Conceptual Clustering Research Articles

Политический нарратив: концепции, типологии и структуры

The article is devoted to the study of the narrative turn effects in humanitarian knowledge on the formation and development of the narrativity theory in relation to various types of discourse. The paper focuses on analytical and theoretical studies of the political narrative, its state of research, systematization of the concepts underlying its interpretations, and identification of the modern trends specifics in the development of political narrative. The authors managed to demonstrate the capacity of the narrative paradigm to describe the narrative nature of mentality and human activity through the prism of narrative in a socio-cultural context. Criteria have been established that make it possible to define a political narrative from the perspective of various interpretations. The theory of metanarrative is systematized and the stages of its transformation in the mirror of postmodernism and metamodern concepts are characterized. Five essential characteristics of the metanarrative are given: usefulness, ubiquity, invisibility, commitment, rigidity. The value component of metanarratives in the aspect of Russian and foreign cultures is represented by the example of basic concepts. The structure of ideologies and the family of ideologies as conceptual cluster formations are presented and described. The theories of the counternarrative are considered and the authors criteria for differentiating the alternative narrative and the counternarrative are put forward. The general structure of the narrative is demonstrated, taking into account narrative levels and narrative acts. The structure of the political narrative is described and the conditions for the successful implementation of the political narrative are identified, including the key elements of the political narrative: the main character, plot, setting, conflict, resolution. The most effective tools of the current manipulative strategy of political storytelling are analyzed – the tactics of spinning or “unwinding” the positive aspects of the message, and the techniques of using spin – cherry picking, gaslighting, non-denial denial and others. The grounds for distinguishing between fakes and factoids are presented. Classification parameters for the narrative analysis of political discourse are proposed.

Formal Logic Enabled Personalized Federated Learning through Property Inference

Recent advancements in federated learning (FL) have greatly facilitated the development of decentralized collaborative applications, particularly in the domain of Artificial Intelligence of Things (AIoT). However, a critical aspect missing from the current research landscape is the ability to enable data-driven client models with symbolic reasoning capabilities. Specifically, the inherent heterogeneity of participating client devices poses a significant challenge, as each client exhibits unique logic reasoning properties. Failing to consider these device-specific specifications can result in critical properties being missed in the client predictions, leading to suboptimal performance. In this work, we propose a new training paradigm that leverages temporal logic reasoning to address this issue. Our approach involves enhancing the training process by incorporating mechanically generated logic expressions for each FL client. Additionally, we introduce the concept of aggregation clusters and develop a partitioning algorithm to effectively group clients based on the alignment of their temporal reasoning properties. We evaluate the proposed method on two tasks: a real-world traffic volume prediction task consisting of sensory data from fifteen states and a smart city multi-task prediction utilizing synthetic data. The evaluation results exhibit clear improvements, with performance accuracy improved by up to 54% across all sequential prediction models.

Introducing Industrial Clusters in Multi-Node Energy System Modelling by the Application of the Industry–Infrastructure Quadrant

To reach climate neutrality and circularity targets, industry requires infrastructure guaranteeing available, accessible, affordable, and sustainable supply of renewable energy and resources. The layout and operation of the required grids are a key topic in energy system modelling, a research field under constant development to tackle energy transition challenges. Although industry is a core player, its transformation and related policy initiatives are not yet fully reflected, resulting in a research gap. The industrial cluster concept, stimulating local cross-sectoral co-operation, circularity, and optimisation, offers untapped potential to improve the spatial representation of industry in energy system models and paves the way for cluster transition research. This paper introduces the Industry–Infrastructure Quadrant to visualise the relationship between industry and infrastructure presence by means of five distinct area categories. A complementary methodology integrates industrial clusters for multi-node selection in energy system models, solely relying on open-source data and cluster algorithms (DBSCAN). A case study applied to Belgium results in ten nodes to represent the territory, accurately reflecting crucial infrastructure elements and future needs whilst improving industry representation in terms of space and composition. The work serves as a first step towards a deeper understanding of the prominence of industrial clusters in sustainable energy systems.

