Cluster Analysis Algorithm Research Articles

Customer Churn Analysis Prediction Based on Cluster Analysis and Machine Learning Algorithms

Customer Churn Analysis Prediction Based on Cluster Analysis and Machine Learning Algorithms

Reasonable Energy-Abandonment Operation of a Combined Power Generation System with an Ultra-High Proportion of Renewable Energy

With large-scale grid-connected renewable energy, new power systems require more flexible and reliable energy storage power sources. Pumped storage stations play an important role in peak shaving, valley filling, and promoting renewable energy consumption. This paper presents the reasonable energy-abandonment operation of a combined power generation system (CPGS), in which a pumped storage station is the core control power, with an ultra-high proportion of renewable energy. Firstly, based on the seasonal characteristics of wind, solar, and load demand, typical days are selected through improved clustering analysis algorithms. Then, a daily optimal scheduling model for combined power generation systems (CPGS) is developed with the goals of economy, low-carbon, and stable operation. Finally, the correlation between the energy-abandonment rate and pumped storage station peak shaving and system optimization operation indicators is obtained by a reasonable energy-abandonment calculation method considering source-grid-load coordination. Taking the operation data of an energy base in the western region of China as an example, when the penetration rate of renewable energy is 60–70% in the future, the operating cost on the power side is greatly affected by the construction of the source side. When the system operates at a planned reasonable energy-abandonment rate of 2%, electricity regulation, load tracking, and daily operating costs all show better performance.

Economic Supply Chain Management System of Farm Products Industry Based on the Belt and Road Under the Background of Big Data

Under the strategic background of "One Belt, One Road", the future farm products industry will also become part of the country's core competitiveness. Promoting the transformation of enterprises in the supply chain is conducive to cultivating the country's core competitiveness and creating good output conditions for my country's manufacturing industry. Under the national strategy of "One Belt, One Road", the agricultural product industry is facing new development opportunities. This paper analyzes the background, significance and existing basic problems of the "One Belt, One Road". Starting from the economic development of the national agricultural product industry, it uses the cluster analysis algorithm to analyze the current overall economic situation of the agricultural product industry, points out the current problems in the industry, and believes that the future farm The product industry is facing greater pressure for transformation and upgrading. Finally, combined with the strategic background of "One Belt One Road", from the perspectives of society, government and enterprises, analyze its supply chain process, extract relevant problems, and put forward relevant countermeasures and suggestions from the perspective of intra-chain, extra-chain and their interactive integration .

РАЗРАБОТКА МЕХАНИЗМОВ ПОВЫШЕНИЯ ДОСТУПНОСТИ ЛЕКАРСТВЕННОЙ ПОМОЩИ В УСЛОВИЯХ ЧРЕЗВЫЧАЙНЫХ СИТУАЦИЙ С ИСПОЛЬЗОВАНИЕМ ГЕОИНФОРМАЦИОННЫХ ТЕХНОЛОГИЙ

The article considers the issues of availability of reserves of medicinal products intended to provide medicinal supplies to victims in the elimination of the medical and sanitary consequences of emergencies, taking into account the internal risks for territories with complex climatic and geographical characteristics. The purpose of the study was to develop approaches to increase the territorial availability of medicinal products in the Tomsk Region reserve in emergency situations. In the course of the research, content analysis of data from regulatory documents and scientific publications on the medicinal provision of the population in emergency situations was used; correlation analysis of geo-information, spatial-geographical and natural-climatic indicators characterizing the territorial availability of medicinal products. The typologization of the Tomsk region according to the degree of transport accessibility was performed using cluster analysis algorithms. It has been established that the main climatic and geographic factors creating the internal risks for the drug supply system of the disaster medicine service of the Tomsk region include dangerous hydrological phenomena, forest and peat fires, frosts and hurricanes. More than 85% of the territory of the Tomsk region belongs to hard-to-reach areas equated by climatic conditions to those of the Far North, which poses risks for the drug supply system. Based on the data of the geo-information services, the road accessibility of the Tomsk region districts in relation to the regional center has been considered, two directions «Eastern» and « Northern» have been identified including 5 and 11 respectively. It was established that direct interaction between the districts belonging to these areas bypassing the regional center is extremely difficult. In the process of typologization of the districts according to the degree of transport accessibility in the direction "East", one group with a high degree of transport accessibility between 3 districts that are remote from the regional center has been identified. Two groups of districts have been identified in the North direction. The first group, characterized by high transport accessibility, included 5 districts, including the regional center. The second group with low transport accessibility included 6 districts. The optimal area for the placement of an additional point of medicines of the Tomsk region reserve has been justified, taking into account the transport infrastructure of the districts, natural and climatic risks, socio-demographic factors and the standard time for the organization of drug provision in an emergency situation. With the use of geo-information technologies, a model of decentralization of the reserve has been developed to increase the availability of drug care by the regional disaster medicine service of the Tomsk region in the context of internal risks.

