Szabolcs Rozsnyai

Event-driven rules for sensing and responding to business situations

Event-based systems have been developed and used to implement networked and adaptive business environments based on loosely coupled systems in order to respond faster to critical business events. In this paper, we introduce a rule management system which is able to sense and evaluate events in order to respond to changes in a business environment or customer needs. It enables users to graphically compose comprehensive event-triggered rules, which can be used to control the processing of services. For the definition of a rule set, users can independently define event conditions, event patterns and correlation-related information which can be combined for modeling complex business situations. We have fully implemented the proposed system with a service-oriented approach and illustrate our approach with an order management business case.

Concepts and models for typing events for event-based systems

Event-based systems are increasingly gaining widespread attention for applications that require integration with loosely coupled and distributed systems for time-critical business solutions. In this paper, we show concepts and models for representing, structuring and typing events. We discuss existing event models in the field and introduce the event model of the event-based system SARI for illustrating various typing concepts. The typing concepts cover topics such as type inheritance and exheritance, dynamic type inferencing, attribute types, as well as the extendibility and addressability of events. We show how the typing concepts evolved and depend on the implemented event-based systems which use different approaches for the event processing such as graphical approaches, or approaches, that use Java code, SQL code, or ECA (event-condition-action) rules.

Event Cloud - Searching for Correlated Business Events

Market players that can respond to critical business events faster than their competitors will end up as winners in the fast moving economy. Event-based systems have been developed and used to implement networked and adaptive business environments based on loosely coupled systems. In this paper, we introduce Event Cloud, a system that allows searching for business events in a variety of contexts that also take the relationships between events into consideration. Event Cloud supports knowledge workers in their daily operations in order to perform investigations and analyses based on historical events. It enables users to search in large sets of historical events which are correlated and indexed in a data staging process with an easy-to-use search interface. For improving the search results, we propose an index based ranking system. We present an architecture for the Event Cloud system, which supports a continuous near real-time integration of business events with the aim of decreasing the time it takes to make them available for searching purposes. We have fully implemented the proposed architecture and discuss implementation details.

Investigating clinical care pathways correlated with outcomes

Clinical care pathway analysis is the process of discovering how clinical activities impact patients in their care journeys, and uses the discovered knowledge for various applications including the redesign and optimization of clinical pathways. We present an approach for mining clinical care pathways correlated with patient outcomes that involves a combination of clustering, process mining and frequent pattern mining. Our approach is implemented as a set of interactive tools in the business process insight (BPI) platform, a a collaborative software as a service platform, that provides an event-driven process-aware analytics toolset. After interactively utilizing the individual clustering, process mining, and frequent pattern mining capabilities in BPI, users can overlay frequent patterns, ranked according to their correlation with a particular patient outcome, on a mined model of the patient population with that outcome. We have tested our approach for mining care pathways correlated with outcomes on electronic medical record data obtained from a US based healthcare provider on congestive heart failure (CHF) patients. Experimental results show that the tools we have developed and implemented can provide new insights to facilitate the improvement of existing clinical care pathways.

Discovering event correlation rules for semi-structured business processes

In this paper we describe an algorithm to discover event correlation rules from arbitrary data sources. Correlation rules can be useful for determining relationships between events in order to isolate instances of a running business process for the purposes of monitoring, discovery and other applications. We have implemented our algorithm and validate our approach on events generated by a simulator that implements a real-world inspired export compliance regulations scenario consisting of 24 activities and corresponding event types. This simulated scenario involves a wide range of heterogeneous systems (e.g. Order Management, Document Management, E-Mail, and Export Violation Detection Services) as well as workflow-supported human-driven interactions (Process Management System). Experimental results demonstrate that our algorithm achieves a high level of accuracy in the detection of correlation rules. This paper confirms that our algorithm is a step towards semi-automating the task of detecting correlations. We also demonstrate how correlation rules discovered by our algorithm can be used to create aggregation nodes that allow more efficient querying, filtering and analytics. The results in this paper encourage future directions such as distributed statistics calculation, and scalability in terms of handling massive data sets.

