K-medoids Clustering Algorithm Research Articles

Сравнение методик кластеризации частично упорядоченных множеств

In this paper, we compare three approaches of clustering partial ordered subsets of a set of items. First approach was k-medoids clustering algorithm with distance function based on Levenshtein distance. The second approach was k-means algorithm with cosine distance as distance function after vectorization of partial orders. And the third one was k-medoids algorithm with Kendall's tau as a distance function. We use Adjusted Rand Index as a measure of quality of clustering and find out that clustering with all three methods get stable results when variance of number of items ranked is high. Vectorization of partial orders get best results if number of items ranked is low.

The Improvement of K-Medoids Clustering Algorithm under Semantic Web

K-medoids clustering algorithm is an efficient algorithm in classifying cluster categories. Based on algorithm analysis, this paper first improves the selection of K center point and then sets up a web model of ontology data set object with the aim of demonstrating through experiment evaluation that the improved algorithm can greatly enhance the accuracy of clustering results under semantic web.

A Keypoints of Fingers Detection Method Based on K-Medoids Clustering and K-Vector Algorithm

The paper presents a calculation method to get the probable key points in hands outline. First, use K-vector to calculate the K-slope of all points in outline and get a series of peak point sets. Then, cluster the peak point sets which have been obstained by K-medoids algorithm and get the location of every fingertips and probable points; At last, distinguish the fingertips and webbed points by vector cross product operation. The experiment shows that this method achieve the precise positionong of the fingertips and finger webbed points.

A K-Medoids Clustering Algorithm Based on Particle Swarm Algorithm with Simulated Annealing

Be aimed at the problems that K-medoids algorithm is easy to fall into the local optimal value and basic particle swarm algorithm is easy to fall into the premature convergence, this paper joins the Simulated Annealing (SA) thought and proposes a novel K-medoids clustering algorithm based on Particle swarm optimization algorithm with simulated annealing. The new algorithm combines the quick optimization ability of particle swarm optimization algorithm and the probability of jumping property with SA, and maintains the characteristics that particle swarm algorithm is easy to realize, and improves the ability of the algorithm from local extreme value point. The experimental results show that the algorithm enhances the convergence speed and accuracy of the algorithm, and the clustering effect is better than the original k-medoids algorithm.

The K-Medoids Clustering Algorithm with Membrane Computing

The K-medoids clustering algorithm is realized by a P system in this paper. Because the membrane system has great parallelism and lower computational time complexity, it is suitable for solving combinatorial problems like the clustering problem. A P system with all the rules to solve the K-medoids algorithm was constructed. The specific P system is associated with the dissimilarity matrix between n objects. This system can get one possible classifications in a non-deterministic way. Through example test, it is appropriate for cluster analysis. This is a new attempt in applications of membrane system and it provides new ideas and methods for cluster analysis. DOI: http://dx.doi.org/10.11591/telkomnika.v11i4.2383

A GIS (geographical information system)-based spatial data mining approach for optimal location and capacity planning of distributed biomass power generation facilities: A case study of Tumkur district, India

This paper primarily intends to develop a GIS (geographical information system)-based data mining approach for optimally selecting the locations and determining installed capacities for setting up distributed biomass power generation systems in the context of decentralized energy planning for rural regions. The optimal locations within a cluster of villages are obtained by matching the installed capacity needed with the demand for power, minimizing the cost of transportation of biomass from dispersed sources to power generation system, and cost of distribution of electricity from the power generation system to demand centers or villages. The methodology was validated by using it for developing an optimal plan for implementing distributed biomass-based power systems for meeting the rural electricity needs of Tumkur district in India consisting of 2700 villages. The approach uses a k-medoid clustering algorithm to divide the total region into clusters of villages and locate biomass power generation systems at the medoids. The optimal value of k is determined iteratively by running the algorithm for the entire search space for different values of k along with demand–supply matching constraints. The optimal value of the k is chosen such that it minimizes the total cost of system installation, costs of transportation of biomass, and transmission and distribution. A smaller region, consisting of 293 villages was selected to study the sensitivity of the results to varying demand and supply parameters. The results of clustering are represented on a GIS map for the region.

