Number Of Clusters Research Articles

Enhancing healthcare decision support through explainable AI models for risk prediction

Electronic health records (EHRs) are a valuable source of information that can aid in understanding a patient’s health condition and making informed healthcare decisions. However, modelling longitudinal EHRs with heterogeneous information is a challenging task. Although recurrent neural networks (RNNs) are frequently utilized in artificial intelligence (AI) models for capturing longitudinal data, their explanatory capabilities are limited. Predictive clustering stands as the most recent advancement within this domain, offering interpretable indications at the cluster level for predicting disease risk. Nonetheless, the challenge of determining the optimal number of clusters has put a brake on the widespread application of predictive clustering for disease risk prediction. In this paper, we introduce a novel non-parametric predictive clustering-based risk prediction model that integrates the Dirichlet Process Mixture Model (DPMM) with predictive clustering via neural networks. To enhance the model’s interpretability, we integrate attention mechanisms that enable the capture of local-level evidence in addition to the cluster-level evidence provided by predictive clustering. The outcome of this research is the development of a multi-level explainable artificial intelligence (AI) model. We evaluated the proposed model on two real-world datasets and demonstrated its effectiveness in capturing longitudinal EHR information for disease risk prediction. Moreover, the model successfully produced interpretable evidence to bolster its predictions.

Understanding and including 'pink-collar' workers in employment-based travel demand models.

The segmentation of commuters into either blue or white-collar workers remains is still common in urban transport models. Internationally, models have started to use more elaborate segmentations, more reflective of changes in labour markets, such as increased female participation. Finding appropriate labour market segmentations for commute trip modelling remains a challenge. This paper harnesses a data-driven approach using unsupervised clustering-applied to 2017-20 South East Queensland Travel Survey (SEQTS) data. Commuter types are grouped by occupational, industry, and socio-demographic variables (i.e., gender, age, household size, household vehicle ownership and worker skill score). The results show that at a large number of clusters (i.e., k = 8) a highly distinct set of commuter types can be observed. But model run times tend to require a much smaller number of market segments. When only three clusters are formed (k = 3) a market segmentation emerges with one female-dominated type ('pink collar'), one male-dominated type ('blue collar') and one with both genders almost equally involved ('white collar'). There are nuances as to which workers are included in each segment, and differences in travel behaviours across the three types. 'Pink collar' workers are mostly comprised of female clerical and administrative workers, community and personal service workers and sales workers. They have the shortest median commutes for both private motorised and active transport modes. The approach and methods should assist transport planners to derive more accurate and robust market segmentations for use in large urban transport models, and, better predict the value of alternative transport projects and policies for all types of commuters.

Cluster mapping in Spain: Exploring the correlation between industrial agglomeration and regional performance

This paper presents a quantitative cluster mapping methodology for traded industries, adapted for the Spanish case; also explores the correlation between the existence of clusters and regional performance. e study is made at NUTS-2 level, and a total of forty-seven out of eighty-eight 2-digits codes for CNAE2009 are analyzed; ICT Index and Industry 4.0 Index are also designed and computed. A six-step methodology is applied departing from cross-industry linkages and implementing clustering algorithms; one set of clusters is elected and mapped over territory. The correlation analysis shows that a high number of clusters based on absolute employment data is positively correlated with variables associated with competitiveness, education, ICT adoption, and Industry 4.0, while no significant correlation is found for GDP per capita nor earning per worker.

VolCID: A MATLAB-based GUI for the detection of volcanic vent clusters

In this work, a MATLAB Graphic User Interphase (GUI) is presented to support the identification of a clustered structure of eruptive centers for which only their spatial coordinates are known. The GUI, called VolCID (Volcano Cluster IDentification), combines an unsupervised classification scheme with user approval at various stages of the analysis. It is not necessary to make any assumption about the total number of clusters expected or about other characteristics of the clusters (as, for example, minimum cluster size) at the beginning of any analysis. Due to its flexibility, VolCID allows exploration of the effect of different distances of reference and to assess the possibility of chaining effects in a systematic and reproducible form. The combination of objective calculations and subjective choices therefore helps the operator to bring forward underlying assumptions (e.g., the expected number of clusters, general shape of each cluster, number of volcanoes per area of specific clusters) that often remain unstated in these types of studies. VolCID also incorporates the possibility of completing the analysis superimposed to an image of geologic significance, as well as to complete analysis of volcanic features on other planets. By facilitating the exploration of distribution schemes, VolCID might be a helpful tool that can be used to further our understanding of volcanic phenomena (or any other geologic feature that can be represented by point-like features) on Earth and other planets as well.

