Representation Methodology Research Articles

Explainable next POI recommendation based on spatial–temporal disentanglement representation and pseudo profile generation

The current research in Point-of-Interest (POI) recommendation primarily aims to decipher users’ transitional patterns to predict their future location visits. Traditional approaches often intertwine various features to model these check-in transitions, which inadvertently compromises the quality of the resulting representations. This issue is compounded in both industrial and academic settings, where user-generated textual data is frequently inaccessible or restricted due to privacy concerns. Such limitations in user profiles pose significant challenges to the effectiveness of subsequent applications. In response to these challenges, the recent rise of Large Language Models (LLMs) offers a novel perspective. Diverging from the conventional approach of leveraging LLMs for semantic-based next check-in predictions, our research investigates the potential of integrating LLMs with sequential recommendation systems. This integration aims to augment feature dimensions and facilitate the generation of explicit explanations. To this end, we introduce CrossDR-Gen, a Cross-sequence Location Disentanglement Representation methodology. CrossDR-Gen is specifically designed for next POI recommendation and explanation generation. It uniquely considers spatial and temporal factors in shaping check-in behaviors, offering a comprehensive global view of location transitions. Crucially, CrossDR-Gen utilizes LLMs for pseudo profile generation in scenarios with limited semantic context, thereby enriching user features without relying on additional textual profiles or conversational data. Our experiments on real-world datasets demonstrate that CrossDR-Gen not only excels in addressing cold-start scenarios but also showcases robust recommendation capabilities. These findings validate the effectiveness of our proposed cooperative paradigm between LLMs and sequential recommendation models, highlighting a promising avenue for future research in POI recommendation systems.

Graph‐Based Representation Approach for Deep Learning of Organic Light‐Emitting Diode Devices

The performance prediction of organic light‐emitting diode (OLED) devices using artificial intelligence is significantly limited due to the lack of representational feature data. This study proposes a novel graph‐based representation methodology to effectively address these challenges. Various graph convolution methods are explored, resulting in an ideal representation of the device parameters in the static equilibrium state, which is crucial for accurate modeling. This representation not only exhibits parameter‐like characteristics but also encapsulates essential physical meanings that enhance interpretability. Additionally, the trained predictive model demonstrates relatively high accuracy, making it a reliable tool for practical applications. Finally, this research serves as a valuable initial study for predicting and designing OLED devices, paving the way for future advancements in the field.

Advancing complex urban traffic forecasting: A fully attentional spatial-temporal network enhanced by graph representation

Accurate urban traffic forecasting is essential for intelligent transportation systems (ITS). However, the majority of existing forecasting methodologies predominantly concentrate on point-based forecasts (e.g., traffic detector forecasts). A limited number of them pay attention to the urban bidirectional road segments and the complex road network topology. To advance accurate traffic forecasting in complex urban scenarios, this paper proposes a Graph Representation enhanced Fully Attentional Spatial-Temporal network (GR-FAST). First, we construct a refined bidirectional road network graph (BRG) to depict the urban road network topology more accurately, particularly focusing on the turning patterns at intersections. Then, we adopt the graph representation methodology and introduce spatial information encoding (SIE) to explicitly characterize the significance of roads and network structure from multiple perspectives. Enhanced by SIE, spatial attention can capture spatial dependencies from both road network topologies and traffic pattern similarities, thereby forming a unified urban spatial cognition. Finally, a multi-scale residual perception (MRP) module is designed to balance the interplay of short-term temporal variability and long-term periodicity. Experiments on a real-world urban dataset from Wuhan, China, demonstrate that GR-FAST outperforms the state-of-the-art deep learning methods, achieving an improvement of 9.19%. Furthermore, ablation studies suggest that the explicit incorporation of complex road spatial topologies can significantly enhance forecasting accuracy.

