Clustering Results Research Articles

A Clustering Ensemble Method for Drug Safety Signal Detection in Post-Marketing Surveillance.

Post-marketing surveillance refers to the process of monitoring the safety of drugs once they reach the market, after the successful completion of clinical trials. In this work, we investigate a computational approach using data mining tools to detect safety signals from post-market safety databases, or in other words, to identify adverse events (AEs) with disproportionately high reporting rates compared to other AEs, associated with a particular drug or a drug class. Essentially, we view this as a problem of cluster analysis-based anomaly detection on post-market safety data, where the goal is to 'unsupervisedly' detect the anomalous or the signal AEs. Our findings demonstrate the potential of using a clustering ensemble method to detect drug safety signals. It employs multiple clustering or anomaly detection algorithms, followed by a performance comparison of the candidate algorithms based on a collection of appropriate measures of goodness of clustering results. The method is general enough to include any number of clustering or anomaly detection algorithms and goodness measures, and performs better than any of the candidate algorithms in identifying the signal AEs. Extensive simulation studies illustrate that the ensemble method detects the AE signals from synthetic post-market safety datasets pretty accurately across the different scenarios explored. Based on the cases reported to the FDA Adverse Event Reporting System (FAERS) between 2013 and 2022, we further demonstrate that the ensemble method successfully identifies and confirms most of the adverse events that are known to occur most frequently in reaction to antiepileptic drugs and -lactam antibiotics.

The role of the O2O blended teaching model in improving the teaching effectiveness of physical education classes

Abstract The deep fusion of Internet technology and education is constantly pushing forward the reform of university education. Traditional educational ideas, concepts, and models cannot keep pace with the times, and hybrid teaching has become a new way of education in colleges and universities. To improve the teaching effect of physical education classes, the study used a blended teaching model and designed a teaching evaluation and performance prediction model under the blended teaching model based on an improved cluster analysis method and attention mechanism. The lab results indicated that under the blended teaching model, students’ performance increased by 12.89 points, and the level of skill mastery and proficiency increased by 26.52 and 28.55%, respectively, with grades more inclined to high score distribution. “Excellent” grade clustering increased by 77.71%, and “Good” grade clustering increased by 19.01%. The minimum error sum of squares of the improved clustering algorithm was 58.18 and 36.25% lower than the other two algorithms, and the clustering results were more relevant. The two-way attention mechanism algorithm predicted higher accuracy results and performed best on all four evaluation metrics, with a prediction accuracy of 98.23%, an accuracy of 98.42%, and an F1 value of 91.78%. This hybrid teaching model is more in line with the characteristics of the physical education teaching discipline, successfully cultivates students’ independent learning ability, stimulates students’ love for physical education courses, and achieves better teaching results.

Simplify to the Limit! Embedding-less Graph Collaborative Filtering for Recommender Systems

The tremendous positive driving effect of Graph Convolutional Network (GCN) and Graph Contrastive Learning (GCL) for recommender systems has become a consensus. GCN encoders are extensively used in recommendation models for capturing high-order connectivities between users and items, whereas GCL accelerates the training of recommendation tasks by adding extra supervision signals from contrastive objectives. However, little attention has been paid on corresponding theories that are truly tailored to recommendation tasks. From the technical perspective, Collaborative Filtering (CF) is seen as an important factor in recommender systems. It is applied to measure user-user, item-item, and user-item similarities rather than to achieve better clustering or node classification results. Besides, heuristic-based data augmentation may not be hold true in the field of recommender systems as it requires additional training costs and introduces noises that will corrupt the interaction graph structure and the semantic information of nodes. To tackle these limitations, we propose a novel Embedding-less Graph Collaborative Filtering (EGCF) for recommendation, which is tailor-made for the problem mentioned for CF and further simplifies existing solutions. Structurally, it consists of two parts: embedding-less GCN and embedding-less GCL. The former improves user-item affinity by streamlining user-type embeddings and carrying out iterative graph convolution. And the latter utilizes three-type contrastive objectives to directly measure the alignment and the uniformity of users, items, and interaction pairs, respectively, avoiding any type of data augmentation or multi-view construction. Even though EGCF has been extremely streamlined, extensive experimental results on three classical datasets demonstrate the effectiveness of EGCF in terms of recommendation accuracy and training efficiency. The code and used datasets are released at https://github.com/BlueGhostYi/ID-GRec.