Identifying influential nodes in complex networks with community structure

It is a fundamental issue to find a small subset of influential individuals in a complex network such that they can spread information to the largest number of nodes in the network. Though some heuristic methods, including degree centrality, betweenness centrality, closeness centrality, the k-shell decomposition method and a greedy algorithm, can help identify influential nodes, they have limitations for networks with community structure. This paper reveals a new measure for assessing the influence effect based on influence scope maximization, which can complement the traditional measure of the expected number of influenced nodes. A novel method for identifying influential nodes in complex networks with community structure is proposed. This method uses the information transfer probability between any pair of nodes and the k-medoid clustering algorithm. The experimental results show that the influential nodes identified by the k-medoid method can influence a larger scope in networks with obvious community structure than the greedy algorithm without reducing the expected number of influenced nodes.

Medoid selection from sub‐tree leaf nodes for k‐medoid clustering‐based hierarchical template tree construction

A method to construct a hierarchical template tree for pedestrian contour detection by iteratively applying a k-medoid clustering algorithm from the lowest level to the highest level was recently proposed and received much attention. Analysed here is the limitation of the method using lower level medoids as points of the next higher level, and proposed is a method of selecting a medoid from the leaf nodes of sub-trees corresponding to the lower level medoids.

Clustering of Patients’ Trajectories with an Auto-Stopped Bisecting K-Medoids Algorithm

Nowadays, long wait, cancellations and resource overload frequently occur in healthcare, especially in those sectors related to the patients passing through the operating theatre in both United States and the European Union. Since more and more hospitals seek to develop the overall patient pathways instead of the effectiveness of “isolated” departments, the most important work has been defining suitable patient groups for employing process management and simulation tools developed in the recent decades. In this study, we proposed a data mining method, an auto-stopped Bisecting K-Medoids clustering algorithm, to classify patients into groups with homogenous trajectories. This method classifies the patient trajectories with two stages. At the first stage, patients are classified by the complexity of outpatient visits; afterwards, the groups obtained at the first stage are further classified by the original information of the trajectories where all medical appointments including outpatient ones are taken into account. By using a real data set collected from a medium-size Belgian hospital, we demonstrate how the proposed approach works and examine which kinds of trajectories are grouped into the same clusters. According to the experimental results, the proposed method can be used to classify patients into manageable groups with homogenous trajectories, which can be used as a base for the process modelling techniques and simulation tools.

An Improved K-Medoids Clustering Algorithm

Because of the traditional K-medoids clustering algorithm the initial clustering center sensitive, the global search ability is poor, easily trapped into local optimal and slow convergent speed; therefore, this article proposes an improved K-medoids clustering algorithm. Differential evolution is a kind of heuristic global search technology population, has strong robustness. Combined with K-medoids clustering algorithm efficiency and the global optimization ability of DE algorithm, not only can effectively overcome the detects of the K-medoids clustering algorithm, but also can raise the global search capability, short the convergence time, effectively improve the clustering quality. Finally, the algorithm is verified stability and robustness by simulation.

Automatic machine embroidery image color analysis system. Part I: Using Gustafson-Kessel clustering algorithm in embroidery fabric color separation

This series of studies aims to propose the automatic machine embroidery image color analysis system to solve the problem of lack of manpower for machine embroidery fabric drafting and to shorten drafting time. The studies included three parts: (1) machine embroidery image color separation, (2) search of repeated pattern images, (3) machine embroidery color analysis system integration. This study aimed to find the optimal clustering algorithm and cluster validity indices for the automatic color separation process of machine embroidery fabric drafting in order to shorten drafting time. To improve image quality for computer analysis, this study used the color hybrid median filter to filter noise and the color bilateral filter to smoothen fabric and embroidery texture for subsequent color separation. By extracting the color a* component and b* component of the machine embroidery image in CIE L*a*b* color system, this study used the Gustafson-Kessel clustering algorithm for color separation. The Gustafson-Kessel clustering algorithm in machine embroidery image color separation can improve color separation accuracy, and its result is compared with that of the clustering algorithms commonly used in the color separation of color images. This study implemented the chromatography of the color separation results, and used the cluster validity indices to prove that the application of Gustafson-Kessel clustering algorithm in the machine embroidery image color separation system has better results than K-means, K-medoid, fuzzy C-means (FCM), and self-organizing map (SOM) clustering algorithm. The results meet the classifications as expected by human eyes.