Data-driven identification of earthquake clusters: Clusters before the 2010 El mayor-Cucapah earthquake MW 7.1, Baja California, Mexico

Seismic clusters in background seismicity have been associated with high stress levels and can be an important precursor to large earthquakes, but there is not a unanimous concept of cluster and most cluster identification methods are cumbersome and involve a priori assumptions. We propose a simple definition of seismic cluster and a straightforward method of identification involving a minimum of parameters that can be objectively determined in a data-driven way according to a principle of low random occurrence. As an illustration, definition and method were applied to the identification of cluster activity from October 1979 to March 2010 in northern Baja California, Mexico, between 118°W to 113°W and 30°N to 33°N, a tectonically complex seismic region with several fault systems. Twenty-one clusters were identified, of which 17 located around the places at the northeastern corner of the study area that would be ruptured on April 4, 2010 by the El Mayor-Cucapah Mw 7.1 earthquake, the largest recorded earthquake in Baja California, Mexico, and the four others occurred within 9 km from its epicenter. Clustering also became slightly more frequent as the time of the earthquake approached, so that if the clustering survey had been carried out before the whole northern Baja California area, the clustering might have identified the future epicentral region as a region of interest to be closely monitored (this earthquake featured foreshock activity starting some 15 days before the main event). Although the reliability of clusters as precursors to large earthquakes is still to be studied, it is certainly useful to have a reliable and simple method to identify and characterize them.

THREE SYSTEMIC APPROACHES TO HUMAN BEHAVIOR, BEHAVIORAL HISTORY AND POPULATION BEHAVIOR: WHAT ROLE FOR SOCIAL ENVIRONMENT AND PUBLIC PARTICIPATION?

Purpose: The research considers the systemic approaches to human behavior, social environment and behavioral history engaging the theoretical representations and conceptualizations of the relevant constructs in the field. Methods and methodology: The research overviews the conceptual clarity using terms and association of terms and following a five-steps research technique: (1) determination of conceptual cluters using JavaScript based programming Mermaid Live Editor; (2) generating three themes of analysis using the tools and services provided by four databases, namely: Wiley Online Library, Sage Journals, JSTOR and the Royal Society Publishing (information retrieved during the period from January 10, 2024 to February 10, 2024); (3) identifying the associated domains and applications of the conceptual clusters by reviewing the relevant scientific literature; (4) examining the concepts and evolutionary advances; (5) framing the patterns of the terminological analysis. Results and findings: The findings of the research patterns the analysis units referring to the relationships between various constructs and discussing the variations and role of concepts. Conclusions: In conclusion, the paper releases significant research outcomes concerning linkage between human behavior, social environment and behavioral history and the occurrence and viewpoints about the three themes of analysis by focusing on the role of social environment and public participation.

Hierarchical matrix factorization for interpretable collaborative filtering

Matrix factorization (MF) is a simple collaborative filtering technique that achieves superior recommendation accuracy by decomposing the user–item interaction matrix into user and item latent matrices. Because the model typically learns each interaction independently, it may overlook the underlying shared dependencies between users and items, resulting in less stable and interpretable recommendations. Based on these insights, we propose “Hierarchical Matrix Factorization” (HMF), which incorporates clustering concepts to capture the hierarchy, where leaf nodes and other nodes correspond to users/items and clusters, respectively. Central to our approach, called hierarchical embeddings, is the additional decomposition of the latent matrices (embeddings) into probabilistic connection matrices, which link the hierarchy, and a root cluster latent matrix. The embeddings are differentiable, allowing simultaneous learning of interactions and clustering using a single gradient descent method. Furthermore, the obtained cluster-specific interactions naturally summarize user–item interactions and provide interpretability. Experimental results on ratings and ranking predictions show that HMF outperforms existing MF methods, in particular achieving a 1.37 point improvement in RMSE for sparse interactions. Additionally, it was confirmed that the clustering integration of HMF has the potential for faster learning convergence and mitigation of overfitting compared to MF, and also provides interpretability through a cluster-centered case study.

