StreamFitter: a real time linear regression analysis system for continuous data streams

  • Authors:

  • Chandima Hewa Nadungodage;Yuni Xia;Fang Li;Jaehwan John Lee;Jiaqi Ge

  • Affiliations:

  • Department of Computer & Information Science, Indiana University-Purdue University Indianapolis;Department of Computer & Information Science, Indiana University-Purdue University Indianapolis;Department of Mathematical Sciences, Indiana University-Purdue University, Indianapolis;Department of Electrical & Computer Engineering, Indiana University-Purdue University, Indianapolis;Department of Computer & Information Science, Indiana University-Purdue University Indianapolis

  • Venue:
  • DASFAA'11 Proceedings of the 16th international conference on Database systems for advanced applications: Part II
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this demo, we present the StreamFitter system for real-time linear regression analysis on continuous data streams. In order to perform regression on data streams, it is necessary to continuously update the regression model while receiving new data. In this demo, we will present two approaches for on-line, multi-dimensional linear regression analysis of stream data, namely Incremental Mathematical Stream Regression (IMSR) and Approximate Stream Regression (ASR). These methods dynamically recompute the regression model, considering not only the data records of the current window, but also the synopsis of the previous data. Therefore, the refined parameters more accurately model the entire data stream. The demo will show that the proposed methods are not only efficient in time and space, but also generate better fitted regression functions compared to the traditional sliding window methods and well-adapted to data changes.