Unsupervised Multivariate Feature-Based Adaptive Clustering Analysis of Epileptic EEG Signals.

Supervised classification algorithms for processing epileptic EEG signals rely heavily on the label information of the data, and existing supervised methods cannot effectively solve the problem of analyzing unlabeled epileptic EEG signals. In the traditional unsupervised clustering algorithm, the number of clusters and the global parameters must be predetermined, and the algorithm's analytical results are combined with a huge number of subjective errors, which affects the detection accuracy. For this reason, this paper proposes an unsupervised multivariate feature adaptive clustering analysis algorithm based on epileptic EEG signals. First, CEEMDAN and CWT are introduced into the epileptic EEG signal after preprocessing for joint denoising to further improve the signal quality. Then, the multivariate feature set of the signal is extracted and constructed, which includes nonlinear, time, frequency, and time-frequency characteristics. To reveal the hidden structures and correlations in the high-dimensional feature data, t-SNE dimensionality reduction is introduced. Finally, the DBSCAN clustering algorithm is optimized using the SSA algorithm to achieve adaptive selection of cluster number and global parameters.It not only enhances the clustering performance and reliability of the clustering results, but also avoids subjective errors in the analysis results. It provides a pre-theoretical foundation for the successful development of future seizure prediction devices and has good application prospects in clinical diagnosis and daily monitoring of patients.

Design of music recommendation system based on EDA and K-means cluster analysis

This paper describes the construction and analysis of a music recommendation system, through visual and exploratory analysis of the open source dataset of music recommendation system in the UCL database, exploring the relationship between music popularity and music acoustic properties, using the K-means clustering algorithm for clustering analysis of genres and songs, and finally building a music recommendation system that can, based on the user input of music data, to match the ten music songs that best meet the listener's taste. The analyses in this paper show that there is a certain correlation between music popularity and the indicators of sound, danceability, energy, instrumentality, liveliness, loudness, language, rhythm, mood, duration and tone. The indicators of energy, liveliness, loudness and tempo are positively correlated with the popularity of music, while the indicators of mood and duration are negatively correlated with the popularity of music. In addition, this paper uses K-means clustering algorithm to cluster analysis of genres and songs, and shows the similarity and difference between different genres and songs through visualisation, which provides important data support for the construction of music recommendation system. Finally, this paper establishes a music recommendation system based on K-means clustering algorithm, which can match the ten music songs that best meet the listener's taste according to the music data input by the user. The system is intelligent and personalised, and can recommend music works that are more in line with the user's taste based on the user's preferences and historical listening records. The results of this paper provide useful reference and support for the algorithms and models of the music recommendation system, which can be further researched in depth in the future to improve its intelligence and personalisation level and provide users with better music services.

Identification and prediction of association patterns between nutrient intake and anemia using machine learning techniques: results from a cross-sectional study with university female students from Palestine

