Natural event summarization

Abstract

Event mining is a useful way to understand computer system behaviors. The focus of recent works on event mining has been shifted to event summarization from discovering frequent patterns. Event summarization seeks to provide a comprehensible explanation of the event sequence on certain aspects. Previous methods have several limitations such as ignoring temporal information, generating the same set of boundaries for all event patterns, and providing a summary which is difficult for human to understand.In this paper, we propose a novel framework called natural event summarization that summarizes an event sequence using inter-arrival histograms to capture the temporal relationship among events. Our framework uses the minimum description length principle to guide the process in order to balance between accuracy and brevity. Also, we use multi-resolution analysis for pruning the problem space. We demonstrate how the principles can be applied to generate summaries with periodic patterns and correlation patterns in the framework. Experimental results on synthetic and real data show our method is capable of producing usable event summary, robust to noises, and scalable.

META: Multi-resolution Framework for Event Summarization

“…Wang et al [13] further extended Kiernan's work by presenting the inter-segment relationship with HMM. Jiang et al [4] provided a richer summarization of the events by providing the event relationship network (ERN) based on the logs, which captures the temporal dynamics. To the extent of our knowledge, existing research mainly focused on developing approaches to find the optimal summarization results, while our work is to present a comprehensive and extensible framework to facilitate the multi-resolution summarization for the system analysts.…”

“…Some research efforts have been working on providing various summarization methods [8,13,4,9,17]. Each of them defines its own way of summarizing events/documents.…”

Temporal Mining for Distributed Systems

“…Its implementation depends on the concrete algorithm and all the previous event summarization papers can be regarded as proposing a concrete describe operator. For example, [JPL11] summarize the events with periodic and inter-arrival relationships. The describe operation is written as Υ name (F ), where name is the name of summarization algorithm used for describing the events.…”

“…As a showcase, we leverage the algorithm proposed in [JPL11] as the summarization algorithm, which summarizes the events from the perspective of inter-arrival temporal relationship.…”

“…We also attach corresponding summarization results (3rd row) by implementing the describe operator according to [JPL11] 2 .…”