The Watersheds Speak: The Voice of Ecosystems in Northern New York's Environmental Movements☆

AbstractThis paper explores how connectivity to place has brought life to contemporary environmental struggles in what is now known as New York State. Layers of memory, colonization, and stewardship are embedded within a community's relationship with their environment. By focusing on two case studies, the authors will illustrate how this relationship shaped successful place‐based resistance. Throughout our work, we offer the alternative methodology of ecocentric storytelling and artistic representation to elevate the voice of both ecosystems and all their inhabitants. Both case studies are centered on environmental/social movements that foreground the inherent personhood of the natural world. The first will focus on the Ashokan Reservoir, in the Catskill Mountains, on land guarded by the Lenape. It is a location that has suffered multiple waves of colonization and successfully resisted a large damming operation. The second case study will be centered on the Upper St. Lawrence River/Kaniatarowanénhne Watershed, Haudenosaunee Territory, where rural communities are uniting to grant Rights to the Rivers that bring life to their communities. Both cases exemplify the story of communities that defended their livelihoods and environments by uniting with their most reliable allies, the Water, and all their more‐than‐human guardians.

A novel enhancing method for terahertz imaging of integrated circuits flaw detection

This research paper addresses the current gap in understanding the intricate damage types within packaged integrated circuits (ICs). We introduce a novel and multiscale latent low-rank representation (M-LatLRR) methodology, meticulously tailored to enhance the quality of terahertz (THz) IC images and reveal latent features. By integrating THz imaging with our proposed M-LatLRR method, we aim to facilitate the precise identification of damage types within inside the packaged ICs. Firstly, the multiscale Gaussian functions are used to remove the blur and the components are obtained via averaging. Secondly, the image matrices are utilized to extract base matrices and detail matrices. With LatLRR, the multiscale detail matrices are extracted at several representation levels. The final enhanced image is reconstructed by average strategy for dealing with the detail and pre-enhancement parts. The M-LatLRR framework is universal and can be effectively applied to extract multi-level features of packaged IC images. In a comparative analysis, our method demonstrates superior capabilities in determining the failure types of fractured wire bonds, cracks, and delamination of the dielectric layer, outperforming alternative methodologies.

A review of graph neural network applications in mechanics-related domains

Mechanics-related tasks often present unique challenges in achieving accurate geometric and physical representations, particularly for non-uniform structures. Graph neural networks (GNNs) have emerged as a promising tool to tackle these challenges by adeptly learning from graph data with irregular underlying structures. Consequently, recent years have witnessed a surge in complex mechanics-related applications inspired by the advancements of GNNs. Despite this process, there is a notable absence of a systematic review addressing the recent advancement of GNNs in solving mechanics-related tasks. To bridge this gap, this review article aims to provide an in-depth overview of the GNN applications in mechanics-related domains while identifying key challenges and outlining potential future research directions. In this review article, we begin by introducing the fundamental algorithms of GNNs that are widely employed in mechanics-related applications. We provide a concise explanation of their underlying principles to establish a solid understanding that will serve as a basis for exploring the applications of GNNs in mechanics-related domains. The scope of this paper is intended to cover the categorisation of literature into solid mechanics, fluid mechanics, and interdisciplinary mechanics-related domains, providing a comprehensive summary of graph representation methodologies, GNN architectures, and further discussions in their respective subdomains. Additionally, open data and source codes relevant to these applications are summarised for the convenience of future researchers. This article promotes an interdisciplinary integration of GNNs and mechanics and provides a guide for researchers interested in applying GNNs to solve complex mechanics-related tasks.

A framework for fully automated reconstruction of semantic building model at urban-scale using textured LoD2 data

