Search

Monitoring abnormal patterns in data streams is an important research area for many applications. In this paper we present a new approach MAPS(Monitoring Abnormal Patterns over data Streams) to model and identify the abnormal patterns over the massive data streams. Compared with other data streams, ICU streaming data have their own features: pseudo-periodicity and polymorphism. MAPS first extracts patterns from the online arriving data streams and then normalizes them according to their pseudo-periodic semantics. Abnormal patterns will be detected if they are satisfied the predicates defined in the clinician-specifying normal patterns. At last, a real application demonstrates that MAPS is efficient and effective in several important aspects.

This demo paper describes a clinical information system: Data Stream Engine based Clinical information system (DSEC). In DSEC, data stream technology as well as traditional computer mechanism is used to process medical data and improve the quality of service in hospitals. The novel features of DSEC include: 1) A complete data stream processing and querying architecture for medical application 2) being able to detect data changes in a simple and effective way; 3) a load shedding mechanism used to avoid system crash when high data rate occurs.

Detecting changes in data streams is very important for many applications. This paper presents a hybrid method for detecting data stream changes in intensive care unit. In the method, we first use query processing to detect all the potential changes supporting semantics in big granularity, and then perform similarity matching, which has some features such as normalized subsequences and weighted distance. Our approach makes change detection with a better trade-off between sensitivity and specificity. Experiments on ICU data streams demonstrate its effectiveness.

The load shedding problem is vital to a Data Stream Management System (DSMS). This paper presents the design, implementation, and evaluation of a load shedding method under the guide of the feedback control theory, in order to solve practical problems in medical environment. Thus, the using of operator selectivity, which has been proven not stable enough, is avoided. This paper focuses on the restriction of memory resource, this prevents the overflow of both CUP and memory resource. Our method can well support ad-hoc queries, while it is not so in a DSMS using current load shedding method because of the instability of operator selectivity. Our method also ensures a higher query precision when the system is over loaded and is easy to be implemented. The analytical and experimental results show that our method can be applied to medical data stream systems efficiently.

This paper proposes a general consistency management methodology within ontology-driven WIS development. The contributions of this paper are follows: 1) introduces the concept structures for representing model ontologies, mappings ontology and their instances. These ontologies and their instances greatly enhance the representation and reasoning capabilities of mapping and also simplify the manipulation 2) Defines and implements a series of general operators to manipulate mapping and model ontologies. 3) Formalizes the consistency management problem and gives an algorithm basing on general operators. 4) Implements a prototype and integrates it with WISE.