SOFTWARE MODULES CLUSTERING: A LITERATURE REVIEW

Maintenance takes the largest cost within software engineering processes. Because understanding the software code is complex and requires a deep understanding of the structure of building the code. To face Such a problem and other problems such as research purposes, or retrieval, restructuring, and understanding the behavior of software entities, today the software component, such as objects, classes, or files that have similarity in feature can be grouped together using unsupervised learning method called software module clustering. This method gives a clear picture of the components of the program and the details of the dependencies between these components, and thus it gives a structure that can be relied upon effectively. This research paper will present a compilation of studies published in this field over the past 23 years. We review 36 well-known research papers in the literature that examines software module clustering techniques and obtain useful results and data. Research containing duplicate ideas was excluded. We will attempt to answer the following questions: What are the modern methods used in the concept of clustering? What are the applications of clustering in software engineering? What are various clustering algorithms? What evaluation methods are available for assessing the quality of clustering techniques?

What is conceptual disruption?

Recent work on philosophy of technology emphasises the ways in which technology can disrupt our concepts and conceptual schemes. We analyse and challenge existing accounts of conceptual disruption, criticising views according to which conceptual disruption can be understood in terms of uncertainty for conceptual application, as well as views assuming all instances of conceptual disruption occur at the same level. We proceed to provide our own account of conceptual disruption as an interruption in the normal functioning of concepts and conceptual schemes. Moreover, we offer a multilevel taxonomy thereof, where we distinguish between instances of conceptual disruptions occurring at different levels (conceptual scheme, conceptual clusters, and individual concepts), taking on different forms (conceptual gaps and conceptual conflicts), and leading to different degrees of severity (extending from mild to severe). We also provide detailed accounts through historical examples of how conceptual gaps and conceptual conflicts can occur at different times in the very same process of conceptual disruption. Finally, we make the case that different kinds of conceptual engineering can provide meaningful ways to assess and overcome distinct types of conceptual disruption.

C4O: chain-based cooperative clustering using coati optimization algorithm in WSN

In order to provide sensing services to low-powered IoT devices, wireless sensor networks (WSNs) organize specialized transducers into networks. Energy usage is one of the most important design concerns in WSN because it is very hard to replace or recharge the batteries in sensor nodes. For an energy-constrained network, the clustering technique is crucial in preserving battery life. By strategically selecting a cluster head (CH), a network's load can be balanced, resulting in decreased energy usage and extended system life. Although clustering has been predominantly used in the literature, the concept of chain-based clustering has not yet been explored. As a result, in this paper, we employ a chain-based clustering architecture for data dissemination in the network. Furthermore, for CH selection, we employ the coati optimisation algorithm, which was recently proposed and has demonstrated significant improvement over other optimization algorithms. In this method, the parameters considered for selecting the CH are energy, node density, distance, and the network’s average energy. The simulation results show tremendous improvement over the competitive cluster-based routing algorithms in the context of network lifetime, stability period (first node dead), transmission rate, and the network's power reserves.

Self-Adjusted Consensus Clustering with Agglomerate Algorithms

This paper reports of theoretical and computational results related to an original concept of consensus clustering involving what we call the projective distance between partitions. This distance is defined as the squared difference between a partition incidence matrix and its image over the orthogonal projection in the linear space spanning the other partition incidence matrix. It appears, provided that the ensemble clustering is of a sufficient size, agglomerate clustering with the semi-average within-cluster similarity criterion effectively solves the problem of consensus partition and, moreover, of the number of clusters in it.