PurposeThis study utilized data mining and machine learning (ML) techniques to identify new patterns and classifications of the associations between nutrient intake and anemia among university students.MethodsWe employed K-means clustering analysis algorithm and Decision Tree (DT) technique to identify the association between anemia and vitamin and mineral intakes. We normalized and balanced the data based on anemia weighted clusters for improving ML models’ accuracy. In addition, t-tests and Analysis of Variance (ANOVA) were performed to identify significant differences between the clusters. We evaluated the models on a balanced dataset of 755 female participants from the Hebron district in Palestine.ResultsOur study found that 34.8% of the participants were anemic. The intake of various micronutrients (i.e., folate, Vit A, B5, B6, B12, C, E, Ca, Fe, and Mg) was below RDA/AI values, which indicated an overall unbalanced malnutrition in the present cohort. Anemia was significantly associated with intakes of energy, protein, fat, Vit B1, B5, B6, C, Mg, Cu and Zn. On the other hand, intakes of protein, Vit B2, B5, B6, C, E, choline, folate, phosphorus, Mn and Zn were significantly lower in anemic than in non-anemic subjects. DT classification models for vitamins and minerals (accuracy rate: 82.1%) identified an inverse association between intakes of Vit B2, B3, B5, B6, B12, E, folate, Zn, Mg, Fe and Mn and prevalence of anemia.ConclusionsBesides the nutrients commonly known to be linked to anemia—like folate, Vit B6, C, B12, or Fe—the cluster analyses in the present cohort of young female university students have also found choline, Vit E, B2, Zn, Mg, Mn, and phosphorus as additional nutrients that might relate to the development of anemia. Further research is needed to elucidate if the intake of these nutrients might influence the risk of anemia.

A systematic review of symbiotic organisms search algorithm for data clustering and predictive analysis

Abstract In recent years, the field of data analytics has witnessed a surge in innovative techniques to handle the ever-increasing volume and complexity of data. Among these, nature-inspired algorithms have gained significant attention due to their ability to efficiently mimic natural processes and solve intricate problems. One such algorithm, the symbiotic organisms search (SOS) Algorithm, has emerged as a promising approach for clustering and predictive analytics tasks, drawing inspiration from the symbiotic relationships observed in biological ecosystems. Metaheuristics such as the SOS have been frequently employed in clustering to discover suitable solutions for complicated issues. Despite the numerous research works on clustering and SOS-based predictive techniques, there have been minimal secondary investigations in the field. The aim of this study is to fill this gap by performing a systematic literature review (SLR) on SOS-based clustering models focusing on various aspects, including the adopted clustering approach, feature selection approach, and hybridized algorithms combining K-means algorithm with different SOS algorithms. This review aims to guide researchers to better understand the issues and challenges in this area. The study assesses the unique articles published in journals and conferences over the last ten years (2014–2023). After the abstract and full-text eligibility analysis, a limited number of articles were considered for this SLR. The findings show that various SOS methods were adapted as clustering and feature selection methods in which CSOS, discrete SOS, and multiagent SOS are mostly used for the clustering applications, and binary SOS, binary SOS with S-shaped transfer functions, and BSOSVT are used for feature selection problems. The findings also revealed that, of all the selected studies for this review, only a few studies specifically focused on hybridizing SOS with K-means algorithm for automatic data clustering application. Finally, the study analyzes the study gaps and the research prospects for SOS-based clustering methods.

Metabolic profiling of bacteria with the application of polypyrrole-MOF SPME fibers and plasmonic nanostructured LDI-MS substrates

Here we present application of innovative lab-made analytical devices such as plasmonic silver nanostructured substrates and polypyrrole-MOF solid-phase microextraction fibers for metabolic profiling of bacteria. For the first time, comprehensive metabolic profiling of both volatile and non-volatile low-molecular weight compounds in eight bacterial strains was carried out with utilization of lab-made devices. Profiles of low molecular weight metabolites were analyzed for similarities and differences using principal component analysis, hierarchical cluster analysis and random forest algorithm. The results showed clear differentiation between Gram positive (G+) and Gram negative (G−) species which were identified as distinct clusters according to their volatile metabolites. In case of non-volatile metabolites, differentiation between G+ and G− species and clustering for all eight species were observed for the chloroform fraction of the Bligh & Dyer extract, while methanolic fraction failed to recover specific ions in the profile. Furthermore, the results showed correlation between volatile and non-volatile metabolites, which suggests that lab-made devices presented in the current study might be complementary and therefore, useful for species differentiation and gaining insights into bacterial metabolic pathways.