Improved MRF rail surface defect segmentation method based on clustering features

Aiming at the characteristics of small number and many types of rail surface defect samples, as well as the problems of unstable transfer learning effect and threshold segmentation being easily affected by environmental factors in real scenes, an improved Markov defect segmentation method with zero samples is proposed. Firstly, the collected data is processed by Gabor function to highlight the defect features and reduce the data dimension to obtain the reduced dimension feature map; Kmeans clustering is performed on the processed feature map to reduce the distribution of data and reduce the influence of reflection and shadow, and the clustering result is used as the pre-classification matrix; an improved Markov random field two-layer graph model is constructed and inferred through the reduced dimension feature map and the pre-classification matrix; the local geometric structure of the defect part is analyzed according to the eigenvalues of the classification matrix inferred by the model; finally, the defect area is marked and the defect segmentation is completed. The experimental part uses a self-sampling data set, and the final conclusion is drawn based on the comparative experiment and ablation experiment. The experimental results show that the pixel accuracy, average pixel accuracy, weighted intersection-over-union ratio, and average intersection-over-union ratio of this method on the self-sampling data set are respectively 93.6%、80.7%、89.4%、68.2% , which exceeds the accuracy of other comparative detection algorithms.

A Method for Predicting the Impact of Labor Education at Institutions of Higher Education Based on a Two-Stage Clustering Algorithm

In order to develop more accurate teaching plans and methods based on students’ characteristics and needs, and improve the effectiveness of labor education, a two-stage clustering algorithm based method for predicting the effectiveness of labor education in universities is studied. A two-stage clustering algorithm-based method for predicting the effectiveness of labor education in universities has been proposed. Through in-depth analysis of labor education data in universities, principal component analysis (PCA) was used to select 16 most representative variables from the original features. Subsequently, self-organizing map (SOM) neural network was used for preliminary clustering, and the optimal number of clusters was determined to be 8. Furthermore, the K-means clustering algorithm performs fine clustering based on the initial centers provided by SOM, significantly improving the stability and accuracy of clustering. At the same time, processing the outliers in the clustering results has reduced their impact on the clustering effect. Finally, the processed data was input into an improved XGBoost prediction model, which achieved a true positive rate (TPR) of over 85% on the ROC curve when predicting the effectiveness of labor education in universities. While maintaining a low false positive rate (FPR), the model outperformed other comparison methods significantly. This achievement not only validates the effectiveness of the proposed method in feature selection, clustering optimization, and prediction model construction, but also confirms its ability to accurately predict the labor education effectiveness of students in different grades, providing strong support for higher education institutions to optimize teaching plans and improve educational outcomes.

Evaluation of germplasm resources of Sarcomyxa edulis in the Changbai Mountains of Northeast China