INDUSTRIAL TERRITORIES IN DNIPRO CITY: TRANSFORMATION STRATEGY INTO RECREATIONAL ASSETS

In the modern world, with increasing urbanization, efficient land use becomes critically important for the sustainable development of cities. The reorganization of outdated industrial territories emerges as a key direction in this process, requiring a combination of various approaches to preserve industrial function, reprofile enterprises, and adapt production to new conditions. This study is built upon a broad methodological foundation, encompassing both general scientific and specialized methods within the fields of architecture, urban planning, and urban studies. Specifically, methods such as literary, statistical, and cartographic analyses, synthesis, and generalization are employed for data collection and processing. Specialized methods, including experimental design, mental mapping, and phenomenological analysis, are also utilized to achieve the research objectives. A cluster approach was employed in developing the concept of spatial development for recreation on former industrial territories. Taking into account various factors such as economic opportunities, sociocultural factors, urban planning considerations, ecological and infrastructural requirements, as well as legal aspects, is crucial for the successful reorganization of industrial territories for recreational purposes. Achieving a balance among these factors enables the creation of resilient, functional, and culturally rich spaces that contribute to improving the quality of life for residents and the development of cities. The research encompasses a strategy for planning the reorganization of industrial territories for recreational use, which involves the identification and analysis of territorial potential and condition, planning functional transformations, architectural transformation, sustainability and ecological considerations, public participation, infrastructure transformations, economic resilience, multimodality, and adaptability. The cluster approach involves the formation of flexible entities enriched with innovative, dynamic spaces encompassing various functional zones, while preserving industrial heritage and creating green, environmentally sustainable spaces that align with the needs of modern society. Community engagement and project flexibility are key elements in creating inclusive and sustainable ecological spaces that enhance the quality of life and attract new visitors. The article analyzes the urban planning situation in the city of Dnipro, identifying shortcomings in the formation of industrial zones. The study emphasizes the need to develop strategies to address issues related to the placement of production zones in the city, which lead to limitations in territorial development and the fragmentation of the urban fabric. Based on this, a concept of a recreational cluster for industrial tourism within the industrial hub "Fabrichny" is developed, envisioning the transformation of industrial territory into recreational facilities. An analysis of the site revealed valuable architectural objects that could become the main attractors of the recreational cluster. These objects served as the basis for the development of a conceptual project for the considered area.

A conceptual clustering method for large-scale group decision-making with linguistic truth-valued lattice implication algebra

The increasing complexity of decision-making environments has led to a rise in the involvement of decision-makers (DMs) in group decision-making problems. Clustering is widely used in large-scale group decision-making (LSGDM) to categorize DMs into smaller groups. Ensuring reasonable decision-making results requires providing explanations for the generated groups during the clustering process. To address the clustering problem in LSGDM within uncertain linguistic environments, this paper proposes a conceptual clustering method based on the linguistic concept lattice. The method efficiently manages comparable and incomparable linguistic information. To achieve interpretable clustering results for DMs, attribute and expert induction matrices are first introduced. Cluster stability analysis is then employed to automatically determine the optimal number of clusters. Second, linguistic truth-valued aggregation operators are proposed to aggregate the linguistic evaluation information of DMs in each cluster. In addition, a consensus reaching process is conducted within each cluster, and a feedback mechanism is established to iteratively update clusters when consensus cannot be reached. Finally, numerical examples and comparative analyses are presented that verify the effectiveness of the proposed approach in effectively addressing the LSGDM problem within uncertain linguistic environments.

Independence of the Indonesian Defense Industry and Challenges in Defense Budget Allocation

This research aims to explore key issues related to Indonesia's economic potential, particularly its Gross Domestic Product (GDP) reaching US$9.3 trillion. The International Monetary Fund (IMF) projections indicate that by 2030, Indonesia will be one of the top economic powers, competing with India and Brazil. The primary focus of the research is the defense sector as a strategic potential requiring a proactive response. The research objectives involve exploring the level of self-sufficiency in the defense industry, understanding challenges related to defense budget allocation, and identifying factors influencing global competitiveness. The research methodology is descriptive with a qualitative document-based approach, specifically focusing on strategic State-Owned Enterprises (SOEs) within Defense ID, such as PT Len, PT Pindad, PT PAL, PT DI, and PT Dahana. Data analysis encompasses information from these SOEs, utilizing theoretical frameworks including Competitive Theory, Penta-Helix Theory, Defense Industry Cluster Concept, and Central Government Budgeting Model. Research findings indicate that SOEs within Defense ID can be considered self-sufficient in the global defense industry competitiveness. However, it is noteworthy that the Defense Budget allocation is still below 1% of the GDP, creating its own challenges. The conclusion of the research is that Indonesia has significant potential to develop a self-reliant defense industry capable of global competition. Recommendations include enhancing collaboration, developing Pentahelix or Septahelix concepts, and implementing Defense Industry Clustering to improve innovation, efficiency, and competitiveness in the sector. Additionally, the Defense Budget Allocation Policy significantly influences the progress of this industry, requiring proactive measures to ensure alignment with the desired economic growth. Thus, the research findings provide a foundation for the Government, Stakeholders, and Future Researchers to develop strategies and policies supporting optimal growth in the Indonesian Defense sector.