Large-Scale Distributed Storage System for Business Provenance

In todays complex business environment, applications span across loosely coupled systems generating massive amounts of business artifacts at various levels of granularity. Monitoring and analyzing these artifacts enables access to critical process information to improve the effectiveness of business operations. Tracking, capturing, storing and processing such large volumes of data, however, is difficult and resource intensive with current relational database technologies. Hence, designers are forced to make trade-offs in deciding the type and the granularity level of the data to be captured. Nevertheless, the amount of historical data that carries important insight about the business processes that need to be captured is growing. A solution that is capable of handling massive business provenance data is necessary. In this paper, using cloud as opposed to relational databases to manage this massive amount of business provenance data is proposed and a cloud-based business provenance architecture based on HBase/Hadoop technology is introduced.

Proactive Business Process Compliance Monitoring with Event-Based Systems

Business processes spanning across organizational boundaries inside and outside an enterprise are increasingly becoming common practice in todays networked business environments. Service level agreements (SLAs) are negotiated between enterprises to measure, ensure and enforce service fulfillment and quality in this dynamic context. Often, SLA violations are directly associated with penalty costs, making it crucial to stick to agreed SLAs and proactively intervene in case of potential violations. Thus, a framework is required which allows for (1) efficient business process compliance monitoring, and (2) taking immediate action in case of compliance violations in order to minimize the business impact. In this paper we present a novel compliance monitoring framework based on a Complex Event Processing (CEP) engine. It allows modeling business processes as event flows, whereby events reflect state changes in a process or the business environment. Compliance checkpoints are added to an event flow and signify aspects which may be relevant to monitor, such as the relative timeframe between two events. Upon these, monitoring rules are defined to detect compliance violations and automatically trigger corrective actions.

Event data warehousing for Complex Event Processing

In the last few years, Complex Event Processing (CEP) has emerged as a new paradigm for event-driven applications. The research focus in this area has so far been primarily on operational issues and not on the ex-post analysis of event data. On the other side, approaches like data warehousing have proven useful in the past to extract further valuable information from the data available within an organization. In this paper, we elaborate the concept of a fully-implemented event data warehouse as an add-on for CEP that allows to efficiently archive and query valuable event data for later analysis. We outline the overall architecture and describe the relevant meta-models for an integrated data management approach. The data management itself is implemented using a RDBMS, and its schema is automatically synchronized with the CEP models. Finally, we present a real-world use case to illustrate the application of the event data warehouse in practice.

SARI-SQL: Event Query Language for Event Analysis

Complex Event Processing (CEP) systems are capable of processing large amounts of events, utilizing them to monitor, steer and optimize business in real time. The lack of tracking events and maintaining the causal relationships and traceability between those events, as well as aggregating them to higher-level events, is a problem that is currently investigated by many research groups. In this paper, we present SARI-SQL, which is a domain-specific event-query language, (EQL) that is designed for business analysts to easily gain insight into business events. SARI-SQL enables the retrieval of near real-time events and can process historical events, metrics and scores for analytical purposes. We introduce the SARI-SQL syntax and show infrastructural components for the query engine. We further show examples to illustrate the query language, and propose a reference implementation for the query engine.

Semantic Event Correlation Using Ontologies

Complex event processing (CEP) is a software architecture paradigm that aims at low latency, high throughput, and quick adaptability of applications for supporting and improving event-driven business processes. Events sensed in real time are the basic information units on which CEP applications operate and react in self-contained decision cycles based on defined processing logic and rules. Event correlation is necessary to relate events gathered from various sources for detecting patterns and situations of interest in the business context. Unfortunately, event correlation has been limited to syntactically identical attribute values instead of addressing semantically equivalent attribute meanings. Semantic equivalence is particularly relevant if events come from organizations that use different terminologies for common concepts. n nIn this paper, we introduce an approach that uses semantic technologies, in our case ontologies, for the definition of event correlations to facilitate semantic event correlation derived from semantic equivalence, inherited meaning, and relationships between different terms or entities. We evaluate the practical application of three types of semantic correlation based on use cases that are relevant to the real-world domain of industrial production automation. Major results of the evaluation show that semantic correlation enables functions for CEP that traditional syntactic correlation does not allow at all.