Speaker Model Clustering to Construct Background Models for Speaker Verification by Gökay DİŞKEN, Zekeriya TÜFEKCİ, Ulus ÇEVİK

“…First, a UBM is trained, and speaker models are adapted from the UBM. Then, the $k$-means algorithm with the Euclidean distance measure is applied to the speaker models. …”
Get full text

DFDA-AD: An Approach with Dual Feature Extraction Architecture and Dual Attention Mechanism for Image Anomaly Detection by Babak Masoudi

“…Two attention mechanisms are improved and developed in this paper, which provide more important feature maps for clustering by K-means algorithm. The evaluation of the model's performance was done on the MVTec AD data set, and the results of the evaluations for anomaly detection and localization were satisfactory compared to several other approaches that have been recently proposed.…”
Get full text

Task offloading and resource allocation in vehicle heterogeneous networks with MEC by Haibo ZHANG, Qiuji LUAN, Jiang ZHU, Fangwei LI

“…Based on the advantages of high-bandwidth and low-latency brought by mobile edge computing (MEC),which could provide IT service environment and cloud computing capability,combined with the long-term evolution unlicensed (LTE-U) technology,the task offloading decision and resource allocation issues in vehicle heterogeneous network were studied.Considering the link differentiation requirements,which were the high capacity of vehicle-to-roadside unit (V2I) links and the super reliability of vehicle-to-vehicle (V2V) links,quality of service (QoS) was modeled as the combination of capacity and latency.Firstly,the improved K-means algorithm was used to cluster the request vehicles according to different QoS to determine the communication mode.Secondly,the LTE-U technology based on non-competition period (CFP) which was combined with carrier aggregation (CA) technology,and the distribution Q-Learning algorithm were adopted to allocate the channel and power.The simulation results show that the proposed mechanism can maximize the V2I link traversal capacity while ensuring the reliability of the V2I link.…”
Get full text

Real-Time Fire Detection Method Based on Computer Vision for Electric Vehicle Charging Safety Monitoring by Yuchen Gao, Qing Yang, Shiyu Zhang, Dexin Gao

“…Therefore, a target detection model based on the improved YOLOv5 (You Only Look Once) algorithm is proposed for the features generated by lithium battery combustion, using the K-means algorithm to cluster and analyse the target locations within the dataset, while adjusting the residual structure and the number of convolutional kernels in the network and embedding a convolutional block attention module (CBAM) to improve the detection accuracy without affecting the detection speed. …”
Get full text

Research on HMM based link prediction method in heterogeneous network by Rong QIAN, Jianting XU, Kejun ZHANG, Hongyu DONG, Fangyuan XING

“…In order to solve the problem that incomplete mining of structural information and semantic information in heterogeneous networks, a link prediction method combining meta-path-based analysis and hidden Markov model was proposed for link prediction of heterogeneous network.Considering that clustering could effectively capture the structural information of heterogeneous network, the k-means algorithm was improved to obtain the initial clustering center method based on the minimum distance mean square error, and it was applied to the hidden Markov model, first-order cluster hidden markov model (C-HMM⁽¹⁾) link prediction method, and a link prediction method for heterogeneous network with second-order cluster hidden Markov model (C-HMM⁽²⁾) were designed.Further, considering the feature information of the data, a link prediction method called ME-HMM that combined the maximum entropy model and the second-order Markov model was proposed.The experimental results show that the ME-HMM has higher link prediction accuracy than the C-HMM, and the ME-HMM method has better performance than the C-HMM method because it fully considers the feature information of the data.…”
Get full text

Research on the Method of Coke Optical Tissue Segmentation Based on Adaptive Clustering by Huaiguang Liu, Liheng Zhang, Shiyang Zhou, Li Fang

“…The experimental results show that compared with the traditional K-means algorithm, FCM algorithm, and Meanshift algorithm, the adaptive clustering algorithm proposed in this paper is more accurate in the segmentation of various tissue components in COT images, and the accuracy of tissue segmentation reaches 94.3500%.…”
Get full text

A Lightweight Intrusion Detection Method Based on Fuzzy Clustering Algorithm for Wireless Sensor Networks by Hongchun Qu, Libiao Lei, Xiaoming Tang, Ping Wang

“…Our method is highlighted by the fusion of fuzzy c-means algorithm, one-class SVM, and sliding window procedure to effectively differentiate network attacks from abnormal data. …”
Get full text

Regional Spatial Mean of Ionospheric Irregularities Based on K-Means Clustering of ROTI Maps by Yenca Migoya-Orué, Oladipo E. Abe, Sandro Radicella

“…The results obtained could be adapted by appropriate K-means algorithms to a real-time scenario, as has been performed for other applications. …”
Get full text

An Improved Ant Colony Optimization Based on an Adaptive Heuristic Factor for the Traveling Salesman Problem by Pengzhen Du, Ning Liu, Haofeng Zhang, Jianfeng Lu

“…In the AHACO, three main improvements are proposed to improve the performance of the algorithm. First, the k-means algorithm is introduced to classify cities. The AHACO provides different movement strategies for different city classes, which improves the diversity of the population and improves the search ability of the algorithm. …”
Get full text

Healthcare professionals and the public sentiment analysis of ChatGPT in clinical practice by Lizhen Lu, Yueli Zhu, Jiekai Yang, Yuting Yang, Junwei Ye, Shanshan Ai, Qi Zhou