The CityGML Level of Detail 3 (LoD3), a widely adopted standard for three-dimensional (3D) city modeling, has been accessible for an extended period. However, its comprehensive implementation remains limited due to challenges such as insufficient automation and inconsistent data quality. This research introduces an innovative and fully automated framework aimed at urban-scale semantic building model reconstruction. The proposed framework addresses three critical challenges: (1) proposing facade layout graph model to formalize the geometry and topological relationships of semantic entities on building facades, thereby promoting the deduction of structural completeness and the reconstruction of semantic facade models; (2) establishing a mapping relationship between texture images, semantic entities, and building shells guided by the facade layout graph to ensure consistent correlations among the geometry, semantics, and topology of building models; (3) developing an efficient representation methodology for semantic building models utilizing a parameter set derived from the facade layout graph. The proposed framework has been successfully validated by reconstructing 8,681 buildings from three different locations in Berlin. The results demonstrate an outstanding reconstruction accuracy of 91%, with a time efficiency of only 3.42 s per building. Visual analysis further confirms that the framework effectively fulfills the application prerequisites of 3D GIS. The code of the proposed framework is available in the repository: https://github.com/wangyuefeng2017/LoD3Framework-.

Importance of general systems theory for scientific training

The article investigates the possibility of formation of creative competencies of students within the framework of scientific personnel training. It is noted that the available training courses within the framework of general training of scientific personnel allow to prepare a specialist who is able to use existing knowledge, but such a specialist is not trained in creativity. For the formation of creative competences it is necessary to take a training course of the general theory of systems, the purpose of which is to create a universal methodology of reliable representation of models of objects, based on the similarity (isomorphism) of the forms and laws of the universe. A detailed argumentation of the connection between creativity and holistic representation of being, the empirical manifestation of which is isomorphism, is given. The mechanism of knowledge translation within the framework of the mechanism of multisystem integration of human being, the mechanism of knowledge borrowing by human being from sensually comprehended forms of reality is investigated. It is stated that in order to meet the set tasks of cognition, a version of the general theory of systems, extended by the metaphysical component, is required. Further, starting from the metaphysical description of the foundations of the phenomenon of isomorphism observed in the practice of cognition, it is possible to form complete collections of patterns – typical forms, templates of objects and their relations. At present, the work on such a version of the general theory of systems is being finalized. The article concludes with a discussion of the problematics of teaching general systems theory. It emphasizes two components: the issue of positioning the training course "General Theory of Systems" in relation to other courses, as well as taking into account the specifics of the subject area of students in different specialties (directions).

Visual reconstructions of endometriosis pain: An interdisciplinary visual methodology for illness representation.

Endometriosis is a chronic condition in which tissue resembling the endometrium grows outside of the womb, causing severe chronic pain. Research demonstrates the physical, emotional and quality of life impact on people with endometriosis, but pain is reportedly difficult to communicate, resulting in lengthier diagnosis. This work aimed to gain insight into the value of imagery production as a pain communication strategy through a novel synergy of psychological and linguistic/socio-semiotic approaches. A qualitative, multimodal, participant-generated imagery study. Interpretative phenomenological analysis (IPA) and conceptual metaphor and metonymy analysis were utilized to examine visual representations of endometriosis pain. Data were collected in two focus groups with four and six women, respectively; all with a diagnosis of endometriosis, aged 25-40 years old (M = 34.5, SD = 4.2) and a mean diagnosis delay of 8.4 years (SD = 3.6). The overarching theme across visual representations was 'Pain as Physical Violence' with 'colour as emotional representation', 'texture as sensory qualities' and 'materials as sensation' as sub-themes. These are realized through metaphorical and metonymical relations in both the visual representations as well as the accompanying linguistic representation of the process. This study demonstrates the value of a creative mixed-methodologies approach to capture experiential aspects of pain and its impact that are not verbalized in linguistic accounts alone. This can facilitate a deeper understanding of one's pain, acting as a medium for therapeutic adjustment to occur, while facilitating effective and empathic patient-professional conversations surrounding pain.

