clustering algorithm, which is a statistical analysis method for research in classifications, plays an important role in data mining *** algorithm based on similarity, and is easy to combine with other methods in opti...
详细信息
clustering algorithm, which is a statistical analysis method for research in classifications, plays an important role in data mining *** algorithm based on similarity, and is easy to combine with other methods in optimization. In this review, signal clustering algorithm is introduced by discussing of the clustering parametric in different signal clustering *** order to develop traditional algorithm, we introduce a series of improvement,development and application of the methods in recent years. Finally, we make an outlook of the future direction and content of the research in this field.
Our group recommender system was targeted at a scenario that requires the adoption of group recommendation techniques to conserve computational resources. The profile aggregation strategy was used in our work to imple...
详细信息
ISBN:
(纸本)9781509048724
Our group recommender system was targeted at a scenario that requires the adoption of group recommendation techniques to conserve computational resources. The profile aggregation strategy was used in our work to implement this group recommendation system. Key to our work, user clustering is also the first step of our work. The accuracy of user clustering could be improved once we processed the data set by the SVD (Singular Value Decomposition) algorithm. The data set in this experiment was 1M user rating data available from MovieLens. According to the metric we used for evaluating the quality of user clustering, we discovered that the bisecting K-means algorithm outperformed the DBSCAN algorithm on the dataset within the experimental settings.
clustering in vehicular ad hoc networks (VANETs) is a challenging issue due to the highly dynamic vehicle mobility and frequent communication disconnections problems. Recent years' research have proven that mobili...
详细信息
ISBN:
(纸本)9781509002245
clustering in vehicular ad hoc networks (VANETs) is a challenging issue due to the highly dynamic vehicle mobility and frequent communication disconnections problems. Recent years' research have proven that mobility-based clustering mechanisms considering speed, moving direction, position, destination and density, were more effective in improving cluster stability. In this paper, we propose a new mobility-based and stability-based clustering algorithm (MSCA) for urban city scenario, which makes use of vehicle's moving direction, relative position and link lifetime estimation. We evaluate the performance of our proposed algorithm in terms of changing maximum lane speed and traffic flow rate. Our proposed algorithm performs well in terms of average cluster head lifetime and average number of clusters.
In allusion to increase the speed and accuracy rate of clustering algorithm, the paper proposes mixed clustering algorithm with artificial fish swarm and improved K-means. Firstly, it leads artificial fish swarm algor...
详细信息
In allusion to increase the speed and accuracy rate of clustering algorithm, the paper proposes mixed clustering algorithm with artificial fish swarm and improved K-means. Firstly, it leads artificial fish swarm algorithm to clustering algorithm and proposes artificial fish swarm clustering algorithm. Secondly, it improves traditional K-means algorithm and gives improved K-means algorithm. Finally, it gets mixed clustering algorithm with artificial fish swarm and K-means.
Vehicular Ad Hoc Networks (VANET) is a wireless mobile ad hoc networks established on the inter-vehicle communication. Due to the special characteristics and the restrictions of roads, VANET shows characteristics of u...
详细信息
Vehicular Ad Hoc Networks (VANET) is a wireless mobile ad hoc networks established on the inter-vehicle communication. Due to the special characteristics and the restrictions of roads, VANET shows characteristics of uneven nodes density, fast moving, high dynamic topology. So it is difficult to establish a stable link between nodes, and the reliability of data transmission declines rapidly with the increase of hops. Therefore, ensuring data reliability and rapid distribution is always a difficult point in the VANET research. Through analysis of vehicular ad hoc networks environment, we propose the node connectivity and the connectivity strength. By dividing the road environment into the segment area and the intersection area, respectively calculating and predicting the connectivity according to the road environment, we propose the zone based adaptive clustering algorithm (ZACA).
In order to explore the main influencing factor of safety management of construction project proposed a based on clustering algorithm of safety management factors analysis method,and the main factors of influencing th...
详细信息
In order to explore the main influencing factor of safety management of construction project proposed a based on clustering algorithm of safety management factors analysis method,and the main factors of influencing the construction safety management through the practical *** a factor of 16 questionnaire as the basic data,using multivariate statistical analysis method of factor analysis method analyzes the results of the survey,six main affecting factors of construction safety and project management,and then analyzes the main factor *** the analysis of the factors of the characteristic value of the stone figure and factor analysis of the tree,by factor analysis and cluster analysis of the results summed up the impact of construction safety management of the five main areas.
Affinity Propagation(AP) algorithm is a relatively new clustering algorithm that can handle large datasets to obtain more satisfactory *** paper introduces a detection mechanism for application-layer DDoS attack by us...