“…This study was divided into five steps: data collection, data cleaning, validation of relevance, sentiment analysis, and content analysis using the K-means algorithm. This study comprised 3130 comments amounting to 1,593,650 words. …”
Get full text

An Inspiration Recommendation System for Automotive Styling Design Based on User Behavior Data and Group Preferences by Wanxin Cai, Mingqing Yang, Li Lin

“…Therefore, a novel inspiration recommendation (IR) system based on multi-level mining of user behavior data is proposed. Firstly, the K-means algorithm is employed to cluster users based on a variety of features. …”
Get full text

Optimizing Cluster Heads for Energy Efficiency in Large-Scale Heterogeneous Wireless Sensor Networks by Yi Gu, Qishi Wu, Nageswara S. V. Rao

“…The results from an extensive set of experiments on a large number of simulated sensor networks illustrate the performance superiority of the proposed solution over the clustering schemes based on k -means algorithm.…”
Get full text

Analyzing the customer purchase data of an online shopping store by data mining: A real case study in Iran by Nima Moradi, Mosayeb Jalilian

“…For clustering methods, the results show that the K-Means algorithm with K=3 is more efficient regarding the average within centroid distance for each cluster. …”
Get full text

Convalescing Cluster Configuration Using a Superlative Framework by R. Sabitha, S. Karthik

“…Data clustering is one such descriptive data mining technique which guides in partitioning data objects into disjoint segments. K-means algorithm is a versatile algorithm among the various approaches used in data clustering. …”
Get full text

A Novel Shape Classification Approach Based on Branch Length Similarity Entropy by Sang-Hee Lee, Cheol-Min Park

“…The methodology consists of two steps: the t-distributed Stochastic Neighbor Embedding (t-SNE) technique, which projects feature vectors into a two-dimensional space, and the k-means algorithm, which groups these points into clusters. …”
Get full text

Spectral clustering-based energy-efficient resource allocation algorithm in heterogeneous cellular ultra-dense network by Xue WANG, Jing LIU, Jiani SUN, Jizhen ZHANG, Zhihong QIAN

“…In order to solve problems of high power consumption, spectrum shortage and low energy efficiency in the ultra-intensive 5G mobile communication scenario, a resource allocation algorithm based on the maximum energy efficiency for the two-layer heterogeneous cellular non-orthogonal multiple access network was proposed.The original NP-hard optimization problem on the downlink communication link of ultra-dense scene was divided into two subproblem, such as frequency resource allocation and power allocation, which became a deterministic constraint optimization problem.The frequency resource allocation scheme of different user groups was obtained by using base station clustering based on the improved k-means algorithm and users grouping based on spectral clustering algorithm.The fraction of energy efficiency optimization was transformed into a solvable continuous convex optimization problem and power distribution was realized by Dinkelbach method, and the Lagrange multiplier iterative algorithm, respectively.Jointly optimize system energy efficiency in terms of base station clustering, user grouping, resource block allocation and power allocation, which minimized the inter-cluster interference and intra-cluster interference of the base station efficiently.The simulation results show that the proposed algorithm is better on energy efficiency and computational efficiency compared with existing algorithms.…”
Get full text

3D WSN clustering routing algorithm based on IHBA optimized fuzzy C-means by Jing YANG, Biyang XU, Weizheng LI, Wei LI, Zhaoqiang ZHU, Tao QIN

“…Aiming at the problem that clustering routing algorithm in a three-dimensional (3D) scene has high energy consumption and short network lifetime, a clustering routing algorithm IFCRA for 3D wireless sensor networks based on improved honey badger algorithm optimized fuzzy C-means was proposed.The network clustering, cluster head election, and data transmission stages were optimized respectively.Firstly, the improved honey badger algorithm was used to optimize the fuzzy C-means algorithm, solving the problem of fuzzy C-means easily falling into local optima, and the topological structure was divided based on the distance and energy characteristics of nodes.Secondly, the optimal cluster head function was constructed by combining the energy and relative distance of nodes within the cluster to balance the cluster head load.Finally, an adaptive transmission mechanism was used to search for relay nodes, and the optimal transmission path function was constructed by combining node distance and energy.The improved honey badger algorithm was used to optimize the data transmission energy consumption.Simulation results show that IFCRA has reasonable clustering, balanced energy consumption, and long lifespan in a 3D heterogeneous scene.…”
Get full text

A hybrid approach to conangtent generation based on user experience using generative AI elements by Mohammad Javad Shayegan, Hossein Hosseinpour

“…Results show that the enhanced K-means algorithm paired with GPT-3.5 Turbo achieved the best performance, yielding a Perplexity score of 3.476. …”
Get full text

BEHAVIORAL SEGMENTATION OF CUSTOMERS: RFM MODEL AND K-MEANS APPLICATION by Thanh L.T.P., Liaw S.-Y., Hien N.T.

“…The main goal of this study is to propose a customer segmentation approach based on RFM (recency, frequency, and monetary) model and K-means algorithm, and then use the one-way ANOVA test to explore the difference in recency, frequency, and monetary attributes in customers of different ages. …”
Get full text

Determination of Renewable Energy Growth Using Cluster Analysis and Multi-Criteria Decision-Making Methods by Eşiyok Sevgi, Antmen Z. Figen

“…In the study, 38 countries were divided into three clusters for each year with the 13 variable K-means algorithm for the years 2018–2022. The criteria of the countries in the determined clusters were weighted with the CRITIC (Criteria Importance Through Intercriteria Correlation) method for each year during the study period. …”
Get full text