Multi-Dimensional Regression Analysis of Time-Series Data Streams* Export

@misc{citeulike:3225158, abstract = {Real-time production systems and other dynamic environments often generate tremendous (potentially infinite) amount of stream data; the volume of data is too huge to be stored on disks or scanned multiple times. Can we perform on-line, multi-dimensional analysis and data mining of such data to alert people about dramatic changes of situations and to initiate timely, high-quality responses? This is a challenging task. In this paper, we investigate methods for online, multi-dimensional regression analysis of time-series stream data, with the following contributions: (1) our analysis shows that only a small number of compressed regression measures instead of the complete stream of data need to be registered for multi-dimensional linear regression analysis, (2) to facilitate on-line stream data analysis, a partially materialized data cube model, with regression as measure, and a tilt time frame as its time dimension, is proposed to minimize the amount of data to be retained in memory or stored on disks, and (3) an exception-guided drilling approach is developed for on-line, multi-dimensional exceptionbased regression analysis. Based on this design, algorithms are proposed for efficient analysis of time-series data streams. Our performance study compares the proposed algorithms and identifies the most memory- and time- efficient}, author = {Yixin and Dong, Guozhu and Han, Jiawei and Benjamin, I. and Wah, W. and Jianyong}, citeulike-article-id = {3225158}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.19.8295}, keywords = {mining, streams}, posted-at = {2008-09-11 13:21:53}, priority = {2}, title = {Multi-Dimensional Regression Analysis of Time-Series Data Streams*}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.19.8295} }

TY - ELEC ID - citeulike:3225158 L3 - citeulike-article-id:3225158 N2 - Real-time production systems and other dynamic environments often generate tremendous (potentially infinite) amount of stream data; the volume of data is too huge to be stored on disks or scanned multiple times. Can we perform on-line, multi-dimensional analysis and data mining of such data to alert people about dramatic changes of situations and to initiate timely, high-quality responses? This is a challenging task. In this paper, we investigate methods for online, multi-dimensional regression analysis of time-series stream data, with the following contributions: (1) our analysis shows that only a small number of compressed regression measures instead of the complete stream of data need to be registered for multi-dimensional linear regression analysis, (2) to facilitate on-line stream data analysis, a partially materialized data cube model, with regression as measure, and a tilt time frame as its time dimension, is proposed to minimize the amount of data to be retained in memory or stored on disks, and (3) an exception-guided drilling approach is developed for on-line, multi-dimensional exceptionbased regression analysis. Based on this design, algorithms are proposed for efficient analysis of time-series data streams. Our performance study compares the proposed algorithms and identifies the most memory- and time- efficient TI - Multi-Dimensional Regression Analysis of Time-Series Data Streams* KW - mining KW - streams AU - Yixin AU - Dong, Guozhu AU - Han, Jiawei AU - Benjamin, I AU - Wah, W AU - Jianyong UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.19.8295 ER -