Explore the vast range of titles available.

Research on the problem of clustering tends to be fragmented across the pattern recognition, database, data mining, and machine learning communities. Addressing this problem in a unified way, Data Clustering: Algorithms and Applications provides complete coverage of the entire area of clustering, from basic methods to more refined and complex data clustering approaches. It pays special attention to recent issues in graphs, social networks, and other domains.The book focuses on three primary aspects of data clustering: Methods, describing key techniques commonly used for clustering, such as feature selection, agglomerative clustering, partitional clustering, density-based clustering, probabilistic clustering, grid-based clustering, spectral clustering, and nonnegative matrix factorization Domains, covering methods used for different domains of data, such as categorical data, text data, multimedia data, graph data, biological data, stream data, uncertain data, time series clustering, high-dimensional clustering, and big data Variations and Insights, discussing important variations of the clustering process, such as semisupervised clustering, interactive clustering, multiview clustering, cluster ensembles, and cluster validation In this book, top researchers from around the world explore the characteristics of clustering problems in a variety of application areas. They also explain how to glean detailed insight from the clustering process-including how to verify the quality of the underlying clusters-through supervision, human intervention, or the automated generation of alternative clusters.

This study addresses the challenge of planogram compliance in convenience stores by proposing a scalable, automated shelf monitoring system deployed across over 7,000 7-Eleven stores in Taiwan. Traditional manual audits are labor-intensive, error-prone, and costly, creating a growing need for reliable, automated solutions. To address this challenge, the proposed system integrates computer vision and deep learning techniques into a unified pipeline capable of detecting shelves, recognizing products, and comparing shelf layouts against digital planograms through a customized alignment algorithm. The system further incorporates multi-image stitching to overcome spatial constraints and construct virtual shelves that closely replicate real-world environments, improving adaptability and accuracy. Three large-scale datasets were developed to support model training and validation: 15,232 images for shelf detection, 99,135 images for product detection, and 471 product categories averaging 210 images each for classification. Automated labeling and clustering processes were introduced to substantially reduce manual annotation time. Experimental results demonstrate that the YOLOv8-based detection models achieve exceptional precision and recall across all stages. For shelf detection, the model achieved 99.23% precision, 98.93% recall, and 99.41% mAP@50, while product detection reached 94.61% precision, 93.02% recall, and 95.7% mAP@50—both surpassing transformer-based alternatives such as Deformable DETR. ResNet101 and FAN-based Transformer models achieved 99.86% accuracy on real-world retail datasets, indicating strong model stability. In the few-shot experiments, the FAN-based model showed strong adaptability and generalization, maintaining high accuracy with only five samples per class and achieving 98.39% Top-1 and 99.48% Top-5 accuracy on unseen products, demonstrating excellent transfer learning and real-time recognition capability. The system offers high accuracy, scalability, and real-time efficiency, making it a strong alternative to manual audits and a driver of smart retail innovation.

The improvement of comfort monitoring resources is pivotal for a better understanding of personal perception in indoor and outdoor environments and thus developing personalized comfort models maximizing occupants’ well-being while minimizing energy consumption. Different daily routines and their relation to the thermal sensation remain a challenge in long-term monitoring campaigns. This paper presents a new methodology to investigate the correlation between individuals’ daily Thermal Sensation Vote (TSV) and environmental exposure. Participants engaged in the long-term campaign were instructed to answer a daily survey about thermal comfort perception and wore a device continuously monitoring temperature and relative humidity in their surroundings. Normalized daily profiles of monitored variables and calculated heat index were clustered to identify common exposure profiles for each participant. The correlation between each cluster and expressed TSV was evaluated through the Kendall tau-b test. Most of the significant correlations were related to the heat index profiles, i.e., 49% of cases, suggesting that a more detailed description of physical boundaries better approximates expressed comfort. This research represents the first step towards personalized comfort models accounting for individual long-term environmental exposure. A longer campaign involving more participants should be organized in future studies, involving also physiological variables for energy-saving purposes.