Exploratory Analysis of Distributional Data Using the Quantile Method

The quantile method transforms each complex object described by different histogram values to a common number of quantile vectors. This paper retraces the authors’ research, including a principal component analysis, unsupervised feature selection using hierarchical conceptual clustering, and lookup table regression model. The purpose is to show that this research is essentially based on the monotone property of quantile vectors and works cooperatively in the exploratory analysis of the given distributional data.

Collective shared value creation as emergent strategy for cluster management organizations

PurposeCluster management organizations (CMOs) have emerged over the past few decades as intermediaries that support the competitiveness of place-based clusters of economic activity. Despite their economic origins, policymakers are now starting to experiment with a broader use for cluster policies that seeks to leverage CMOs to tackle societal challenges in approaches aligned with the concept of creating shared value (CSV). However, there remains a void in conceptual understanding around the specific roles that CMOs might play in overcoming the barriers faced by their members for CSV, which this paper aims to address. Bridging this gap presents an opportunity for cluster practitioners and policymakers in a context in which environmental and social sustainability are at the top of policy agendas.Design/methodology/approachBased on analysis of literature around collaborative approaches to CSV for mitigating transaction costs, the authors define the contours of a new conceptual framework for the roles that CMOs can play in fostering collective CSV. The authors illustrate how the different components of the framework are reflected in emerging cluster practice in the context of a new wave of European cluster-based projects tackling CSV elements.FindingsThe resulting framework reconciles the concepts of clusters and CSV by explicitly positioning CMOs as intermediaries for facilitating the CSV strategies of their members. CMOs embrace emergent strategy making that targets (tangible and intangible) collective CSV capabilities and addresses collective CSV challenges. Collective CSV can provide a theoretical anchor guiding future cluster policies to fully leverage the transformative potential of CMOs. This conceptual framework opens a promising empirical research agenda, particularly around evaluating the plurality of impacts of CMOs.Originality/valueBy stressing the social impact of CMOs alongside their well-understood economic impacts, and by enabling a categorization of functions that can support the monitoring of CMO activities toward collective CSV strategies, the framework provides a novel basis for inspiring further empirical research into the evidencing of these roles.

Absenteeism Among Nurses: A Bibliometric Analysis of the Past 25 Years of Research

In the wake of the COVID-19 pandemic, healthcare has been pushed to the limits. Nursing staff, being an essential aspect of healthcare, has had to bear the brunt of the situation. And as such, in the context of healthcare, the absenteeism of nurses is of great concern. Consequently, there is, in present times, a great need for research into absenteeism, its incidence, precedents and antecedents. This bibliometric review contributes to the literature by examining how the research has been carried out by identifying the relevant references, authors, topics and journals. Having compiled major research, in the last 25 years, in the field of nurse absenteeism, this review seeks to lay the groundwork for prospective research in the field. Analysing 87 articles published globally since 1996, this article reveals the contours of research by indicating the locales of the undertaken research, locating it within various themes and conceptual clusters. The article also reviews the antecedents of nurse absenteeism as indicated in the research. This review has been conducted in view of the near absence of bibliometric work in the field of nurse absenteeism. In systematizing the published literature, the state of the present research and the contours of future research are also laid out.

Intelligent mapping of geochemical anomalies: Adaptation of DBSCAN and mean-shift clustering approaches

Cluster analysis can be used to organize samples and generate ideas regarding the multivariate geochemistry of given dataset. Traditional clustering techniques have the drawbacks of high computational complexity and poor adaptability to big data. Hence, there has recently been much focus on creating better clustering algorithms. Although many clustering algorithms have been applied, and some produce notable clustering results, the performance efficiency of algorithms is often highly dependent on the values the user chooses for the parameters. Currently, density-based spatial clustering of applications with noise clustering (DBSCAN) is widely utilized in image processing, bioinformatics, and social network analysis owing to its ability to detect clusters of various shapes. Even though partitional clustering techniques may be effective when the number of clusters K is known in advance, they cannot implement non-convex clustering and rapidly return to a local optimum. This study proposes the concept of DBSCAN clustering for stream sediment geochemical data. In this respect, the geochemical data collected from Varcheh district, SW Iran, were processed using the clr transformation before applying DBSCAN. Then, PCA was used to minimize the dimension of variables and specify the mineralization-related elements. In the following, one of the PCs connected with mineralization (PC2) was chosen for further analysis. DBSCAB, Mean-shift and Fuzzy K-means algorithms were used to monitor the multi-element geochemical anomalies linked to MVT PbZn deposits in the study area. According to Davies-Bouldin and Silhouette as two validation metrics, it can be deduced that the three SCB models are advantageous, however, the model generated by DBSCAN is preferable to the model generated by Mean-shift and Fuzzy K-means.