A method for chromatin domain partitioning based on hypergraph clustering

For many years, multi-scale models of chromatin domains, such as A/B compartments, sub-compartments, topologically associated domains (TADs), sub-TADs, and loops have been popular. However, existing methods can only identify structures at a single scale and cannot partition multi-scale structures. In this paper, we proposed a method (TORNADOES) for chromatin domain partitioning based on hypergraph clustering. First, we use a density clustering algorithm to identify TADs at different scales based on Hi-C data with different resolutions. Then, by combining ChIP-seq data features and TAD results at different scales, we generate a hypergraph based on these TADs. Finally, we partition the chromatin domain structure at different scales, including A/B, A1, A2, B1, B2, and B3 based on the Laplacian matrix feature of the hypergraph. Similarity comparison experiments and ChIP-seq signal enrichment analysis are performed on the A/B region and sub-TAD levels, respectively, demonstrating that our method can identify chromatin domains with distinct features and provide a deeper understanding of the organizational patterns and functional differences in TADs at the genomic hierarchical structure. Comparative analysis of multiple cell line data shows that TORNADOES can better classify different numbers and types of compartments by changing the factors ChIP-seq data and clustering number used to characterize TAD compared to other methods. Source code for the TORNADOES method can be found at https://github.com/ghaiyan/TORNADOES.

The Three Hundred: Contrasting cluster galaxy density in hydrodynamical and dark matter simulations

Cluster number counts at visible and IR wavelengths will be a key cosmological probe in the next decade thanks to the Euclid satellite mission. For this purpose, the performance of cluster detection algorithms, which at these wavelengths are sensitive to the spatial distributions of the cluster galaxy members and their luminosity functions, need to be accurately characterized. Using hydrodynamical and dark-matter-only simulations, we studied a complete sample of massive clusters beyond 7 (5) times 1014 odot $ at redshift 0 (1) on a $(1.48 \ Gpc )^3$ volume. We find that the mass resolution of the current hydrodynamical simulations (1.5 times 10$^9$ M$_ odot $) is not enough to characterize the luminosity function of the sample in the perspective of Euclid data. Nevertheless, these simulations are still useful to characterize the spatial distribution of the cluster substructures assuming a common relative mass threshold for the different flavours and resolutions. By comparing with the dark-matter-only version of these simulations, we demonstrate that baryonic physics preserves significantly low-mass subhalos (galaxies), as has also been observed in previous studies with less statistics. Furthermore, by comparing the hydro simulations with higher resolution dark-matter-only simulations of the same objects and taking the same limit in subhalo mass, we find galaxy density profiles that are significantly more cuspy towards the centre of the clusters, where the low-mass substructures tend to concentrate. We conclude that using a dark-matter-only simulation may lead to some biases on the spatial distribution and density of galaxy cluster members. Based on the preliminary analysis of few high-resolution hydro simulations we conclude that a mass resolution of 1.8 times 10$^8$ h$^ $ M$_ odot $ will be needed for simulations to approach the expected magnitude limits for the Euclid survey. These simulations are currently under way.

Empirically contrasting urine drug screening-based opioid use disorder treatment outcome definitions.

A lack of consensus on the optimal outcome measures to assess opioid use disorder (OUD) treatment efficacy and their precise definition and computation has hampered the pooling of research data for evidence synthesis and meta-analyses. This study aimed to empirically contrast multiple clinical trial definitions of treatment success by applying them to the same dataset. Data analysis used a suite of functions, developed as a software package for the R language, to operationalize 61 treatment outcome definitions based on urine drug screening (UDS) results. Outcome definitions were derived from clinical trials that are among the most influential in the OUD treatment field. Outcome functions were applied to a harmonized dataset from three large-scale National Drug Abuse Treatment Clinical Trials Network (CTN) studies, which tested various medication for OUD (MOUD) options (n = 2492). Hierarchical clustering was employed to empirically contrast outcome definitions. The optimal number of clusters identified was three. Cluster 1, comprising eight definitions focused on detecting opioid-positive UDS, did not include missing UDS in outcome calculations, potentially resulting in inflated rates of treatment success. Cluster 2, with the highest variability, included 10 definitions characterized by strict criteria for treatment success, relying heavily on UDS results from either a brief period or a single study visit. The 43 definitions in Cluster 3 represented a diverse range of outcomes, conceptualized as measuring abstinence, use reduction and relapse. These definitions potentially offer more balanced measures of treatment success or failure, as they avoid the extreme methodologies characteristic of Clusters 1 and 2. Clinical trials using urine drug screening (UDS) for objective substance use assessment in outcome definitions should consider (1) incorporating missing UDS data in outcome computation and (2) avoiding over-reliance on UDS data confined to a short time frame or the occurrence of a single positive urine test following a period of abstinence.