ABSTRACT In this study, we evaluated 16 strains of Sarcomyxa edulis from the Changbai Mountains, China through antagonism tests and ISSR molecular marker technique for cytological and molecular biological evaluation. Q-type and R-type cluster analyses were utilised to evaluate and classify agronomic traits according to the test guidelines for distinctness, uniformity, and stability. The tested strains showed antagonistic reactions. The differences in antagonistic reactions among different germplasm resources are helpful for evaluating genetic diversity and provide a basis for better development and utilisation of fungal germplasm resources. A total of 330 bands were amplified by 21 ISSR primers, including 321 polymorphic bands. The percentage of polymorphic bands observed was 97.07%. Using PopGen32 software, the effective number of alleles (Ne), Nei's gene diversity index (H), and Shannon's diversity index (I) were calculated, yielding average values of 1.6638, 0.3796, and 0.5480, respectively. In addition, the genetic similarity coefficient of 16 strains ranged from 0.4306–0.7778. According to the Q-type clustering, the tested strains were classified into two categories based on the Euclidean distance of 8.52. While R-type cluster analysis revealed a significant correlation between the traits. At the Euclidean distance of 6.298, the traits could be classified into three distinct categories. According to the findings, all 16 strains demonstrated high genetic diversity. The ISSR clustering results revealed that strains with shorter genetic distances were more similar in specific traits. However, there were notable differences between the results of comprehensive trait clustering (Q-type clustering) and ISSR clustering.

Clustering Functional Data With Measurement Errors: A Simulation-Based Approach.

Clustering analysis of functional data, which comprises observations that evolve continuously over time or space, has gained increasing attention across various scientific disciplines. Practical applications often involve functional data that are contaminated with measurement errors arising from imprecise instruments, sampling errors, or other sources. These errors can significantly distort the inherent data structure, resulting in erroneous clustering outcomes. In this article, we propose a simulation-based approach designed to mitigate the impact of measurement errors. Our proposed method estimates the distribution of functional measurement errors through repeated measurements. Subsequently, the clustering algorithm is applied to simulated data generated from the conditional distribution of the unobserved true functional data given the observed contaminated functional data, accounting for the adjustments made to rectify measurement errors. We illustrate through simulations show that the proposed method has improved numerical performance than the naive methods that neglect such errors. Our proposed method was applied to a childhood obesity study, giving more reliable clustering results.

Clustering of Student Expertise Fields Using the K-Means Algorithm (Case Study: STMIK Kaputama Binjai)

Grouping students' fields of expertise in higher education is an important issue that can provide significant benefits for students and educational institutions. STMIK Kaputama is one of the universities that has students with various fields of expertise, but the absence of data that informs the field of expertise of students is very unfortunate. Research data was obtained through questionnaires distributed to students, which included information about study programs, Grade Point Average (GPA), and areas of expertise. Clustering analysis was conducted using Matlab software to validate and implement the clustering results. The results show that the K-Means algorithm is effective in grouping students into clusters that have similar characteristics. The first cluster consists of students with expertise in programming and database, the second cluster focuses on students with networking expertise, and the third cluster includes students with various combinations of expertise.This study also found a tendency that students with certain expertise have a higher GPA than students with other expertise.

K-means-ten clustering algorithm based on min-max criterion and region division

Aiming at the problem of unstable clustering results and low solution accuracy of -means- + algorithm, an I--means-+ clustering algorithm based on min-max criterion and region division is proposed. Firstly, the min-max criterion is proposed to calculate the distance from each data point to the nearest center, and the data point with the largest distance is preferentially selected as the new cluster center to avoid the situation where multiple initial centers are clustered in the same cluster; secondly, the data points in the split cluster are divided into different regions, and a data point is selected in each region as the candidate center to increase the diversity of the candidate centers; finally, for the clusters that fail to pair, the new split cluster is reselected by gain to pair with the original deleted cluster again to improve the pairing success rate and further reduce the objective function value. Experimental results show that compared with the I means-+ algorithm, the proposed algorithm has an average improvement in solution accuracy −k −and a more stable clustering result kunder the premise of basically equivalent operation efficiency compared with 6.47%-means and kmeans+ algorithms, the proposed algorithm has higher solution accuracy.