Spatial heterogeneity in soil physicochemical properties is central to sustainable land management and productivity, yet remains poorly quantified across many potato landscapes. This study quantified and mapped the spatial variability of key soil attributes across Colombia’s main potato-growing departments (Antioquia, Cundinamarca, Cauca, Boyacá, Santander, Nariño, among others), and delineated edaphic homogeneous zones to guide site-specific management. We compiled and quality-controlled 3,137 georeferenced soil samples and performed exploratory multivariate analysis. Unsupervised K-means clustering was then applied to generate edaphic zones, which were characterized by pH, organic matter, texture proxies, macro- and micronutrients (with emphasis on phosphorus and iron). A multi-focus modeling approach integrating soil variables and a genetic algorithm was applied to optimize potato recommendations within an open-access platform.Three clusters emerged with clear contrasts in nutrient status and dispersion, both within and among clusters. Using crop nutritional requirements as benchmarks, 59% of the mapped area was classified as highly suitable for potato cultivation. Linking soil information with climatic covariates and available yield records revealed multiple associations between productivity and edaphic climatic conditions, underscoring the importance of genotype by environment management interactions. The outputs support an evidence-based fertilization recommendation system and a suitability (aptitude) model, both integrated into a digital decision-support platform for Colombia’s potato sector. The proposed framework provides a reproducible pathway from raw soil data to actionable zoning and management guidance capable of scaling to new regions, incorporating additional variables (e.g., spectral signatures), and updating recommendations as new data accrue. The aligning agronomic decisions with spatial soil variability, the approach enables more efficient input use, reduced environmental burdens, and improved resilience of potato production systems. Findings generalize to Andean tuber systems with comparable edaphoclimatic mosaics regionally .

Real-time and accurate longitudinal rip detection of a conveyor belt is crucial for the safety and efficiency of an industrial haulage system. However, the existing longitudinal detection methods possess drawbacks, often resulting in false alarms caused by tiny scratches on the belt surface. A method of identifying the longitudinal rip through three-dimensional (3D) point cloud processing is proposed to solve this issue. Specifically, the spatial point data of the belt surface are acquired by a binocular line laser stereo vision camera. Within these data, the suspected points induced by the rips and scratches were extracted. Subsequently, a clustering and discrimination mechanism was employed to distinguish the rips and scratches, and only the rip information was used as alarm criterion. Finally, the direction and maximum width of the rip can be effectively characterized in 3D space using the principal component analysis (PCA) method. This method was tested in practical experiments, and the experimental results indicate that this method can identify the longitudinal rip accurately in real time and simultaneously characterize it. Thus, applying this method can provide a more effective and appropriate solution to the identification scenes of longitudinal rip and other similar defects.

Spatially explicit consideration of species distribution can significantly add to our understanding of species coexistence. In this paper, we evaluated the relative importance of habitat heterogeneity and other clustering processes (e.g., dispersal limitation, collectively called the non‐habitat clustering process) in explaining the spatial distribution patterns of 341 tree species in three stem‐mapped 25–50 ha plots of tropical, subtropical, and temperate forests. Their relative importance was estimated by a method that can take one mechanism into account when estimating the effects of the other mechanism and vice versa. Our results demonstrated that habitat heterogeneity was less important in explaining the observed species patterns than other clustering processes in plots with flat topography but was more important in one of the three plots that had a complex topography. Meanwhile, both types of clustering mechanisms (habitat or non‐habitat) were pervasive among species at the 50‐ha scale across the studied plots. Our analyses also revealed considerable variation among species in the relative importance of the two types of mechanism within each plot and showed that this species‐level variation can be partially explained by differences in dispersal mode and growth form of species in a highly heterogeneous environment. Our findings provide new perspectives on the formation of species clustering. One important finding is that a significant species–habitat association does not necessarily mean that the habitat heterogeneity has a decisive influence on species distribution. The second insight is that the large species‐level variation in the relative importance of the two types of clustering mechanisms should not be ignored. Non‐habitat clustering processes can play an important role on species distribution.