How time weathers galaxies: The temporal impact of the cluster environment on galaxy formation and evolution

Abstract We illuminate the altered evolution of galaxies in clusters compared to central galaxies by tracking galaxies in the IllustrisTNG300 simulation as they enter isolated clusters of mass 1013 < M200, mean/M⊙ < 1015 (at z = 0). We demonstrate significant trends in galaxy properties with residence time (time since first infall) and that there is a population of galaxies that remain star-forming even many Gyrs after their infall. By comparing the properties of galaxies at their infall time to their properties at z = 0, we show how scaling relations, like the stellar-to-halo mass ratio, shift as galaxies live in the cluster environment. Galaxies with a residence time of 10 Gyr increase their stellar-to-halo mass ratio, by around 1 dex. As measurements of the steepest slope of the galaxy cluster number density profile (Rst), frequently used as a proxy for the splashback radius, have been shown to depend strongly on galaxy selection, we show how Rst depends on galaxy residence time. Using galaxies with residence times less than one cluster crossing time (≈5 Gyr) to measure Rst leads to significant offsets relative to using the entire galaxy population. Galaxies must have had the opportunity to ‘splash back’ to the first caustic to trace out a representative value of Rst, potentially leading to issues for galaxy surveys using UV-selected galaxies. Our work demonstrates that the evolution of cluster galaxies continues well into their lifetime in the cluster and departs from a typical central galaxy evolutionary path.

The novel hierarchical clustering approach using self‐organizing map with optimum dimension selection

AbstractIntroductionData clustering is an important field of machine learning that has applicability in wide areas, like, business analysis, manufacturing, energy, healthcare, traveling, and logistics. A variety of clustering applications have already been developed. Data clustering approaches based on self‐organizing map (SOM) generally use the map dimensions (of the grid) ranging from 2 × 2 to 8 × 8 (4–64 neurons [microclusters]) without any explicit reason for using the particular dimension, and therefore optimized results are not obtained. These algorithms use some secondary approaches to map these microclusters into the lower dimension (actual number of clusters), like, 2, 3, or 4, as the case may be, based on the optimum number of clusters in the specific data set. The secondary approach, observed in most of the works, is not SOM and is an algorithm, like, cut tree or the other.MethodsIn this work, the proposed approach will give an idea of how to select the most optimal higher dimension of SOM for the given data set, and this dimension is again clustered into the lower actual dimension. Primary and secondary, both utilize the SOM to cluster the data and discover that the weight matrix of the SOM is very meaningful. The optimized two‐dimensional configuration of SOM is not the same for every data set, and this work also tries to discover this configuration.ResultsThe adjusted randomized index obtained on the Iris, Wine, Wisconsin diagnostic breast cancer, New Thyroid, Seeds, A1, Imbalance, Dermatology, Ecoli, and Ionosphere is, respectively, 0.7173, 0.9134, 0.7543, 0.8041, 0.7781, 0.8907, 0.8755, 0.7543, 0.5013, and 0.1728, which outperforms all other results available on the web and when no reduction of attributes is done in this work.ConclusionsIt is found that SOM is superior to or on par with other clustering approaches, like, k‐means or the other, and could be used successfully to cluster all types of data sets. Ten benchmark data sets from diverse domains like medical, biological, and chemical are tested in this work, including the synthetic data sets.

Real-time soft body dissection simulation with parallelized graph-based shape matching on GPU

