Highlights - A framework for designing process querying methods is proposed

 - The framework is positioned for broader Process Analytics and Business Intelligence

 - The framework is grounded in use cases from the Business Process Management field

 - The framework is informed by and validated via a systematic literature review

 - The framework structures the state of the art and points to gaps in existing research

 Abstract 

The volume of process-related data is growing rapidly: more and more business operations are being supported and monitored by information systems. Industry 4.0 and the corresponding industrial Internet of Things are about to generate new waves of process-related data, next to the abundance of event data already present in enterprise systems. However, organizations often fail to convert such data into strategic and tactical intelligence. This is due to the lack of dedicated technologies that are tailored to effectively manage the information on processes encoded in process models and process execution records. Process-related information is a core organizational asset which requires dedicated analytics to unlock its full potential. This paper proposes a framework for devising process querying methods, i.e., techniques for the (automated) management of repositories of designed and executed processes, as well as models that describe relationships between processes. The framework is composed of generic components that can be configured to create a range of process querying methods. The motivation for the framework stems from use cases in the field of Business Process Management. The design of the framework is informed by and validated via a systematic literature review. The framework structures the state of the art and points to gaps in existing research. Process querying methods need to address these gaps to better support strategic decision-making and provide the next generation of Business Intelligence platforms.

Process querying: Enabling business intelligence through query-based process analytics

This paper explores the structure of research papers in software engineering. Using text mining, we study 35,391 software engineering (SE) papers from 34 leading SE venues over the last 25 years. These venues were divided, nearly evenly, between conferences and journals. An important aspect of this analysis is that it is fully automated and repeatable. To achieve that automation, we used topic modeling (with LDA) to mine 10 topics that represent much of the structure of contemporary SE. The 10 topics presented here should not be "set in stone" as the only topics worthy of study in SE. Rather our goal is to report that (a) text mining methods can detect large scale trends within our community; (b) those topic change with time; so (c) it is important to have automatic agents that can update our understanding of our community whenever new data arrives.

Finding Trends in Software Research

We address the problem of offline handwritten diagram recognition. Recently, it has been shown that diagram symbols can be directly recognized with deep learning object detectors. However, object detectors are not able to recognize the diagram structure. We propose Arrow R-CNN, the first deep learning system for joint symbol and structure recognition in handwritten diagrams. Arrow R-CNN extends the Faster R-CNN object detector with an arrow head and tail keypoint predictor and a diagram-aware postprocessing method. We propose a network architecture and data augmentation methods targeted at small diagram datasets. Our diagram-aware postprocessing method addresses the insufficiencies of standard Faster R-CNN postprocessing. It reconstructs a diagram from a set of symbol detections and arrow keypoints. Arrow R-CNN improves state-of-the-art substantially: on a scanned flowchart dataset, we increase the rate of recognized diagrams from 37.7 to 78.6%.

/pdf/arrow-r-cnn-for-handwritten-diagram-recognition-2sm3csdkc4.pdf

Arrow R-CNN for handwritten diagram recognition

Despite the widespread availability of process modeling tools, the first version of a process model is often drawn by hand on a piece of paper or whiteboard, especially when several people are involved in its elicitation. Though this has been found to be beneficial for the modeling task itself, it also creates the need to manually convert hand-drawn models afterward, such that they can be further used in a modeling tool. This manual transformation is associated with considerable time and effort and, furthermore, creates undesirable friction in the modeling workflow. In this paper, we alleviate this problem by presenting a technique that can automatically recognize and convert a sketch process model into a digital BPMN model. A key driver and contribution of our work is the creation of a publicly available dataset consisting of 502 manually annotated, hand-drawn BPMN models, covering 25 different BPMN elements. Based on this data set, we have established a neural network-based recognition technique that can reliably recognize and transform hand-drawn BPMN models. Our evaluation shows that our technique considerably outperforms available baselines and, therefore, provides a valuable basis to smoothen the modeling process.

Sketch2BPMN: Automatic Recognition of Hand-Drawn BPMN Models

Offline hand-drawn diagram recognition is concerned with digitizing diagrams sketched on paper or whiteboard to enable further editing. Our proposed DiagramNet model addresses this recognition problem. We combine shape detection and visual arrow-relation detection to recognize arrows between shape pairs. A shape degree predictor predicts the number of in- and outgoing arrows in each direction. An optimization procedure uses the generated predictions to find the set of globally coherent arrows. Previous offline methods focus on clean images from online datasets with nicely layouted diagrams. We show that our approach is effective in the domain of camera-captured diagrams with chaotic layouts and various recognition challenges such as crossing arrows. To that end, we introduce a new dataset of hand-drawn business process diagrams that originate from textual process modeling tasks. Our evaluation shows that DiagramNet considerably outperforms prior state-of-the-art in this challenging domain.

DiagramNet: Hand-Drawn Diagram Recognition Using Visual Arrow-Relation Detection

We propose Arrow R-CNN for recognizing the symbols and structure of offline handwritten flowcharts. Arrow R-CNN extends the Faster R-CNN object detection system with an arrow keypoint predictor. This keypoint predictor is used to reconstruct the flowchart structure. We propose a network architecture and data augmentation methods that allow us to train a very deep model on a small publicly available flowchart dataset. Evaluation results show that Arrow R-CNN outperforms existing offline systems by a wide margin. For comparison with existing online flowchart recognizers, we propose an extension for mapping strokes to recognized symbols. Results show that Arrow R-CNN also achieves state of the art in online recognition, even though it does not explicitly leverage stroke information. An ablation study reveals that data augmentation guided by domain knowledge is key to high accuracy on such a small dataset.

Arrow R-CNN for Flowchart Recognition

It is understandable that sponsors of research activities are interested in assessing the work of scholars they (plan to) support although this is not a simple undertaking. Until today, there is no obvious approach for objectively measuring and comparing the quality of research results from different disciplines. Hence, counting publications has been used for long time as a substitute to deal with this challenge and only recent technological advances have fostered the usage of so-called citation indices (such as the h-index) for this purpose. Although this approach is as disputed as all previous ideas in this context, we feel it is about time to investigate the expressiveness of modern citation analysis approaches in computer science more closely. In order to do that, we have chosen the area of software engineering and created a first comprehensive ranking, illustrating citation values of world class scholars by analyzing the work of almost 700 researchers in this field. We have found that top h-index scores in software engineering are around 60 while top-notch g-indices start at around 130 when Google Scholar and Publish or Perish, the quasi standard tools for this purpose are used. Clearly, the results of our study are influenced by the coverage of these tools so that we have also analyzed Google Scholar and found it having a very high coverage of software engineering publications. Hence we are convinced to have collected good quality results that will allow our community to better judge and use citation numbers in the future.

/pdf/analyzing-citation-frequencies-of-leading-software-4lqu458zo2.pdf

Bernhard Schäfer

Papers

Arrow R-CNN for handwritten diagram recognition

Arrow R-CNN for Flowchart Recognition

Sketch2BPMN: Automatic Recognition of Hand-Drawn BPMN Models

Analyzing Citation Frequencies of Leading Software Engineering Scholars

DiagramNet: Hand-Drawn Diagram Recognition Using Visual Arrow-Relation Detection