Effects of Light Conditions on Tunnel Construction Workers’ Quality of Life and Work Productivity

Higher lighting intensity promotes workers’ productivity but few studies focus on the problems caused by lower lighting intensities at tunnel construction sites without natural light. The purpose of this study is to explore the relationship between light intensity and workers’ sleep quality, alertness, vitality, and work productivity through a theoretical structural equation model based on the literature review. Data were collected through a questionnaire survey, and 5792 valid responses were adopted for the analysis. First, the results showed that greater lighting intensity promoted workers’ productivity directly and indirectly through three mediators: sleep quality, alertness, and vitality. Then, the whole sample was classified into four groups: high intensity/high comfort, moderate intensity/moderate comfort, moderate intensity/low comfort, and low intensity/low comfort. The clustered results showed that the lighting conditions of tunnel construction sites were generally poor and that lighting comfort promoted workers’ productivity to some extent. Besides, the influence of lighting intensity on productivity declined with improved lighting conditions while the impact of lighting intensity on workers’ physiological and psychological status showed differing trends as lighting conditions varied. However, the relationships between workers’ physiological and psychological status and productivity remained stable regardless of the lighting conditions. The findings could provide a reference for developing corresponding measures to promote workers’ productivity.

Clustering Urban Roads Using Local Binary Patterns to Enhance the Accuracy of Traffic Flow Prediction

Many studies employ a dynamic statistical approach to model the relationship between roads for predicting traffic flow. Statistical relationships among roads pertain to the associations between road segments in nearby areas. Roads with similar traffic patterns usually define the relationship. Previous studies have shown that adjacent roads demonstrated a similar traffic pattern on the same day and time interval. Studying similar patterns between roads in surrounding areas provides information about the traffic state among roads in a cluster. Furthermore, the results of this finding may be considered to increase the performance of prediction of traffic conditions. In general, road segment roads are correlated with other road segments. They are connected upstream, downstream, or both; they are connected upstream, downstream, or both. To address this, we propose a Local Binary Pattern (LBP) to explore road connections by clustering congestion features between the target roads and their neighboring roads. The outcomes show that the roads with high connectivity obtained from the clustering process were utilized to predict traffic conditions using the SVM approach. The David-Bouldin Index of LBP with k-means shows the lowest compared with PCA k-means and k-means only. By evaluating the clustering result using the Davies-Bouldin Index and assessing the prediction results with SVM, we discovered that incorporating LBP with K-Means improved outcomes in identifying highly connected roads using K-Means.

Large-scale multi-view spectral clustering based on two-stage well-distributed anchor selection

Spectral clustering has attracted much attention because of its good clustering effect, but its high computational cost makes it difficult to apply to large-scale multi-view clustering. In response to this issue, a simple and efficient large-scale multi-view spectral clustering algorithm is proposed, which is based on a Two-stage Well-distributed Anchor Selection strategy (TWAS). Firstly, the data set is divided into several disjoint sample blocks to get the global well-distributed anchor candidate. Then, the algorithm proceeds to select anchor points within each local candidate anchor set. This two-stage anchor selection strategy facilitates the identification of anchors with significant representativeness at a reduced computational expense, thereby adeptly capturing the intrinsic data structure. Secondly, the present study devises an adaptive near-neighbor graph learning approach to construct an anchor-based intra-view similarity matrix. Finally, the multiple views are fused to obtain a consistent inter-view similarity matrix, and the clustering results are obtained. Extensive experiments demonstrate the effectiveness, efficiency, and stability of the TWAS algorithm.