Developing a Soil Salinity Model from Landsat 8 Satellite Bands Based on Advanced Machine Learning Algorithms

AbstractSoil salinity is a significant challenge in numerous regions across the globe, including Egypt. The potential consequences encompass negative impacts on crop yield, human well-being, and eco-logical systems. The utilization of remote sensing and machine learning techniques is increasingly becoming recognized as cost-effective methodologies for the cartographic representation of soil salinity. The present work employed Landsat 8 satellite imaging data and sophisticated machine learning techniques to delineate and assess soil salinity levels in Sharkia Governorate, Egypt. In this work, several machine learning techniques were employed to forecast the salinity values of Total Dissolved Solids (TDS) in the designated geographical region. These algorithms encompassed support vector machines (SVM), regression trees, Gaussian linear regression, and tree-based ensemble in addition to linear regression analysis. A variety of instances were generated to develop an optimal model that accurately characterizes the salinity TDS values within the study area. This was achieved by utilizing the band values extracted from the Landsat 8 satellite imagery. The approach that demonstrated the highest performance was observed when employing the Blue, Red, and shortwave infrared bands in conjunction with the SVM-Quadratic SVM model. This particular configuration yielded an R2 value of 0.86 and an RMSE value of 175.98. The findings of this work demonstrate the feasibility of precisely mapping soil salinity through the utilization of Landsat 8 satellite imaging data and machine learning techniques. The provided data can be utilized to identify regions characterized by elevated levels of soil salinity, as well as for the formulation of effective approaches aimed at addressing this issue.

Persistent homology of featured time series data and its applications

<p>Recent studies have actively employed persistent homology (PH), a topological data analysis technique, to analyze the topological information in time series data. Many successful studies have utilized graph representations of time series data for PH calculation. Given the diverse nature of time series data, it is crucial to have mechanisms that can adjust the PH calculations by incorporating domain-specific knowledge. In this context, we introduce a methodology that allows the adjustment of PH calculations by reflecting relevant domain knowledge in specific fields. We introduce the concept of featured time series, which is the pair of a time series augmented with specific features such as domain knowledge, and an influence vector that assigns a value to each feature to fine-tune the results of the PH. We then prove the stability theorem of the proposed method, which states that adjusting the influence vectors grants stability to the PH calculations. The proposed approach enables the tailored analysis of a time series based on the graph representation methodology, which makes it applicable to real-world domains. We consider two examples to verify the proposed method's advantages: anomaly detection of stock data and topological analysis of music data.</p>

Panoptic Segmentation with Convex Object Representation

Abstract The accurate representation of objects holds pivotal significance in the realm of panoptic segmentation. Presently, prevalent object representation methodologies, including box-based, keypoint-based and query-based techniques, encounter a challenge known as the ‘representation confusion’ issue in specific scenarios, often resulting in the mislabeling of instances. In response, this paper introduces Convex Object Representation (COR), a straightforward yet highly effective approach to address this problem. COR leverages a CNN-based Euclidean Distance Transform to convert the target instance into a convex heatmap. Simultaneously, it offers a parallel embedding method for encoding the object. Subsequently, COR characterizes objects based on the distinctive embedding vectors of their convex vertices. This paper seamlessly integrates COR into a state-of-the-art query-based panoptic segmentation framework. Experimental findings validate that COR successfully mitigates the representation confusion predicament, enhancing segmentation accuracy. The COR-augmented methods exhibit notable improvements of +1.3 and +0.7 points in PQ on the Cityscapes validation and MS COCO panoptic 2017 validation datasets, respectively.

Skin Imaging: A Digital Twin for Geometric Deviations on Manufactured Surfaces

Closed-loop manufacturing is crucial in Industry 4.0, since it provides an online detection–correction cycle to optimize the production line by using the live data provided from the product being manufactured. By integrating the inspection system and manufacturing processes, the production line achieves a new level of accuracy and savings on costs. This is far more crucial than only inspecting the finished product as an accepted or rejected part. Modeling the actual surface of the workpiece in production, including the manufacturing errors, enables the potential to process the provided live data and give feedback to production planning. Recently introduced “skin imaging” methodology can generate 2D images as a comprehensive digital twin for geometric deviations on any scanned 3D surface including analytical geometries and sculptured surfaces. Skin-Image has been addressed as a novel methodology for continuous representation of unorganized discrete 3D points, by which the geometric deviation on the surface is shown using image intensity. Skin-Image can be readily used in online surface inspection for automatic and precise 3D defect segmentation and characterization. It also facilitates search-guided sampling strategies. This paper presents the implementation of skin imaging for primary engineering surfaces. The results, supported by several industrial case studies, show high efficiency of skin imaging in providing models of the real manufactured surfaces.

