Data-Driven Annotation of Textual Process Descriptions Based on Formal Meaning Representations

Semantic parsing is more popular than ever. One reason is that we have a rising number of semantically annotated corpora. Another reason is that there is new AI technology to be explored. In this talk I will present a new corpus of open-domain texts annotated with formal meaning representations. Using a parallel corpus, the resource is developed not only for English, but also for Dutch, German and Italian. The meaning representations comprise logical operators to assign scope, comparison operators, and non-logical symbols. The non-logical symbols are completely grounded in WordNet concepts and VerbNet-style roles. I will contrast two methods for semantic parsing on this corpus: a traditional technique using a categorial grammar and lambda-calculus, and an ultra-modern way using a (surprise, surprise) neural network. Guess which one performs better!

This paper is on control-flow extraction which is a part of workflow-extraction. Workflow extraction is the transformation of a textual process description into a formal workflow model. A textual process description is a form of representation for procedural knowledge. Procedural knowledge is know how to do it-knowledge. Workflow extraction opens the use of existing workflow retrieval methods for procedural knowledge which is expressed in natural language. In this paper we extended our existing workflow-extraction-framework with components for control-flow extraction. We present and evaluate two different approaches for control-flow extraction. For the evaluation of the control-flow we developed special evaluation functions which are based on the trace index of workflows. We show that our approaches for control-flow extraction are beneficiai compared with a sequential control-flow.

Large Language Models (LLMs) are emerging as powerful tools for automating business process modeling, promising to streamline the translation of textual process descriptions into Business Process Model and Notation (BPMN) diagrams. However, the extent to which these Al systems can produce high-quality BPMN models has not yet been rigorously evaluated. This paper presents an early evaluation of five LLM-powered BPMN generation tools that automatically convert textual process descriptions into BPMN models. To assess the external quality of these Al-generated models, we introduce a novel structured evaluation framework that scores each BPMN diagram across three key process model quality dimensions: clarity, correctness, and completeness, covering both accuracy and diagram understandability. Using this framework, we conducted experiments where each tool was tasked with modeling the same set of textual process scenarios, and the resulting diagrams were systematically scored based on the criteria. This approach provides a consistent and repeatable evaluation procedure and offers a new lens for comparing LLM-based modeling capabilities. Given the focused scope of the study, the results should be interpreted as an exploratory benchmark that surfaces initial observations about tool performance rather than definitive conclusions. Our findings reveal that while current LLM-based tools can produce BPMN diagrams that capture the main elements of a process description, they often exhibit errors such as missing steps, inconsistent logic, or modeling rule violations, highlighting limitations in achieving fully correct and complete models. The clarity and readability of the generated diagrams also vary, indicating that these Al models are still maturing in generating easily interpretable process flows. We conclude that although LLMs show promise in automating BPMN modeling, significant improvements are needed for them to consistently generate both syntactically and semantically valid process models.

LightGCAN: A lightweight graph convolutional attention network for user preference modeling and personalized recommendation

Automatic building extraction from optical imagery remains a challenge due to, for example, the complexity of building shapes. Semantic segmentation is an efficient approach for this task. The latest development in deep convolutional neural networks (DCNNs) has made accurate pixel-level classification tasks possible. Yet one central issue remains: the precise delineation of boundaries. Deep architectures generally fail to produce fine-grained segmentation with accurate boundaries due to their progressive down-sampling. Hence, we introduce a generic framework to overcome the issue, integrating the graph convolutional network (GCN) and deep structured feature embedding (DSFE) into an end-to-end workflow. Furthermore, instead of using a classic graph convolutional neural network, we propose a gated graph convolutional network, which enables the refinement of weak and coarse semantic predictions to generate sharp borders and fine-grained pixel-level classification. Taking the semantic segmentation of building footprints as a practical example, we compared different feature embedding architectures and graph neural networks. Our proposed framework with the new GCN architecture outperforms state-of-the-art approaches. Although our main task in this work is building footprint extraction, the proposed method can be generally applied to other binary or multi-label segmentation tasks.

Cross-subject emotion recognition in brain-computer interface based on frequency band attention graph convolutional adversarial neural networks

Predicting DNA sequence splice site based on graph convolutional network and DNA graph construction

Video-based human action recognition is one of the most important and challenging areas of research in the field of computer vision. Human action recognition has found many pragmatic applications in video surveillance, human-computer interaction, entertainment, autonomous driving, etc. Owing to the recent development of deep learning methods for human action recognition, the performance of action recognition has significantly enhanced for challenging datasets. Deep learning techniques are mainly used for recognizing actions in images and videos comprising of Euclidean data. A recent development in deep learning methods is the extension of these techniques to non-Euclidean data or graph data with many nodes and edges. Human body skeleton resembles a graph, therefore, the graph convolutional network (GCN) is applicable to the non-Euclidean body skeleton. In the past few years, GCN has emerged as an important tool for skeleton-based action recognition. Therefore, we conduct a survey using GCN methods for action recognition. Herein, we present a comprehensive overview of recent GCN techniques for action recognition, propose a taxonomy for the categorization of GCN techniques for action recognition, carry out a detailed study of the benchmark datasets, enlist relevant resources and open-source codes, and finally provide an outline for future research directions and trends. To the best of authors' knowledge, this is the first survey for action recognition using GCN techniques.

