An Exploratory Study and Application of Data Mining: Railway Alarm Data
The railway industry generates large data but there are few researches on railway data analysis. The paper presented an exploratory study and application of data mining from railway alarm data. The railway alarm data is analyzed to find the correlation between alarm items and between railway bureaus when alarm occurred and predict the alarm occurring. The paper proposed an alternative measurement mode with three values: support, Kulc and balance to mine the correlation from alarm data analysis, and the results finally indicated the very possibility of associated railway bureaus.
KeywordsData mining Association rules Railway alarm data
This work was supported by National Key Research and Development Plan of China (2016YFB0502604, 2016YFC0803000), International Scientific and Technological Cooperation and Academic Exchange Program of Beijing Institute of Technology (GZ2016085103), Frontier and interdisciplinary innovation program of Beijing Institute of Technology (2016CX11006).