A method for identifying the operating state of deteriorated insulators based on partial discharge signal reconstruction algorithm and spatial electric field distribution law

In order to improve the accuracy of identifying the working status of degraded insulators and thereby contribute to preventive maintenance and repair, this proposes a method for identifying the operating status of deteriorated insulators based on a partial discharge signal reconstruction algorithm and spatial electric field distribution law. Firstly, the partial discharge signal of the deteriorated insulator is obtained through capacitive sensor technology and clustering analysis algorithm. Then, the mapping features between the partial discharge signal and the electric field distribution are extracted using the SN-EMD algorithm to determine the spatial electric field distribution on the surface of the insulator. Finally, a state recognition model is established based on the hidden Markov model to identify the operating status of deteriorated insulators. The experimental results show that the method designed in this paper has superior performance in terms of accuracy.

A task offloading approach based on risk assessment to mitigate edge DDoS attacks

Edge computing mitigates the high latency and other issues associated with cloud computing, but it also introduces new risks. One such issue is DDoS attacks on edge servers, which arise when edge tasks are offloaded. There is a dearth of research on countermeasures for these kinds of DDoS attacks. Consequently, we present EDM_TOS, a task offloading strategy. This technique makes sure that tasks are transferred to dependable edge nodes based on the edge nodes' risk assessment mechanism, thereby reducing edge DDoS attacks. On the other hand, EDM_TOS works better than current approaches in five domains of evaluating awareness and capacity limitations for resolving edge DDoS attacks and issues with edge computing offloading. The edge computing task offloading algorithm, weighted cluster analysis algorithm, and risk assessment algorithm, respectively, are used to implement the risk assessment module, cluster analysis module, and task offloading module of the EDM_TOS mechanism. The three modules were evaluated experimentally. The findings demonstrate that the task offloading module's transactions per second (TPS) and the risk assessment module's malicious detection rate (MDR) respectively outperform to those of other policies. The cluster analysis module's clustering effect is superior to that of the conventional K-means algorithm. EDM_TOS performs better than current methods for thwarting edge DDoS attacks, according to performance assessments on five datasets. EDM_TOS can successfully fend off edge DDoS attacks, as demonstrated by practice.

Exploring elucidation of red dye mixtures on woolen historical textiles via non-destructive spectroscopic analysis and multivariate cluster analysis

One of the foremost challenges facing analysis of historical textiles is that the gold standard technique—high performance liquid chromatography (HPLC)—is inherently destructive. This is especially problematic considering many historical textiles are exceptionally fragile, be it from age, poor care over time, etc. One proposed solution to this is the implementation of non-destructive, namely spectroscopic, techniques, such as diffuse reflectance (Fiber Optic Reflectance Spectroscopy, FORS). In this work, 204 well-provenanced red Norwich textiles were measured with FORS and analyzed to attempt to determine chromophore combinations used to dye the original textiles. To these ends, cluster analysis algorithms and spectroscopic domain knowledge were coupled with selective HPLC validation to assess overall ability of FORS to discern changes in chromophore combinations. It was found that the near infrared (NIR) region of the spectrum contained little meaningful information in multivariate space, while the VIS region, particularly 380–469 nm, showed a narrow visible region that was primarily responsible for clustering behavior that correlates with HPLC-validated samples. This indicates that FORS shows promise for discerning chromophores in textile swatches. Additionally, X-ray fluorescence (XRF) analysis was used to confirm that the observed FORS spectral inflection point shift in the 600 nm region did not correlate with the presence or type of mordant used when dying these textiles. From this work, three main conclusions can be drawn: (1) FORS adequately identifies visual infon, which shows reasonable correlation to HPLC-validated dye recipes, warranting further investigation, and indicating utility for cois or use for those with visual impairments; (2) XRF analysis confirms that the ~ 600 nm inflection point shift and mordant are not correlated when measuring dyed textiles, unless mordant is present below analytical detection limits or not present at all; (3) many documented structural-to-spectral relationships established in the conservation literature are too weak in dyed textiles for statistical analysis and, by extension, expert spectral identification.

Retracted: Data Analysis of College Students’ Mental Health Based on Clustering Analysis Algorithm

Retracted: Data Analysis of College Students’ Mental Health Based on Clustering Analysis Algorithm

Analyzing shifts in age-related macular degeneration research trends since 2014: A bibliometric study with triple-map Sankey diagrams (TMSD).