Background and ObjectiveInteractive soft tissue dissection has been a fundamental procedure in virtual surgery systems. Existing cutting algorithms involve complex topology changes of simulation meshes, which can increase simulation overhead and produce visual artifacts. In this paper, we proposed a novel graph-based shape-matching method that allows for real-time, flexible, progressive, and discontinuous cuts on soft tissue. MethodsWe employed shape-matching constraints within the position-based dynamics (PBD) framework, a widely adopted approach for real-time simulation applications. The soft tissue was effectively modeled using overlapping clusters, each governed by shape-matching constraints. The dissection process was bifurcated into two distinct stages. In the first stage, the surgical scalpel presses the surface of the soft tissue. The soft tissue is cut apart when the surface pressure exceeds a threshold, entering the second stage. To address the discrepancy between the visual mesh and the simulation model during cluster separation, we developed an Aggregate Finding Connected Components (AFCC) algorithm, optimized for GPU computation and integrated with a background grid. This approach also avoids ghost forces and fragmentation artifacts. To control the increase in the number of clusters, we also propose a merging strategy that can run in parallel. ResultsOur simulation outcomes demonstrated that the AFCC dissection algorithm effectively manages cluster separation and expansion with robustness. There were no ghost forces between the cutting surface and unrealistic fragments. Our simulation capability extended to supporting intricate and discontinuous cutting routes. Our dissection simulation maintained real-time performance even with over 100,000 particles constituting the soft tissue. ConclusionsOur real-time and robust surgical dissection simulation technique enables the performance of complex cuts in various surgical scenarios, demonstrating its potential in virtual surgery applications.

Prediction of wind energy location by parallel programming using MPI-based KMEANS clustering algorithm

ABSTRACT Renewable energy resources, like the power of the wind, are the essential sources of energy in today’s world. To keep down the greenhouse energy degasification and stop global warming, it is very important to predict the exact location where maximum wind energy is generated. The total amount of electrical power produced by a turbine relies more on the speed of the wind, pressure created by the wind, and weather conditions. The proposed method predicts the maximum power generated in a particular location using the current conditions of the weather, pressure generated through wind, and its speed. The speed of the wind and pressure of the wind are clustered using the Message Passing Interface (MPI) based KMEANS clustering algorithm. The system minimizes the amount of time required for clustering and it is done by MPI. Clustering is formed with the help Euclidean distance of each point. The wind data is collected and formed into three clusters such as low, medium, and high based on the speed of the wind. The parameters for evaluation are considered as the speed of the wind, and the direction of the wind are determined. The results show that the speed of the wind varies up to 25 km/s with the power of 2000 watts. The proposed method compares the execution time of sequential and MPI-based KMEANS clustering for different numbers of clusters. The sequential KMEANS clustering algorithm takes 3% to 7% more time compared to MPI-based KMEANS clustering algorithm. For the maximum cluster size of 5, the MPI-based KMEANS clustering algorithm produces the result in 0.42 s.

Scaling law of real traffic jams under varying travel demand

The escalation of urban traffic congestion has reached a critical extent due to rapid urbanization, capturing considerable attention within urban science and transportation research. Although preceding studies have validated the scale-free distributions in spatio-temporal congestion clusters across cities, the influence of travel demand on that distribution has yet to be explored. Using a unique traffic dataset during the COVID-19 pandemic in Shanghai 2022, we present empirical evidence that travel demand plays a pivotal role in shaping the scaling laws of traffic congestion. We uncover a noteworthy negative linear correlation between the travel demand and the traffic resilience represented by scaling exponents of congestion cluster size and recovery duration. Additionally, we reveal that travel demand broadly dominates the scale of congestion in the form of scaling laws, including the aggregated volume of congestion clusters, the number of congestion clusters, and the number of congested roads. Subsequent micro-level analysis of congestion propagation also unveils that cascade diffusion determines the demand sensitivity of congestion, while other intrinsic components, namely spontaneous generation and dissipation, are rather stable. Our findings of traffic congestion under diverse travel demand can profoundly enrich our understanding of the scale-free nature of traffic congestion and provide insights into internal mechanisms of congestion propagation.

A Novel Cervical Sagittal Classification for Asymptomatic Population Based on Cluster Analysis.

Cross-sectional study. To propose a novel cervical sagittal classification for asymptomatic people so as to deepen the understanding of cervical sagittal alignment. Cervical spine sagittal morphology varies in people. There is a lack of widely-accepted cervical sagittal classification method. In all, 183 asymptomatic subjects were included. A series of global and segmental cervical sagittal parameters were measured. Subjects with cervical lordosis (CL)<0 degrees were incorporated directly into the kyphosis (K) group. For subjects with CL ≥0 degrees, a two-step cluster analysis was used to arrive at the optimal number of clusters. The results of the expressions for the subtypes were derived by graphing. The 60 randomly selected lateral cervical spine films were evaluated by 4 spine surgeons at 4-week intervals using our classification method, the Toyama classification method and the Donk classification method. The 3 classification methods' reliability was expressed by the intra-group correlation coefficient (ICC), and convenience was expressed by the measuring time. Finally, the distribution of 4 subtypes was depicted, and sagittal parameters were compared among subtypes. Four subtypes of the cervical spine were suggested: Large lordosis (LL): CL≥-1.5×T1 slope (TS)+70°; Small lordosis (SL): -1.5×TS+50°≤CL<-1.5×TS+70°; Straight (S): 0°≤CL<-1.5×TS+50°; and K: CL<0°. The measuring time for our classification method was significantly less than the Toyama classification method (P<0.001). Our classification method showed high inter-observer reliability (ICC=0.856) and high to excellent intra-observer reliability (ICC between 0.851 and 0.913). SL was the most common type (37.7%). Men had more LL type and women had more S type and K type. The proportion of S and K increased with age. Cervical sagittal parameters were significantly different among the subtypes except for C4 vertebral body (VB) angle (P=0.546), C2-C7 SVA (P=0.628) and NT (P=0.816). We proposed a novel cervical sagittal classification for an asymptomatic population, which proved to be simple to implement with satisfactory reliability.