The classification and recognition of radar clutter is helpful to improve the efficiency of radar signal processing and target detection. In order to realize the effective classification of uniform circular array (UCA) radar clutter data, a classification method of ground clutter data based on the chaotic genetic algorithm is proposed. In this paper, the characteristics of UCA radar ground clutter data are studied, and then the statistical characteristic factors of correlation, non-stationery and range-Doppler maps are extracted, which can be used to classify ground clutter data. Based on the clustering analysis, results of characteristic factors of radar clutter data under different wave-controlled modes in multiple scenarios, we can see: in radar clutter clustering of different scenes, the chaotic genetic algorithm can save 34.61% of clustering time and improve the classification accuracy by 42.82% compared with the standard genetic algorithm. In radar clutter clustering of different wave-controlled modes, the timeliness and accuracy of the chaotic genetic algorithm are improved by 42.69% and 20.79%, respectively, compared to standard genetic algorithm clustering. The clustering experiment results show that the chaotic genetic algorithm can effectively classify UCA radar’s ground clutter data.

Over the last ten years, due to the increase in frequency and severity of climate change effects, resilience in buildings has become a growing topic in the current global discussion on climate change adaptation. Designing both sustainable and resilient constructions would help to face such effects; however, sustainability and resilience in design have been mostly treated separately so far. Since sustainability has been considered more than resilience, paying deeper attention to the latter is indispensable to reducing building vulnerability. The purpose of this article is to examine the commonalities between the sustainability and resilience of buildings using two different approaches: (i) a systematic literature review, taking into consideration a 10-year period for selecting records, and (ii) an analysis of five green building rating systems and five resilience rating systems and guidelines selected according to their popularity and number of certified buildings. There is an overlap in some indicators between the two domains at the building level, as shown by the results from both paths. These aspects could assist in considering sustainability and resilience from the very beginning of the design process. This will ensure that buildings may be designed more effectively by considering and enhancing the synergies between the two domains. This paper targets potential stakeholders who may be interested in including such an integrated implementation in their designs.

In a smart factory, setting a production plan, relocating production equipment, and producing small batches of various products in real-time at a low cost is essential. This study discusses clustering customer orders with the same or similar process routes using a sequential pattern-mining technique, simplifying the overall production process and reducing the relocation and restructuring of equipment and machines in smart factories. We present a similarity measure to evaluate the similarity between two process routes and mathematically formulate integer programming to solve the problem of clustering similar routes. Considering process routes with alternatives, we use sequential pattern-mining techniques to cluster customer orders and determine alternative routes related to customer orders. We propose two sequential pattern-mining algorithms to expedite customer orders in smart factories. Algorithm 1 cluster customer orders and finds frequent process route patterns based on the similarity of the process routes. Algorithm 2 determines frequent sequential patterns based on the frequency of customer orders. We compared the results of 0–1 integer programming and Algorithm 1 and evaluated the algorithms' running time and memory space. This study demonstrates how data-mining techniques can be integrated into manufacturing systems to simplify process routes and reduce the complexity of the manufacturing process in the customer order phase.

Predicting rice yield in a timely, precise, and efficient manner is crucial for directing agricultural output and creating food policy. The goal of this work was to create a stable, high-precision estimate model for the yield prediction of multi-genotype rice combined with dynamic growth processes. By obtaining RGB and multispectral data of the rice canopy during the whole development stage, several bands of reflectance, vegetation index, canopy height, and canopy volume were retrieved. These remote sensing properties were used to define several curves of the rice-growing process. The k-shape technique was utilized to cluster the various characteristics based on rice growth features, and data from different groups were subsequently employed to create a yield estimation model. The results demonstrated that, in comparison to utilizing solely spectral and geometric factors, the accuracy of the multi-genotype rice estimate model based on dynamic process clustering was much higher. With a root mean square error of 315.39 kg/ha and a coefficient of determination of 0.82, the rice yield calculation based on canopy volume temporal characteristics was the most accurate. The proposed approach can support precision agriculture and improve the extraction of characteristics related to the rice growth process.