Age-related macular degeneration (AMD) is the primary cause of vision impairment in older adults, especially in developed countries. While many articles on AMD exist in the literature, none specifically delve into the trends based on document categories. While bibliometric studies typically use dual-map overlays to highlight new trends, these can become congested and unclear with standard formats (e.g., in CiteSpace software). In this study, we introduce a unique triple-map Sankey diagram (TMSD) to assess the evolution of AMD research. Our objective is to understand the nuances of AMD articles and show the effectiveness of TMSD in determining whether AMD research trends have shifted over the past decade. We collected 7465 articles and review pieces related to AMD written by ophthalmologists from the Web of Science core collection, accumulating article metadata from 2014 onward. To delve into the characteristics of these AMD articles, we employed various visualization methods, with a special focus on TMSD to track research evolution. We adopted the descriptive, diagnostic, predictive, and prescriptive analytics (DDPP) model, complemented by the follower-leading clustering algorithm (FLCA) for clustering analysis. This synergistic approach proved efficient in identifying and showcasing research focal points and budding trends using network charts within the DDPP framework. Our findings indicate that: in countries, institutes, years, authors, and journals, the dominant entities were the United States, the University of Bonn in Germany, the year 2021, Dr Jae Hui Kim from South Korea, and the journal "Retina"; in accordance with the TMSD, AMD research trends have not changed significantly since 2014, as the top 4 categories for 3 citing, active, and cited articles have not changed, in sequence (Ophthalmology, Science & Technology - Other Topics, General & Internal Medicine, Pharmacology & Pharmacy). The introduced TMSD, which incorporates the FLCA algorithm and features in 3 columns-cited, active, and citing research categories-offers readers clearer insights into research developments compared to the traditional dual-map overlays from CiteSpace software. Such tools are especially valuable for streamlining the visualization of the intricate data often seen in bibliometric studies.

Prediction and Management of Building Energy Consumption Based on Building Environment Simulation Design Platform DeST and Meteorological Data Analysis Algorithm

Currently, the carbon emissions of building energy consumption account for a significant portion of all carbon emissions. How to reduce carbon emissions to achieve carbon neutrality is an important current research direction. Therefore this research builds a predictive algorithm model for analyzing energy consumption data of meteorological buildings using DeST platform for energy saving and emission reduction to achieve carbon neutrality. The new model uses Internet of Things and cloud platform technology to build a simulation building platform, and uses the support vector machine algorithm in the analysis algorithm to vectorize building energy consumption data, which can achieve normalization processing of building energy consumption and meteorological data. By processing building energy consumption data, prediction of building energy consumption at the next moment can be achieved. The experimental results show that the precision and accuracy of the new algorithm are higher than genetic algorithm 1 and 0.15 respectively, and 0.6 and 0.07 higher than clustering analysis algorithm respectively. Therefore, applying this algorithm model to building energy consumption prediction can significantly improve the accuracy and precision of the algorithm.

Study of Methods for Constructing Intelligent Learning Models Supported by Artificial Intelligence

INTRODUCTION: As the essential part of intelligent learning, innovative learning model construction is conducive to improving the quality of intelligent new teaching models, thus leading the deep integration of teaching and artificial intelligence and accelerating the change and development of teaching supported by artificial intelligence.OBJECTIVES: Aiming at the current intelligent teaching evaluation design method, there are problems such as more objectivity, poor precision, and a single method of evaluation indexes.METHODS: his paper proposes an intelligent learning construction method based on cluster analysis and deep learning algorithms. First of all, the intelligent learning model construction process is sorted out by clarifying the idea of clever learning model construction and extracting model elements; then, the intelligent learning model is constructed through a K-means clustering algorithm and deep compression sparse self-encoder; finally, the effectiveness and high efficiency of the proposed method is verified through simulation experiment analysis.RESULTS: Solved the problem that the intelligent learning model construction method is not objective enough, has poor accuracy and is not efficient enough.CONCLUSION: The results show that the proposed method improves the model’s accuracy.