Clustering user preferences for personalized teleoperation control schemes via trajectory similarity analysis.

Successful operation of a teleoperated robot depends on a well-designed control scheme to translate human motion into robot motion; however, a single control scheme may not be suitable for all users. On the other hand, individual personalization of control schemes may be infeasible for designers to produce. In this paper, we present a method by which users may be classified into groups with mutually compatible control scheme preferences. Users are asked to demonstrate freehand motions to control a simulated robot in a virtual reality environment. Hand pose data is captured and compared with other users using SLAM trajectory similarity analysis techniques. The resulting pairwise trajectory error metrics are used to cluster participants based on their control motions, without foreknowledge of the number or types of control scheme preferences that may exist. The clusters identified for two different robots shows that a small number of clusters form stably for each case, each with its own control scheme paradigm. Survey data from participants validates that the clusters identified through this method correspond to the participants' control scheme rationales, and also identify nuances in participant control scheme descriptions that may not be obvious to designers relying only on participant explanations of their preferences.

[Clustering of cognitive domains among older adults with HIV].

Age and HIV are synergistic risk factors for conditions such as HIV-associated neurocognitive disorders (HAND). Yet, it is unclear whether OAWH display different cognitive profiles for HAND. To describe the cognitive patterns of OAWH treated with combined antiretroviral therapy (cART). Cross-sectional study that included 330 participants with HIV, aged 50 years or older, cared for at a tertiary care hospital in Mexico City. A short neuropsychological test battery was used to assess a wide spectrum of cognitive functions. The optimal number of cognitive clusters was determined by the silhouette method and a minimization of the Bayesian information criterion. Participants' mean age was 58.8 years (SD = 6.6), and 12.1% were women. A 3-cluster solution yielded stable Jaccard coefficients (p > 0.70). Cluster 1 showed more significant impairment in visual and verbal memory domains, whereas participants in cluster 3 showed significant impairment in language, and abstraction. Cluster 2 showed no predominance of any domain for alterations. There are different cognitive profiles among OAWH with HAND. These differences may be due to individual patterns of HIV-related and non-HIV-related factors.

Application of The Aggregation and Multicriteria Optimization Method in Assessing the Level of the Knowledge Economy of World Regions in Industry 4.0

The article proposes a new methodology for determining the level of the knowledge economy of countries in terms of the possibilities of using Industry 4.0 technologies. The relevance of the study is due to the transition to a new technological structure – “Industry 4.0” in the world. The aim of the study is to analyze data on the main sub-indices of the knowledge economy associated with Industry 4.0 for countries around the world based on the author’s methodology. As a tool for this analysis, the k-means++ clustering method was used to classify countries according to their characteristics. The assessment of the potential of Industry 4.0 is determined on the basis of 5 criteria characterizing various aspects of the development of countries at the level of economic knowledge. In the calculation process, determining the extent to which the indicators used describe the level of the knowledge economy, and the use of the aggregation formula and multi-criteria optimization, constitutes the scientific novelty of the study. It was concluded that as a result of clustering the countries of the world using the formula Aggregation and Multicriteria Optimization, the number of clusters decreased and the distribution of countries between clusters began to more adequately reflect reality. The results of the study can be used in the process of assessing the competitiveness of countries during the transition to the Industry 4.0 economy, as well as in developing strategies and concepts for countries.