The Issue of Selection of Appropriate Methodology and Methods of Graphical Representation of Data in Biomedical Research

Abstract The development of medicine is based on reliable medical research. This is a process that must be planned in detail and performed in accordance with the accepted study protocol, as any negligence – even a small one – or deviation from the protocol may result in distorted research results and – in consequence – to false conclusions. One of the key stages of research is the selection of the appropriate methodology, particularly in terms of tests that verify the posed hypotheses. Each statistical method has a set of assumptions that have to be met in order for the method to be used. This paper explains why the appropriate selection of methods, verification of the required assumptions, and correct graphical representation of the obtained results are so important from the practical perspective.

A VR-based workflow to assess perception of daylit views-out with a focus on dynamism and immersion

Amongst the elements often overlooked in existing studies, dynamic movement and temporal changes in the content of views-out have been suggested to have a high potential to improve its perceived quality and occupant satisfaction. Moreover, in past and ongoing view-out research, most view quality indicators and representation methodologies primarily rely on static time-independent views and fail to acknowledge the importance of including dynamic features in the viewing content. To address this research gap, the present study introduces a novel VR representation workflow to accurately capture dynamic views in experimental settings. This study is the first to examine VR’s effectiveness in maintaining immersion and dynamism when studying daylit views-out, using original data based on human subjects. The results provide insights into VR’s suitability for representing office window views and assess a new workflow employing a dual fisheye lens and scale model to depict views in physically-based IVE.

An extended stress-based forming limit diagram focusing on the wrinkling phenomenon and the effect of the normal pressure on clamped surfaces

A novel wrinkling limit representation following the pattern of the conventional forming limit curves (FLCs) and the stress-based forming limit curves (SFLCs) as well as the application of the assumed criterion in finite element modelling are discussed in this manuscript. FLCs can partially refer to the wrinkling potential in the area left to the uniaxial tension line, but just like the SFLCs, cannot characterize the limits of the material behavior in a deeper sense, if negative in-plane stress and normal pressure act together on the sheet. This study predicts the wrinkling risk of clamped surfaces with a stress-based criterion through solving the analytical equations of the critical compressive stress causing wrinkling, and the corresponding blank holder pressure. The critical values were calculated based on the Wang and Cao’s theory using Hill48 anisotropic yield function coupled with the Swift hardening law. The results draw a novel wrinkling limit curve representation methodology, in which the minor stress responsible for wrinkling and its ratio to the major stress are distinguished in the function of the applied blank holder pressure. The applicability of the calculated curves was investigated using finite element simulations, which showed that this method provides the opportunity to quantitatively interpret how close a component is to the wrinkling limit. The calculated wrinkling tendencies were verified by standard cup drawing tests supplemented by round shape error measurements on three different automotive steel sheets. It can be stated that the obtained numerical conditions of wrinkling were fitted to the experiments fairly well.

An Efficacy Analysis of Affect Representation Methodologies Used in EEG-Based Emotion Recognition System: An Unsupervised Approach

Brain-Computer Interface (BCI)-based affect recognition in humans utilizing physiological signals is typically guided by a supervised method that relies heavily on the presence of appropriate affect representations. In this paper, we assess the effectiveness of contemporary affect representation methodologies for establishing ground truth for reliable affective BCI systems. On the DEAP dataset, two experiments were conducted to compare the observed emotion clusters based on subjective ratings with the clustering of corresponding physiological responses based on EEG characteristics. The outcomes of the two investigations are inconsistent with one another. The findings suggest the subjective use of emotion categories as opposed to their general application in establishing ground truth for supervised approaches and must encourage researchers to examine affect representation techniques critically before considering them as concrete representations of a person’s affective state.