IMPLEMENTATION OF K-MEANS AND FUZZY C-MEANS CLUSTERING FOR MAPPING TODDLER STUNTING CASES IN GUNUNGKIDUL DISTRICT

Gunungkidul Regency has the highest prevalence of stunted toddlers in the Special Region of Yogyakarta. This study aims to describe the optimal clustering results of toddler stunting cases using the k-means and fuzzy c-means methods and to describe the characteristic of the mapping results of stunting-prone areas for toddlers in Gunungkidul Regency for the years 2020 – 2022. This study maps stunting-prone areas for toddlers across 30 community health centers in Gunungkidul Regency from 2020 to 2022, with variables including the percentage of babies with low birth weight, babies born stunted, babies receiving health services, stunted toddlers, toddlers receiving health services, babies given exclusive breastfeeding, poor couples of reproductive ages, and families with adequate drinking water. The k-means clustering method determines cluster membership using the distance between objects and centroids, while the fuzzy c-means method uses the degree of membership. Cluster evaluation uses the silhouette coefficient, Calinski-Harabasz index, Davies-Bouldin index, and Dunn index to obtain optimal clustering results. The mapping results are presented as a stunting vulnerability map. The findings indicate that the optimal number of clusters is two, with the fuzzy c-means method proving more optimal than the k-means method based on evaluation scores. In 2020, there were 23 community health centers in cluster 0 and 7 in cluster 1. In 2021, there were 21 community health centers in cluster 0 and 9 in cluster 1. In 2022, there were 18 community health centers in cluster 0 and 12 in cluster 1. Generally, community health centers in cluster 0 are less optimal in specific nutrition interventions, such as for infants and toddlers. In contrast, those in cluster 1 are less optimal in sensitive nutrition interventions, such as poverty and water adequacy.

A new nonconvex multi-view subspace clustering via learning a clean low-rank representation tensor

Abstract Recently, low-rank tensor representation has achieved impressive results for multi-view subspace clustering (MSC). The typical MSC methods utilize the tensor nuclear norm as a convex surrogate of the tensor multi-rank to obtain a low-rank representation, which exhibits limited robustness when dealing with noisy and complex data scenarios. In this paper, we introduce an innovative clean low-rank tensor representation approach that combines the idea of tensor robust principal component analysis (TRPCA) with a new nonconvex tensor multi-rank approximation regularization. This integration enhances the robustness of the low-rank representation, resulting in improved performance. Furthermore, to better capture the local geometric features, we employ a higher-order manifold regularization term. To effectively address our new model, we develop an iterative algorithm that has been proven to converge to the desired Karush-Kuhn-Tucker (KKT) point. The numerical experiments on widely used datasets serve to demonstrate the efficacy and effectiveness of our new method.

Comprehensive Study on Optimizing Inland Waterway Vessel Routes Using AIS Data

Inland waterway transport is an important mode of transportation for many countries and regions. Route planning optimization can reduce navigation time, avoid traffic congestion, and improve transportation efficiency. In actual operations, many vessels determine their navigation routes based on the experience of their shipowners. When the captain fails to obtain accurate information, experience-based routes may pose significant navigation risks and may not consider the overall economic efficiency. This study proposes a comprehensive method for optimizing inland waterway vessel routes using automatic identification system (AIS) data, considering the geographical characteristics of inland waterways and navigation constraints. First, AIS data from vessels in inland waters are collected, and the multi-objective Peak Douglas–Peucker (MPDP) algorithm is applied to compress the trajectory data. Compared to the traditional DP algorithm, the MPDP algorithm reduces the average compression rate by 5.27%, decreases length loss by 0.04%, optimizes Euclidean distance by 50.16%, and improves the mean deviations in heading and speed by 23.53% and 10.86%, respectively. Next, the Ordering Points to Identify the Clustering Structure (OPTICS) algorithm is used to perform cluster analysis on the compressed route points. Compared to the traditional DBSCAN algorithm, the OPTICS algorithm identifies more clusters that are both detailed and hierarchically structured, including some critical waypoints that DBSCAN may overlook. Based on the clustering results, the A* algorithm is used to determine the connectivity between clusters. Finally, the nondominated sorting genetic algorithm II is used to select suitable route points within the connected clusters, optimizing objectives, including path length and route congestion, to form an optimized complete route. Experiments using vessel data from the waters near Shuangshan Island indicate that, when compared to three classic original routes, the proposed method achieves path length optimizations of 4.28%, 1.67%, and 0.24%, respectively, and reduces congestion by 24.15%. These improvements significantly enhance the planning efficiency of inland waterway vessel routes. These findings provide a scientific basis and technical support for inland waterway transport.