Football Player Classification

Clustering is an essential tool in data mining research and applications. It is the subject of active research in many fields of study, such as computer science, data science, statistics, pattern recognition, artificial intelligence, and machine learning. Several clustering techniques have been proposed and implemented, and most of them successfully find excellent quality or optimal clustering results in the domains mentioned earlier. However, there has been a gradual shift in the choice of clustering methods among domain experts and practitioners alike, which is precipitated by the fact that most traditional clustering algorithms still depend on the number of clusters provided a priori. These conventional clustering algorithms cannot effectively handle real-world data clustering analysis problems where the number of clusters in data objects cannot be easily identified. Also, they cannot effectively manage problems where the optimal number of clusters for a high-dimensional dataset cannot be easily determined. Therefore, there is a need for improved, flexible, and efficient clustering techniques. Recently, a variety of efficient clustering algorithms have been proposed in the literature, and these algorithms produced good results when evaluated on real-world clustering problems. This study presents an up-to-date systematic and comprehensive review of traditional and state-of-the-art clustering techniques for different domains. This survey considers clustering from a more practical perspective. It shows the outstanding role of clustering in various disciplines, such as education, marketing, medicine, biology, and bioinformatics. It also discusses the application of clustering to different fields attracting intensive efforts among the scientific community, such as big data, artificial intelligence, and robotics. This survey paper will be beneficial for both practitioners and researchers. It will serve as a good reference point for researchers and practitioners to design improved and efficient state-of-the-art clustering algorithms.

Effects of correlation between bounded confidences and node degrees on consensus in opinion dynamics on complex networks

In bounded confidence opinion dynamics, agents only consider other agents when their opinions are sufficiently close. In many works, the confidence bound is assumed as a homogeneous one. However, agents in different situations may be with different confidence bounds in real life. In this work, the correlation of degrees and confidence bounds is introduced into modified Hegselmann–Krause model on complex networks and we mainly focus on whether the population can reach consensus, where all agents hold a same opinion. The number of opinion clusters, the relative size of the largest cluster and the probability of reaching consensus are the measures of the opinion cluster profile. The numerical simulation shows that the existence of the correlation promotes the probability of reaching consensus and even in the case where some agents are with very small confidence bounds. On scale-free and Erdös–Rényi random networks, the negative correlation supports the consensus much more than the positive correlation case. However, two correlations show the same effect on small-world networks, and this effect is also observed in scale-free and Erdös–Rényi random networks when the average degree becomes very large.

K-MEANS CLUSTERING USING ELBOW METHOD IN CASE OF DIABETES MILLETUS TYPE II IN INDONESIA

Indonesia is ranked 7th out of 10 countries with thehighest number of sufferers. BPJS Health include First LevelHealth Facilities (FKTP) and Advanced Referral HealthFacilities (FKRTL) which Type 2 diabetes mellitus is one ofthe ten most common diagnoses at FKRTL visits and ranksthird after follow-up examinations after treatment forconditions other than malignant neoplasms and kidney failurewith a percentage of 3.54% and a total of 62,455 for 2019 to2020. In deciding policies related to the funding of BPJSparticipants who suffer from diabetes mellitus, it is necessaryto have the characteristics of each region so that policy makingis more appropriate. The method used in this study usesclustering analysis using the K-means algorithm for typeII diabetes mellitus in Indonesia from 2015-2020 byprovince. Based on the outcome of the clusteringprovinces in indonesia using K-Means algorithm withoptimization of the determination of the number ofcluster using elbow method formed 3 cluster. 1 st cluster has21 province, the 2 nd cluster has 9 province and the 3 rd clusterhas 4 province.

From Sensors to Insights: An Original Method for Consumer Behavior Identification in Appliance Usage

In light of the energy crisis, extensive research is being conducted to enhance load forecasting, optimize the targeting of demand response programs, and advise building occupants on actions to enhance energy performance. Cluster analysis is increasingly applied to usage data across all consumer types. More accurate consumer identification translates to improved resource planning. In the context of Industry 4.0, where comprehensive data are collected across various domains, we propose using existing sensor data from household appliances to extract the usage patterns and characterize the resource demands of consumers from residential households. We propose a general pipeline for extracting features from raw sensor data alongside global features for clustering device usages and classifying them based on extracted time series. We applied the proposed method to real data from three different types of household devices. We propose a strategy to identify the number of existent clusters in real data. We employed the label data obtained from clustering for the classification of consumers based on data recorded on different time ranges and achieved an increase in accuracy of up to 15% when we expanded the time range for the recorded data on the entire dataset, obtaining an accuracy of over 99.89%. We further explore the data meta-features for a minimal dataset by examining the necessary time interval for the recorded data, dataset dimensions, and the feature set. This analysis aims to achieve an effective trade-off between time and performance.