Name
Abstract | ||
|---|---|---|
Time series like chlorine data in large water distribution network often consist of periodic patterns, for example, the behavior of the chlorine within one day is commonly correlated to that of the next day. If the water quality patterns display a periodicity, discovering these periodicities may reveal interesting information which can be used for better future demand forecasting and decision making. Thus, the subject of this paper is to discover such periodic patterns of the overall multiple time series chlorine data in an accurate picture with time. Traditional periodic analysis techniques mainly focus on discrete symbols, which may not directly be applied to the continuous numerical values of water quality data in our work. In this paper, our core contributions are to employ a new similarity measure which requires no user parameters by using the Minimum Description Length(MDL) Principle for matching patterns in the continuous numerical values application and propose a framework to discover the periodic patterns in the large water distribution network. Furthermore, we evaluate our approaches on a real water distribution network from the Battle of the Water Sensor Network (BWSN). Experiment results show that our periodic pattern discovering methods are effective and can discover interesting periodic time-evolving patterns on the chlorine data. |
Year | DOI | Venue |
|---|---|---|
2010 | 10.1109/ICEBE.2010.21 | ICEBE |
Keywords | Field | DocType |
demand forecasting,minimum description length principle,representative pattern,water quality,summarization,decision making,pattern classification,pattern matching,water quality patterns,periodic pattern,series chlorine data,periodic time evolving pattern,mdl,matching pattern,water distribution network,chlorine data,periodic analysis techniques,data mining,continuous numerical values application,water sensor network,time series,water supply,water quality data,periodic patterns discovery,algorithm design and analysis,minimum description length,erbium,time series analysis,water resources,sensor network | Time series,Data mining,Algorithm design,Similarity measure,Demand forecasting,Computer science,Minimum description length,Pattern matching,Wireless sensor network,Periodic graph (geometry) | Conference |
Volume | Issue | ISSN |
null | null | null |
ISBN | Citations | PageRank |
978-0-7695-4227-0 | 0 | 0.34 |
References | Authors | |
5 | 5 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Hongmei Xiao | 1 | 9 | 1.60 |
| Xiuli Ma | 2 | 92 | 15.47 |
| Xiao-hui Lin | 3 | 203 | 39.35 |
| Shiwei Tang | 4 | 478 | 51.52 |
| Chunhua Tian | 5 | 74 | 16.63 |