Jon Espen Ingvaldsen

Process Mining Manifesto

Process mining techniques are able to extract knowledge from event logs commonly available in today’s information systems. These techniques provide new means to discover, monitor, and improve processes in a variety of application domains. There are two main drivers for the growing interest in process mining. On the one hand, more and more events are being recorded, thus, providing detailed information about the history of processes. On the other hand, there is a need to improve and support business processes in competitive and rapidly changing environments. This manifesto is created by the IEEE Task Force on Process Mining and aims to promote the topic of process mining. Moreover, by defining a set of guiding principles and listing important challenges, this manifesto hopes to serve as a guide for software developers, scientists, consultants, business managers, and end-users. The goal is to increase the maturity of process mining as a new tool to improve the (re)design, control, and support of operational business processes.

Industrial application of semantic process mining

Process mining relates to the extraction of non-trivial and useful information from information system event logs. It is a new research discipline that has evolved significantly since the early work on idealistic process logs. Over the last years, process mining prototypes have incorporated elements from semantics and data mining and targeted visualisation techniques that are more user-friendly to business experts and process owners. In this article, we present a framework for evaluating different aspects of enterprise process flows and address practical challenges of state-of-the-art industrial process mining. We also explore the inherent strengths of the technology for more efficient process optimisation.

Model-Based Business Process Mining

Abstract As companies use enterprise resource planning (ERP) systems to support their business processes, they need to verify that the systems are configured appropriately and used in the most efficient way. This article describes the approach taken and results from a business process mining project at a midsized company in Norway. A newly released tool for analyzing ERP system logs is used to construct the underlying business flows and to provide new insights that can be used by the company to improve the procurement process.

Ontology learning for search applications

Ontology learning tools help us build ontologies cheaper by applying sophisticated linguistic and statistical techniques on domain text. For ontologies used in search applications class concepts and hierarchical relationships at the appropriate level of detail are vital to the quality of retrieval. In this paper, we discuss an unsupervised keyphrase extraction system for ontology learning and evaluate its resulting ontology as part of an ontology-driven search application. Our analysis shows that even though the ontology is slightly inferior to manually constructed ontologies, the quality of search is only marginally affected when using the learned ontology. Keyphrase extraction may not be sufficient for ontology learning in general, but is surprisingly effective for ontologies specifically designed for search.

Financial News Mining: Monitoring Continuous Streams of Text

This paper addresses the problem of extracting, analyzing and synthesizing valuable information from continuous text streams covering financial information. A text mining framework combining elements from information retrieval, information extraction and natural language processing has been implemented. The framework is utilized to extract information regarding key actors in the domain, how they relate to each other, and how these characteristics evolve over time

A text mining approach to integrating business process models and governing documents

As large companies are building up their enterprise architecture solutions, they need to relate business process descriptions to lengthy and formally structured documents of corporate policies and standards. However, these documents are usually not specific to particular tasks or processes, and the user is left to read through a substantial amount of irrelevant text to find the few fragments that are relevant to him. In this paper, we describe a text mining approach to establishing links between business process model elements and relevant parts of governing documents in Statoil, one of Norway’s largest companies. The approach builds on standard IR techniques, gives us a ranked list of text fragments for each business process activity, and can easily be integrated with Statoil’s enterprise architecture solution. With these ranked lists at hand, users can easily find the most relevant sections to read before carrying out their activities.

Unsupervised keyphrase extraction for search ontologies

Ontology learning today ranges from simple frequency counting methods to advanced linguistic analyses of sentence structure and word semantics. For ontologies in information retrieval systems, class concepts and hierarchical relationships at the appropriate level of detail are crucial to the quality of retrieval. In this paper, we present an unsupervised keyphrase extraction system and evaluate its ability to support the construction of ontologies for search applications. In spite of its limitations, such a system is well suited to constantly changing domains and captures some interesting domain features that are important in search ontologies. The approach is evaluated against the project management documentation of a Norwegian petroleum company.

Taming news streams with linked data

The use of smartphones and tablets has increased significantly in the past years and changed the way we consume news. In this paper, we will describe a news stream aggregating system that automatically recognize and disambiguate geo spatial and meaning bearing entities in news text. The system utilizes the entity definitions and associations in WikiData and Geonames to build a semantic representation of the underlying news text. Its mobile user interface allows users to explore news story clusters on an interactive map and retrieve local news streams for any region of the world.

Interactive mobile news recommender system: A preliminary study of usability factors

Interactive news recommender systems allow the user to steer the received recommendations in the desired directions through explicit interaction with the system. It provides a user experience in between a “lean back and let the news wash over me” experience and an “active search and hunt for specific pieces” experience. On the other hand, this added level of interaction might also be perceived as extra burden from the user side and therefore experience a decreased level of user experience. This paper describes a user study which uncovers factors that influence the usability of interactive news recommender system. The user study is carried out by contrasting an experimental system where interaction is granted with a baseline system where interaction is absent. The study demonstrated that test participants find the ability to actively shape its news recommendation strategy a useful and desirable feature. In addition, time, location and user interest as dimensions for interaction seems reasonable. Lastly, it identifies three factors that are of particular importance when designing interactive news recommender systems.

Modeling the Dynamics of Online News Reading Interests

Online news readers exhibit a very dynamic behavior. News publishers have been investigating ways to predict such changes in order to adjust their recommendation strategies and better engage the readers. Existing research focuses on analyzing the evolution of reading interests associated with news categories. Compared to these, we study also how relations among news interests change in time. Observations over a 10-month period on a German news publisher indicate that overall, the relations amid news categories change, but stable periods spanning months are also found. The reasons of these changes and how news publishers could integrate this knowledge in their solutions are subject to further investigation.