Morphological component analysis under non-convex smoothing penalty framework for gearbox fault diagnosis

The sparse representation methodology has been identified to be a promising tool for gearbox fault diagnosis. The core is how to precisely reconstruct the fault signal from noisy monitoring signals. The non-convex penalty has the ability to induce sparsity more efficiently than convex penalty. However, the introduction of non-convex penalty usually influences the convexity of the model, resulting in the unstable or sub-optimal solution. In this paper, we propose the non-convex smoothing penalty framework (NSPF) and combine it with morphological component analysis (MCA) for gearbox fault diagnosis. The proposed NSPF is a unify penalty construction framework, which contains many classical penalty while a new set of non-convex smoothing penalty functions can be generated. These non-convex penalty can guarantee the convexity of the objective function while enhancing the sparsity, thus the global optimal solution can be acquired. The simulation and engineering experiments validate that the NSPF enjoys more reconstruction precision compared to the existing penalties.

Training processes and environmental management: a contribution to education for sustainable development

The objective of this research was to assess the effects of a training process on environmental thinking of teachers at a private university. We worked with the methodology of social representations, mixed approach, in a sample of 20 teachers. Data were analyzed using frequency tables and graphs, content analysis and natural semantic networks. The results show that teachers share reduced, anthropocentric-technical and globalizing social representations. The representation field focuses on nature, its care and administration. This research has implications for teaching; it demonstrated the importance of developing training processes for an adequate approach to environmental and sustainability issues in a teaching team. A limitation of the research was found in the contingency conditions due to the COVID-19 pandemic. The study concludes that updating processes on environmental issues are a resource to enrich teachers' ideas and arouse their interest in education for sustainable development.

ANTHROPOLOGICAL PROBLEMS IN THE PHILOSOPHY OF H. S. SKOVORODA IN THE CONTEXT OF MODERN NATIONAL STATE-FORMING PROCESSES

The philosophical symbolism of H. Skovoroda’s works lies in wisdom, congenial work, seeing the big in the small, unveiling mysteries through the symbolic world. Skovoroda states that to be a human-being is to be a philosopher. The aim of philosophy is to reawaken the main mottos of the Age of Enlightenment (honor, dignity, freedom, justice, solidarity, morality). Creating open society in Ukraine on the basis of these mottos is the aim of the modern national state-building. The aim of the article is to ascertain the influence and meaning of the philosophical and anthropological problems in the works of H. S. Skovoroda for building open society in the modern Ukraine. Methodology of the study is complex and based on the social-philosophical and historic-philosophical analysis of anthropological problems in the works by H. S. Skovoroda and its significance for finding personal freedom and building open society. The leading methodological tools are fundamental research basics of the scientific analysis – principles of objectivity, systematicity, integrity, unity of the historic and the logical, determinism, development, and interrelation. Methodology of social and symbolic representations, ethnic representations, and structural functionalism has been used. Results. As Ukraine is experiencing complex social and political processes, the philosophical works by H. S. Skovoroda acquires special importance: formation and preservation of the Ukrainian national identity, national values, morality, and culture. The society needs politically-minded citizens with clear ideas about the prospects of the state development. Researching works of Skovoroda, one can trace the development of the human spirit: an internally-devastated person devoid of life reference points is gradually adapting to the life in society, overcomes the discrepancies between the singular and the whole finding a compromise in the relations with the world around them. The fates of the people and the nation, language and the state were the key problems which H. S. Skovoroda thoroughly and passionately analyzed through the perspective of universal human values. Conclusions. The modern Ukrainian society needs people in harmony with themselves and the world who work on improving themselves and the society. Such people should be able to overcome both their own internal contradictions and the conflicts in society and the world. Putting significant hope on the human-being as the creator of the new morality, H. Skovoroda believed in the possibility of building solid foundations of the state. An internally rich person and citizen is going to contribute to harmony of the nature, the world, human relations focusing on the issues of good and love as integral attributes of the human existence. The writer considered such personalities to be able to carry peace and unity to the unstable reality full of life challenges and to oppose the alienation, absurdity, and insignificance of the human existence.