Improving Three-Dimensional Building Segmentation on Three-Dimensional City Models through Simulated Data and Contextual Analysis for Building Extraction

Digital twins are increasingly gaining popularity as a method for simulating intricate natural and urban environments, with the precise segmentation of 3D objects playing an important role. This study focuses on developing a methodology for extracting buildings from textured 3D meshes, employing the PicassoNet-II semantic segmentation architecture. Additionally, we integrate Markov field-based contextual analysis for post-segmentation assessment and cluster analysis algorithms for building instantiation. Training a model to adapt to diverse datasets necessitates a substantial volume of annotated data, encompassing both real data from Quebec City, Canada, and simulated data from Evermotion and Unreal Engine. The experimental results indicate that incorporating simulated data improves segmentation accuracy, especially for under-represented features, and the DBSCAN algorithm proves effective in extracting isolated buildings. We further show that the model is highly sensible for the method of creating 3D meshes.

Quality Analysis of Industry-Teaching Integration of Higher Vocational Boutique Courses - Based on Cluster Analysis Algorithm

Abstract With the rapid evolution of the socio-economic landscape and ongoing adjustments in industrial structures, the integration of industry and education has emerged as a pivotal strategy for enhancing vocational education. This approach is crucial for elevating teaching quality and fostering comprehensive student development. In this paper, we utilize cluster analysis to evaluate the integration of industry and education within higher vocational boutique courses. Z school was chosen for the study, where we collected a substantial dataset via questionnaires. We used factor analysis to extract significant influencers of course quality and then applied the K-means clustering algorithm to find the optimal center of mass for the integration. Our analysis distinguishes the disparities among different courses. The study finally classified the course quality into four categories: course A (14.25%), course B (31.25%), course C (33.00%), and course D (21.5). In terms of course quality, the highest-rated course was Course B, with an average score of 90.03, and the lowest-rated course was Course D, with an average score of 70.76, a difference of almost one-quarter. The level of course quality largely influences the results of course evaluation, but the degree of perceived interest and effectiveness are the key factors in improving course quality. The reform and development of higher education are strongly supported and referenced by this study.

Exploring the development of college students’ sports interests in physical education based on cluster analysis

Abstract Good sports interest helps college students to actively participate in physical exercise, excellently complete sports tasks, form good sports habits and have a strong body. Given this, this paper establishes an innovative model of college physical education based on the 4R Theory, fully integrates physical education services with classroom teaching, and designs teaching experiments related to physical education services and sports education in colleges and universities. The subjects were Chengdu University of Technology students. The cluster analysis algorithm was used to cluster students’ sports interests, and the parameter estimation of the multiple linear regression model was carried out by the least squares method to deeply analyze the relevant factors affecting students’ sports interests in colleges and universities. The study found that: students’ sports interest clustering is divided into five categories, the number of students in cluster 1 sports awareness reached 542, and students’ sports interest increased by 0.627 percentage points for every 1 percentage point increase in sports skills, and students’ sports participation interest in the experimental class increased by 46.79% compared to that of the control class after the experiment. Adopting diversified physical education teaching methods and contents can effectively enhance college students’ interest in sports, thus providing guarantee for improving students’ physical fitness.

Research on the Innovation of Education Management Informatization Based on Deep Learning Models

Abstract In the digital era, the rapid development of information technology has not only changed the education model but also provided new tools and paths for education management. This paper establishes the two most commonly used metrics of support and confidence for association rule analysis. Using the Apriori algorithm, all frequent items in the dataset that satisfy the mining conditions are extracted, and through the constraints of the metrics, the eligible association rules are screened, and the association rule results are parsed. Add the interest degree model to reduce the generation of non-interesting rules. The cluster analysis algorithm and it are combined to apply it in the field of education management, which is empirically examined. The results of cluster analysis and association rule mining show that the support degree of rule 1 is 0.4925, and the confidence level is 1, indicating that if 49.25% of all students score between 2 and 4 in question 3, then the score in question 1 is similar to the score in question 3. Based on the weights calculated by the association rules, the teachers’ basic teaching literacy, classroom teaching ability, teaching and research ability, professional ability and after-school counseling were evaluated respectively, and the overall score of teacher 1 was the highest, with a score of 91.213, which accounted for 10% of the total number of students, and the evaluation method adopted in this paper was scientific and fair.