Recognition of Hand Written Kannada Numerals Using K-Medoids

Data Clustering is considered as an interesting approach for finding similarities in data and putting similar data into groups. Clustering partitions a data set into several groups such that the similarity within a group is larger than that among groups. This paper explores the cluster-based classification scheme in the context of recognition of handwritten Kannada numerals. In this paper, K-Medoids clustering algorithm is being used for the classification. The features used for the classification are obtained from the directional chain code information of the contour points of the numerals. The proposed algorithm is experimented on nearly 1000 samples of handwritten Kannada numerals and obtained 97% of recognition accuracy.

Single-document summarization based on semantics

Single-document summarization goals to create a compressed summary while retaining the theme of the original document.Many approaches use statistics and machine learning techniques to extract sentences from a document.Because single document has limited information,the main approaches are of no effect.Therefore,a new single-document summarization framework based on semantics was proposed.First,the sentence-sentence similarity was calculated.After that modified K-Medoids clustering algorithm was used to cluster the sentences.Finally,the most informative sentence was chosen from each cluster to form the summary.The experimental results demonstrate the improvement of the summary quality by using semantics information.

Computational Complexity between K-Means and K-Medoids Clustering Algorithms for Normal and Uniform Distributions of Data Points

Problem statement: Clustering is one of the most important research ar eas in the field of data mining. Clustering means creating groups of ob jects based on their features in such a way that th e objects belonging to the same groups are similar an d those belonging to different groups are dissimila r. Clustering is an unsupervised learning technique. T he main advantage of clustering is that interesting patterns and structures can be found directly from very large data sets with little or none of the background knowledge. Clustering algorithms can be applied in many domains. Approach: In this research, the most representative algorithms K-Mean s and K-Medoids were examined and analyzed based on their basic approach. The best algorithm i n each category was found out based on their performance. The input data points are generated by two ways, one by using normal distribution and another by applying uniform distribution. Results: The randomly distributed data points were taken as input to these algorithms and clusters are found ou t for each algorithm. The algorithms were implemented using JAVA language and the performance was analyzed based on their clustering quality. The execution time for the algorithms in each category was compar ed for different runs. The accuracy of the algorith m was investigated during different execution of the program on the input data points. Conclusion: The average time taken by K-Means algorithm is greater than the time taken by K-Medoids algorithm for both the case of normal and uniform distributions. The r esults proved to be satisfactory.

Clustering of patient trajectories with an auto-stopped bisecting K-Medoids algorithm

Abstract Nowadays, more and more hospitals seek to employ well-developed process management and simulation tools in the healthcare organizations to develop the overall patient pathways. Since it is extremely important to define suitable patient groups for constructing process or simulation models, we proposed a sequence mining method, an auto-stopped Bisecting K-Medoids clustering algorithm, to classify patients into groups with homogeneous trajectories within two stages. At the first stage, patients are classified according to the complexity of the care process. Afterwards, groups obtained at the first stage are further classified with the similarity of the trajectories. The proposed approach was executed with a real data set from a medium-size hospital. According to the experimental results, this method can be used to classify patients into manageable groups, where the most frequent trajectories might be extracted to validate a process modelling technique. In addition, data extracted from those groups could be used to feed our simulation models.

A distributed dynamic clustering algorithm for wireless sensor networks

This paper proposes a distributed dynamic k-medoid clustering algorithm for wireless sensor networks (WSNs), DDKCAWSN. Different from node-clustering algorithms and protocols for WSNs, the algorithm focuses on clustering data in the network. By sending the sink clustered data instead of practical ones, the algorithm can greatly reduce the size and the time of data communication, and further save the energy of the nodes in the network and prolong the system lifetime. Moreover, the algorithm improves the accuracy of the clustered data dynamically by updating the clusters periodically such as each day. Simulation results demonstrate the effectiveness of our approach for different metrics.