A phenological-knowledge-independent method for automatic paddy rice mapping with time series of polarimetric SAR images

Paddy rice, which sustains more than half of the global population, requires accurate and efficient mapping to ensure food security. Synthetic aperture radar (SAR) has become indispensable in this process due to its remarkable ability to operate effectively in adverse weather conditions and its sensitivity to paddy rice growth. Phenological-knowledge-based (PKB) methods have been commonly employed in conjunction with time series of SAR images for paddy rice mapping, primarily because they eliminate the need for training datasets. However, PKB methods possess inherent limitations, primarily stemming from their reliance on precise phenological information regarding paddy rice growth. This information varies across regions and paddy rice varieties, making it challenging to use PKB methods effectively on a large spatial scale, such as the national or global scale, where collecting comprehensive phenological data becomes impractical. Moreover, variations in farming practices and field conditions can lead to differences in paddy rice growth stages even within the same region. Using a generalized set of phenological knowledge in PKB methods may not be suitable for all paddy fields, potentially resulting in errors in paddy rice extraction. To address the challenges posed by PKB methods, this study proposed an innovative approach known as the phenological-knowledge-independent (PKI) method for mapping paddy rice using time series of Sentinel-1 SAR images. The central innovation of the PKI method lies in its capability to map paddy rice without relying on specific knowledge of paddy rice phenology or the need for a training dataset. This was made possible by the incorporation of three novel metrics: VH and VV normalized maximum temporal changes (NMTC) and VH temporal mean, derived from the distinctions between paddy rice and other land cover types in time series of SAR images. The PKI method was rigorously evaluated across three regions in China, each featuring different paddy rice varieties. Additionally, the PKI method was compared with two prevalent phenological-knowledge-based techniques: the automated paddy rice mapping method using SAR flooding signals (ARM-SARFS) and the manual interpretation of unsupervised clustering results (MI-UCR). The PKI method achieved an average overall accuracy of 97.99%, surpassing the ARM-SARFS, which recorded an accuracy of 89.65% due to errors stemming from phenological disparities among different paddy fields. Furthermore, the PKI method delivered results on par with the MI-UCR, which relied on the fusion of SAR and optical image time series, achieving an accuracy of 97.71%. As demonstrated by these findings, the PKI method proves highly effective in mapping paddy rice across diverse regions, all without the need for phenological knowledge or a training dataset. Consequently, it holds substantial promise for efficiently mapping paddy rice on a large spatial scale. The source code used in this study is available at https://code.earthengine.google.com/f82cf10cad64fa3f971ae99027001a6e.

Transcriptomic Analysis of the CNL Gene Family in the Resistant Rice Cultivar IR28 in Response to Ustilaginoidea virens Infection.

Rice false smut, caused by Ustilaginoidea virens, threatens rice production by reducing yields and contaminating grains with harmful ustiloxins. However, studies on resistance genes are scarce. In this study, the resistance level of IR28 (resistant cultivar) to U. virens was validated through artificial inoculation. Notably, a reactivation of resistance genes after transient down-regulation during the first 3 to 5 dpi was observed in IR28 compared to WX98 (susceptible cultivar). Cluster results of a principal component analysis and hierarchical cluster analysis of differentially expressed genes (DEGs) in the transcriptome exhibited longer expression patterns in the early infection phase of IR28, consistent with its sustained resistance response. Results of GO and KEGG enrichment analyses highlighted the suppression of immune pathways when the hyphae first invade stamen filaments at 5 dpi, but sustained up-regulated DEGs were linked to the 'Plant-pathogen interaction' (osa04626) pathway, notably disease-resistant protein RPM1 (K13457, CNLs, coil-coiled NLR). An analysis of CNLs identified 245 proteins containing Rx-CC and NB-ARC domains in the Oryza sativa Indica genome. Partial candidate CNLs were shown to exhibit up-regulation at both 1 and 5 dpi in IR28. This study provides insights into CNLs' responses to U. virens in IR28, potentially informing resistance mechanisms and genetic breeding targets.