Deep temporal clustering features for speech emotion recognition

Deep clustering is a popular unsupervised technique for feature representation learning. We recently proposed the chunk-based DeepEmoCluster framework for speech emotion recognition (SER) to adopt the concept of deep clustering as a novel semi-supervised learning (SSL) framework, which achieved improved recognition performances over conventional reconstruction-based approaches. However, the vanilla DeepEmoCluster lacks critical sentence-level temporal information that is useful for SER tasks. This study builds upon the DeepEmoCluster framework, creating a powerful SSL approach that leverages temporal information within a sentence. We propose two sentence-level temporal modeling alternatives using either the temporal-net or the triplet loss function, resulting in a novel temporal-enhanced DeepEmoCluster framework to capture essential temporal information. The key contribution to achieving this goal is the proposed sentence-level uniform sampling strategy, which preserves the original temporal order of the data for the clustering process. An extra network module (e.g., gated recurrent unit) is utilized for the temporal-net option to encode temporal information across the data chunks. Alternatively, we can impose additional temporal constraints by using the triplet loss function while training the DeepEmoCluster framework, which does not increase model complexity. Our experimental results based on the MSP-Podcast corpus demonstrate that the proposed temporal-enhanced framework significantly outperforms the vanilla DeepEmoCluster framework and other existing SSL approaches in regression tasks for the emotional attributes arousal, dominance, and valence. The improvements are observed in fully-supervised learning or SSL implementations. Further analyses validate the effectiveness of the proposed temporal modeling, showing (1) high temporal consistency in the cluster assignment, and (2) well-separated emotional patterns in the generated clusters.

CAISeg: A Clustering-Aided Interactive Network for Lesion Segmentation in 3D Medical Imaging.

Accurate lesion segmentation in medical imaging is critical for medical diagnosis and treatment. Lesions' diverse and heterogeneous characteristics often present a distinct long-tail distribution, posing difficulties for automatic methods. Currently, interactive segmentation approaches have shown promise in improving accuracy, but still struggle to deal with tail features. This triggers a demand of effective utilizing strategies of user interaction. To this end, we propose a novel point-based interactive segmentation model called Clustering-Aided Interactive Segmentation Network (CAISeg) in 3D medical imaging. A customized Interaction-Guided Module (IGM) adopts the concept of clustering to capture features that are semantically similar to interaction points. These clustered features are then mapped to the head regions of the prompted category to facilitate more precise classification. Meanwhile, we put forward a Focus Guided Loss function to grant the network an inductive bias towards user interaction through assigning higher weights to voxels closer to the prompted points, thereby improving the responsiveness efficiency to user guidance. Evaluation across brain tumor, colon cancer, lung cancer, and pancreas cancer segmentation tasks show CAISeg's superiority over the state-of-the-art methods. It outperforms the fully automated segmentation models in accuracy, and achieves results comparable to or better than those of the leading point-based interactive methods while requiring fewer prompt points. Furthermore, we discover that CAISeg possesses good interpretability at various stages, which endows CAISeg with potential clinical application value.

Cluster approach in marketing

The article discusses the cluster approach in industry. The authors' position on the evolutionary transformation of the cluster concept into an ecosystem one is substantiated. The author's view on the content of the “diamond of competitiveness” for assessing the high-tech industry sector is proposed. On the basis of marketing design, the relationship between the cluster economy and the ecosystem concept is revealed, which will determine the methodological tools of cluster marketing and will allow evaluating its industrial corporations. It is proved that the ecosystem approach is the basis of marketing design. Building the activities of industrial companies on the basis of ecosystem ensures the solution of strategic tasks on a marketing basis.