K-means is still a popular clustering algorithm and active research area. The research is majorly focused at improving efficiency and effectiveness of the method. This paper proposes combined approach of a ranked init...
详细信息
ISBN:
(纸本)9781509020287
K-means is still a popular clustering algorithm and active research area. The research is majorly focused at improving efficiency and effectiveness of the method. This paper proposes combined approach of a ranked initialization and normalization of data values with k-means. Three variations of a score based initialization approach is proposed. Experiments are performed on normalized data to prove the superiority of the proposed algorithm.
Traffic status evaluation is the most important part of Intelligent Traffic System (ITS). However, nowadays, most models used to identify traffic performance are either based on traffic flow theories or machine learni...
详细信息
ISBN:
(纸本)9781509011926
Traffic status evaluation is the most important part of Intelligent Traffic System (ITS). However, nowadays, most models used to identify traffic performance are either based on traffic flow theories or machine learning algorithms and data mining algorithms. On that base, a new traffic performance evaluation model which combines traffic theory and machine learning algorithm is presented. In order to establish such model, characteristics of traffic data is discussed at first. The result of statistical analysis of traffic data shows that the probability density of a large amount of traffic data could be synthetized by a few Gaussian distributions. Therefore, a suitable traffic status evaluation model which combines Gaussian mixture model (GMM) and linear regression (LR) is proposed in this paper. In this model, firstly, by using Gaussian mixture model, historical traffic data will be clustered into three groups with respect to three traffic states, which are free flow, quasi- free flow and crowd flow. Then, the data with respect to crowd flow will be used to train linear regression model, and linear regression model will generate a straight line which analogues the line J in Kerner's traffic theory to classify crowd flow into two parts, which are related to synchronized flow and traffic jam. Through the two steps, traffic performance is described into four states, which are free flow, quasi- free flow, synchronized flow and traffic jam. The result of designed experiment shows that the model could evaluate traffic performance effectively.
The granularity partition for functional modules is a fundamental research topic in robot distributed control technology. How to evaluate the module partition scheme with different granularity, and then obtain the opt...
详细信息
ISBN:
(纸本)9781509041022
The granularity partition for functional modules is a fundamental research topic in robot distributed control technology. How to evaluate the module partition scheme with different granularity, and then obtain the optimum scheme is the urgent problem. In this paper, we proposed a novel evaluation strategy for the granularity partition of functional modules in robotic system using RTM as control platform based on D-S evidence theory. The fuzzy clustering algorithm is primarily used to get the collection of granularity partition schemes for RT Components encapsulated by the platform of OpenRTM. As the two source of evidence, the indices of cohesion and coupling for the robotic system are achieved to measure the degree of module independence by analyzing the correlation matrix of RT Components. Then the Dempster's combination rule and the priority method for utility intervals are applied to obtain the optimal partition granularity. In the end, the effectiveness and progressiveness of the novel evaluation strategy are verified by applying it to the robotic 3D mapping system.
Indoor positioning technologies has boomed recently because of the growing commercial interest in indoor location-based service (ILBS). Due to the absence of satellite signal in Global Navigation Satellite System (GNS...
详细信息
Indoor positioning technologies has boomed recently because of the growing commercial interest in indoor location-based service (ILBS). Due to the absence of satellite signal in Global Navigation Satellite System (GNSS), various technologies have been proposed for indoor applications. Among them, Wi-Fi fingerprinting has been attracting much interest from researchers because of its pervasive deployment, flexibility and robustness to dense cluttered indoor environments. One challenge, however, is the deployment of Access Points (AP), which would bring a significant influence on the system positioning accuracy. This paper concentrates on WLAN based fingerprinting indoor location by analyzing the AP deployment influence, and studying the advantages of coordinate-based clustering compared to traditional RSS-based clustering. A coordinate-based clustering method for indoor fingerprinting location, named Smallest-Enclosing-Circle-based (SEC), is then proposed aiming at reducing the positioning error lying in the AP deployment and improving robustness to dense cluttered environments. All measurements are conducted in indoor public areas, such as the National Center For the Performing Arts (as Test-bed 1) and the XiDan Joy City (Floors 1 and 2, as Test-bed 2), and results show that SEC clustering algorithm can improve system positioning accuracy by about 32.7% for Test-bed 1, 71.7% for Test-bed 2 Floor 1 and 73.7% for Test-bed 2 Floor 2 compared with traditional RSS-based clustering algorithms such as K-means.
In this paper, two phase clustering algorithm is adopted to identify and track event. The first phase clustering is incremental clustering algorithm, and the new event will be identified. The second phase clustering i...