详细信息
ISBN:
(纸本)9781510835368
Affinity Propagation(AP) algorithm is a relatively new clustering algorithm that can handle large datasets to obtain more satisfactory *** paper introduces a detection mechanism for application-layer DDoS attack by using AP *** this detection strategy,we first extract some features from normal users' ***,we cluster these normal users' sessions by AP algorithm to get K ***,we use these models to detect application-layer DDo S attacks.
Compared to traditional fuel vehicles, the structure of pure electric vehicles (BEVs) and the actual driving behaviour of users have changed. Therefore, the original durability evaluation conditions of traditional fue...
详细信息
Compared to traditional fuel vehicles, the structure of pure electric vehicles (BEVs) and the actual driving behaviour of users have changed. Therefore, the original durability evaluation conditions of traditional fuel vehicles cannot fully cover the use of new energy vehicles. In the past, the determination of durability targets was mainly based on user data collection, but this work required a lot of manpower and material resources to meet the engineering requirements. In this paper, the fuzzy clustering method is used to mine the user trajectory to obtain the user-based endurance target of pure electric vehicle, and then according to the durability target, the particle swarm method is used to correlate the user behaviour and the proving ground, and the proving ground test method of electric drive system is developed. Studies have shown that user data mining methods can obtain more user information, so as to better formulate durable target close to users. The particle swarm algorithm can improve the simulation correlation accuracy and reduce the iteration time, which shortens the simulation iteration time by more than 80% compared with polynomials. The test acceleration ratio of 7:1 in relation to user behaviour with the proving ground. During the durability test of the electric drive system of pure electric vehicles, it is found that the test specification can well reflect the user's motor operation during actual driving.
K-means clustering is usually used in image segmentation due to its simplicity and rapidity. However, K-means is heavily dependent on the initial number of clusters and easily falls into local falls into local optimum...
详细信息
K-means clustering is usually used in image segmentation due to its simplicity and rapidity. However, K-means is heavily dependent on the initial number of clusters and easily falls into local falls into local optimum. As a result, it is often difficult to obtain satisfactory visual effects. As an evolutionary computation technique, particle swarm optimization (PSO) has good global optimization capability. Combined with PSO, K-means clustering can enhance its global optimization capability. But PSO also has the shortcoming of easily falling into local optima. This study proposes a new image segmentation algorithm called dynamic particle swarm optimization and K-means clustering algorithm (DPSOK), which is based on dynamic particle swarm optimization (DPSO) and K-means clustering. The calculation methods of its inertia weight and learning factors have been improved to ensure DPSOK algorithm keeping an equilibrium optimization capability. Experimental results show that DPSOK algorithm can effectively improve the global search capability of K-means clustering. It has much better visual effect than K-means clustering in image segmentation. Compared with classic particle swarm optimization K-means clustering algorithm (PSOK), DPSOK algorithm has obvious superiority in improving image segmentation quality and efficiency. (C) 2015 Elsevier GmbH. All rights reserved.
Advances in recent techniques for scientific data collection in the era of big data allow for the systematic accumulation of large quantities of data at various data-capturing sites. Similarly, exponential growth in t...
详细信息
Advances in recent techniques for scientific data collection in the era of big data allow for the systematic accumulation of large quantities of data at various data-capturing sites. Similarly, exponential growth in the development of different data analysis approaches has been reported in the literature, amongst which the K-means algorithm remains the most popular and straightforward clustering algorithm. The broad applicability of the algo-rithm in many clustering application areas can be attributed to its implementation simplic-ity and low computational complexity. However, the K-means algorithm has many challenges that negatively affect its clustering performance. In the algorithm's initialization process, users must specify the number of clusters in a given dataset apriori while the ini-tial cluster centers are randomly selected. Furthermore, the algorithm's performance is susceptible to the selection of this initial cluster and for large datasets, determining the optimal number of clusters to start with becomes complex and is a very challenging task. Moreover, the random selection of the initial cluster centers sometimes results in minimal local convergence due to its greedy nature. A further limitation is that certain data object features are used in determining their similarity by using the Euclidean distance metric as a similarity measure, but this limits the algorithm's robustness in detecting other cluster shapes and poses a great challenge in detecting overlapping clusters. Many research efforts have been conducted and reported in literature with regard to improving the K-means algorithm's performance and robustness. The current work presents an overview and tax-onomy of the K-means clustering algorithm and its variants. The history of the K-means, current trends, open issues and challenges, and recommended future research perspectives are also discussed.(c) 2022 Elsevier Inc. All rights reserved.
暂无评论