In this paper, we propose a model using a graph convolution network for predicting the accuracy of the automatically generated convolution neural network (CNN). In recent years, there have been many excellent results in the automatic designing of deep learning network structures and their hyperparameters. These works mainly use evolutionary algorithms and reinforcement learning methods. But to verify the accuracy of newly generated neural network samples, it requires a lot of computational time and large scale of hardware. Therefore, how to reduce training time and improve the utilization of samples has become an important problem of automatic design for deep learning networks. On the other hand, the effectiveness of the graph convolution network (GCN) has been shown in processing graph data on the tasks of graph node classification, link prediction, and clustering. We are considering using GCN to predict the accuracy of the automatically designed neural networks. We compare GCN, a multi-layer neural network(NN), and CNN on the accuracy prediction task of automatically generated CNN for image classification tasks on the CIFAR-10 dataset. Our experiment demonstrates the advantages of GCN on small datasets compared to the multi-layer neural network and CNN, while GCN achieves the equivalent accuracy to the other methods on large datasets.

Geometric deep learning provides a principled and universal way for the integration of imaging and non-imaging modes in the medical field. Graph Convolutional Networks (GCNs) have been widely explored in a variety of problems, such as disease prediction, segmentation, and matrix completion. Using large, multi-modal data sets, graphs can capture the interaction of individual elements represented as nodes in the graphs. In particular, in medical applications, nodes can represent individuals (patients or healthy controls) in a potentially large population and are accompanied by a set of features, while the edges of the graph contain the associations between subjects in an intuitive way. This representation allows the inclusion of rich imaging and non-imaging information as well as individual subject characteristics in the task of disease classification. This article gives an overview of graph convolutional neural networks, focusing on the application of graph convolutional neural networks in disease prediction, and discusses the challenges and prospects of graph convolutional neural networks in disease prediction.

This paper presents an investigation of an artificial neural network (ANN) to predict the flutter characteristics of a simple wing that under went multiple weight variation changes. The weight variations were used to simulate the addition of non-structural weight or structural repair weight. The test case wing had 24 designated repair areas where repair weights were added as lumped masses. Weight and location of the repairs were randomly generated to increase the sample size and help the generalization of the ANN. The ANN was evaluated for both the flutter speed and flutter frequency for multiple levels of structural damping. The number of test cases required to train the neural network was also evaluated for both the error between the input and output results as well as the generalization of the neural networks.

The chapter presents a description of a novel modelling language called Alvis defined for the design of concurrent especially real-time systems. Alvis combines the advantages of formal methods and practical modelling languages. Based on CCS and XCCS process algebras, Alvis utilizes flexible graphical modelling of interconnections among agents and a high level programming language used for the description of agents behaviour. Each agent in a model can use rule-based systems to support its decisions. A small set of language statements and graphical concepts make Alvis easy to learn and use. The possibility of a formal model verification, makes Alvis a formal modelling language. Alvis modelling environment creates in parallel a model of the considered system and a labelled transition system (LTS graph) that is its formal representation. The LTS graph can be formally verified with the CADP toolbox. A survey of main Alvis features from practical point of view, is given in the chapter.

The Sixth International Symposium on Neural Networks (ISNN 2009)

We investigate the problem of training neural networks from incomplete images without replacing missing values. For this purpose, we first represent an image as a graph, in which missing pixels are entirely ignored. The graph image representation is processed using a spatial graph convolutional network (SGCN) – a type of graph convolutional networks, which is a proper generalization of classical CNNs operating on images. On one hand, our approach avoids the problem of missing data imputation while, on the other hand, there is a natural correspondence between CNNs and SGCN. Experiments confirm that our approach performs better than analogical CNNs with the imputation of missing values on typical classification and reconstruction tasks.

A point process is a collection of positional variables in a given space, while a marked point process (MPP) is basically a point process that measures some additional feature or value at each point. Existing applications such as financial market models and hospitalization models have demonstrated that MPPS are suitable for modeling random events with meaningful values. A big challenge is to study the relationship between different types of events, or how different types of markers affect each other. While Hawkes procedures are an effective and efficient way to model point processes with internal influences, inferential methods of intensity functions can be expensive, especially for multi-dimensional processes. Graph convolutional Network (GCN) is a powerful neural network designed to directly process graphs. A GCN-based model is presented to study potential patterns in the marking process and to predict the location of future events without estimating the Hawkes intensity function. A real data set of water pipe leakage and rupture records over the last 50 years was used in the experiment, modeled as a marker point process, where leakage and rupture are two marker forms. Experimental results on synthetic data sets and real water pipe data sets show that the proposed model is superior to recent state-of-the-art methods.