详细信息
In this paper, two phase clustering algorithm is adopted to identify and track event. The first phase clustering is incremental clustering algorithm, and the new event will be identified. The second phase clustering is refined clustering algorithm, and the new event will be group and tracking. Experimental result shows that event identification and tracking using two phase clustering algorithm is effective.
Improved network lifetime without much increase in the cost contributes popularity to heterogeneous wireless sensor networks. clustering algorithms designed to utilize advantage of heterogeneity of nodes allow these n...
详细信息
Improved network lifetime without much increase in the cost contributes popularity to heterogeneous wireless sensor networks. clustering algorithms designed to utilize advantage of heterogeneity of nodes allow these nodes to be cluster head more times than normal nodes to have load balanced network. Cluster head selection is pivotal for the performance of clustering algorithms as cluster quality in terms of communication distance depends upon the location of selected head in the cluster. Work of this paper analyzes the effect of location of heterogeneous nodes on the performance of clustering algorithms. Worst case, average (random) case and best case for location of heterogeneous nodes are considered for analyzing the effect on the performance of clustering algorithms.
Image segmentation problem is a fundamental task and process in computer vision and image processing applications. It is well known that the performance of image segmentation is mainly influenced by two factors: the s...
详细信息
Image segmentation problem is a fundamental task and process in computer vision and image processing applications. It is well known that the performance of image segmentation is mainly influenced by two factors: the segmentation approaches and the feature presentation. As for image segmentation methods, clustering algorithm is one of the most popular approaches. However, most current clustering-based segmentation methods exist some problems, such as the number of regions of image have to be given prior, the different initial cluster centers will produce different segmentation results and so on. In this paper, we present a novel image segmentation approach based on DP clustering algorithm. Compared with the current methods, our method has several improved advantages as follows: 1) This algorithm could directly give the cluster number of the image based on the decision graph; 2) The cluster centers could be identified correctly; 3) We could simply achieve the hierarchical segmentation according to the applications requirement. A lot of experiments demonstrate the validity of this novel segmentation algorithm.
clustering on multiple manifolds serves as an analysis of the data lying on multiple manifolds. The smoothness and local linearity of data samples are utilized to define the local linear degree which is motivated by P...
详细信息
clustering on multiple manifolds serves as an analysis of the data lying on multiple manifolds. The smoothness and local linearity of data samples are utilized to define the local linear degree which is motivated by Principal Component Analysis (PCA) and Depth First Search (DFS). Then, Multiple Manifolds clustering (LMMC) is proposed on the base of the Local Linear Analysis (LLA) via this definition and neighbor-growing algorithm, which are especially effective under the condition of interactions. Instead of addressing problems of complex optimization and K-means operation, LIVIIVIC is simple and efficient compared with traditional manifold clustering. The algorithm can achieve superior performance on complex subspace and manifolds clustering datasets. Meanwhile, comparative experiments are given to show the effectiveness and efficiency of this algorithm.
We presented in this paper a usage scenario in which cultural resources in a public context, items on display in a historical museum for instance, should be recommended to groups of visitors in response to their inter...
详细信息
ISBN:
(纸本)9783319486741;9783319486734
We presented in this paper a usage scenario in which cultural resources in a public context, items on display in a historical museum for instance, should be recommended to groups of visitors in response to their interest (or preferences), thus conserving computational resources and reducing network traffic. Motivated by the scenario, we set out to design and implement a group recommender system, Museum Guides for Groups (MGG), that provides visitors to a museum with a sequence of items of interest by efficiently clustering visitors of similar user profiles into groups and computing recommendations for each group. Our work in progress was reported, focusing on the system design and the selection of an appropriate clustering algorithm for dividing visitors. We evaluated the efficiency of three candidate clustering techniques, including the bisecting K-Means, DBSCAN, and improved CURE, using the MovieLens dataset with 1M ratings.
A relay-based clustering algorithm (RBC) is proposed in this paper for heterogeneous energy wireless sensor networks. The "Relay" mechanism is introduced to relay the "Cluster Head" position to its...
详细信息
A relay-based clustering algorithm (RBC) is proposed in this paper for heterogeneous energy wireless sensor networks. The "Relay" mechanism is introduced to relay the "Cluster Head" position to its successor by considering the nodes' residual energy. In RBC, only the nodes with highest residual energy in each cluster will be assigned as cluster heads. RBC is easy to implement, and could work in the distributed way. Simulation results show that this new scheme provides higher network performance than the classical and the improved clustering algorithms in heterogeneous networks, in terms of longer lifetime and the amount of effective messages of the network.
暂无评论