%0 Dataset
%T 
%J National Cryosphere Desert Data Center
%I National Cryosphere Desert Data Center(www.ncdc.ac.cn)
%U http://www.ncdc.ac.cn/portal/metadata/12ca6b85-f2eb-424c-a9ad-b85634277bbd
%W NCDC
%A mo yuan qiu
%K Event sequence;causal structure learning;root-cause localization
%X This dataset contains sequence data of alarm events generated by cellular communication networks. Each alarm record contains three elements: alarm event type number, alarm occurrence timestamp, and network element node number that generated the alarm. Compared with event sequence data generated by similar simulations, this dataset comes from real communication network operation and maintenance scenarios, with real and complex characteristics such as alarm storms, cascading propagation, and topological dependencies. It is mainly used for inferring causal relationships of event types, modeling network topological dependencies, root cause alarm localization, and analyzing alarm cascading propagation. It can provide a real data foundation for intelligent operation and maintenance of large-scale communication networks and research on fault root cause localization.