Knowledge-guided Deep Temporal Clustering for Alzheimer's Disease Subtypes in Completed Clinical Trials.

Alzheimer's disease (AD) is a multifaceted neurodegenerative disorder with varied patient progression. We aim to test the hypothesis that AD patients can be categorized into subgroups based on differences in progression. We leveraged data from three randomized clinical trials (RCTs) to develop a knowledge-guided, deep temporal clustering (KG-DTC) framework for AD subtyping. This model combined autoencoders for contextual information capture, k-means clustering for representation formation, and clinical outcome classification for clinical knowledge integration. The derived representations, encompassing demographics, APOE genotype, cognitive assessments, brain volumes, and biomarkers, were clustered using the Gaussian Mixture Model to identify AD subtypes. Our novel KG-DTC framework was developed using placebo data from 2,087 AD patients across three solanezumab clinical trials (EXPEDITION, EXPEDITION2, and EXPEDITION3), achieving high performance in outcome prediction and clustering. The KG-DTC model demonstrated superior clustering structures, especially when combined with k-means clustering loss. External validation with independent clinical trial data showed consistent clustering results, with a 0.33 silhouette score for three clusters. The model's stability was confirmed through a leave-one-out approach, with an average adjusted Rand Index around 0.945. Three distinct AD subtypes were identified, each exhibiting unique patterns of cognitive function, neurodegeneration, and amyloid beta levels. Notably, Subtype 3 (S3) showed rapid cognitive decline across multiple clinical measures (e.g., 0.64 in S1 vs. -1.06 in S2 vs. 15.09 in S3 of average ADAS total change score, p<.001). This innovative approach offers promising insights for understanding variability in treatment outcomes and personalizing AD treatment strategies.

NS-IDBSCAN: An efficient incremental clustering method for geospatial data in network space

The exponential growth of big data presents a significant task for the incremental clustering problem. This paper proposes a density-based total clustering method in network space (NS-IDBSCAN) to cluster newly generated data. Instead of re-clustering the entire massive database, which would be much larger than the new datasets, we perform clustering with only the added data. Based on the current clustering results of the old data, the proposed algorithm checks the role of each newly added point and its neighbors in performing clustering. Depending on the density, an added point can be classified as noise, border, or core. This approach dramatically reduces the response time. Moreover, using a hash table with the key as the element's index corresponding to the point Id to store the cluster Id eliminates the need for searching by direct access, further increasing processing speed. Additionally, the proposed algorithm improves accuracy by reducing the intra-cluster distance when eliminating the phenomenon that the border points may belong to more than one cluster, as presented in the discussion. The proposed algorithm is compared with the spatial data clustering algorithm in network space. Test results on three data sources downloaded from OpenStreetMap, ESRI Open Data, and Inside Airbnb demonstrate that the proposed method significantly speeds up the processing time. The accuracy of the proposed algorithm is measured using eight indicators: Silhouette, BSS, WSS, WB, Davis-Bouldin, Dunn, Calinski-Harabasz, and NMI.

Research on Pseudo-Label Improvement Based on Probabilistic Uncertainty

Abstract The pedestrian re-identification task explored in this study has strong application scenarios in real life. Generally, in actual scenarios, the pedestrian samples taken by the camera are unlabeled, so pseudo-labeling is used. However, low-quality image feature extraction and imperfect clustering will produce low-quality and incorrect clustering results, which will have a negative impact on the re-identification performance. This article proposes a dual extraction method (PLR). The core idea of this study is to explore the distribution differences and similarities between two data sets through the pseudo-label method, so that the model can learn the distribution in one data set and apply it to another data set and accurately identify the target person to be found. Therefore, this article proposes a dual refinement method based on probability uncertainty. PLR first removes unqualified clustering results through probability uncertainty, and then continues to refine the remaining ones. Experiments have shown that compared to newer methods, PLR is approximately 1% higher in most tasks.