeline based stream viewer, and provide initial proof that DASE can be used for ESDA to improve the analysis of stream data.

DASE is an exploratory sequential data analysis framework based on Python and SQLite that implements an advanced stream manipulation algebra.