Although many algorithms have been proposed for the camera-based detection of road features (such as road markings, curbstones and road borders), truly contextual or relational information between the detections is ra...
详细信息
ISBN:
(纸本)9781509018901
Although many algorithms have been proposed for the camera-based detection of road features (such as road markings, curbstones and road borders), truly contextual or relational information between the detections is rarely used. This is all the more surprising, since a lot of potential remains unused, regarding outlier rejection or compensating detection failures, multiple detections, misclassification or fragmentation. The aim of this paper is to present an approach that is suitable for such a task in both online and offline applications as a post-processing step after the actual detection and classification step. This is achieved by adapting a perception-based line-clustering algorithm that groups the pre-classified road features based on their relations and assigns them a final class. The grouped features are then fused to form continuous lines instead of individual dashes or fragmented lines. The evaluation on a 10 km drive in both rural and urban environment, as well as an online test on a short highway driving sequence shows that this approach is very well capable to increase the performance of road feature detection at a low computational cost.
A Wireless Sensor Network (WSN) is composed of distributed sensors with limited processing capabilities and energy restrictions. These unique attributes pose new challenges amongst which prolonging the WSN lifetime is...
详细信息
ISBN:
(纸本)9781509018598
A Wireless Sensor Network (WSN) is composed of distributed sensors with limited processing capabilities and energy restrictions. These unique attributes pose new challenges amongst which prolonging the WSN lifetime is one of the most important. clustering is an energy efficient routing technique that has been widely applied to report data from the WSN nodes to a centralised Base Station. A plethora of different clustering protocols have been proposed. Some protocols are based on equal-sized clusters while others use clusters of unequal size. Some others make use of rotation techniques to reduce the amount of cluster head elections. When different clustering approaches are presented different simulation settings are used. In this paper we perform a comparison study of HEED based clustering protocols that are HEED, UHEED, RUHEED and a novel variation of R-HEED that is ER-HEED. We have considered the same network model, the same energy consumption model and we have compared the lifetime of the protocols by considering various case studies. Our comparison study shows that the selection of the protocol to be used depends on the case study and the WSN lifetime measure that is considered.
Users of search engines are fond of accurate and fast results. Web is now overloaded with lots of pages or documents dealing with same topic. Thus they are often forced to surf through the large and irrelevant set of ...
详细信息
ISBN:
(纸本)9781467394178
Users of search engines are fond of accurate and fast results. Web is now overloaded with lots of pages or documents dealing with same topic. Thus they are often forced to surf through the large and irrelevant set of results. This has forced the IR community to explore such document clustering techniques capable of providing fast and accurate results. Even after being a very effective solution, clustering is yet not deployed on the major search engines. This paper will articulate the requirements of Web Document clustering and reports on the clustering methods belonging in this domain. The focus of ours is on; these methods create their clusters based on the characters or individual terms rather than showcasing them as a single phrase with a meaning and sequence of words. Paper will be reporting general term based and phrase based techniques and will provide conclusion based on their individual efficiency to work with their key methods.
The article presents a particular comparison of selected clustering algorithms of data obtained by interferometrie methods using artificial neural networks. For the purposes of the experiment original data from Szczec...
详细信息
ISBN:
(纸本)9781509025190
The article presents a particular comparison of selected clustering algorithms of data obtained by interferometrie methods using artificial neural networks. For the purposes of the experiment original data from Szczecin Port have been tested. For collecting data authors used the interferometric sonar system GeoSwath Plus 250 kHz. GeoSwath Plus offers very efficient simultaneous swath bathymetry and side scan seabed mapping. During the use of Kohonen's algorithm, the network, during learning, use the Winner Take All rule and Winner Take Most rule. The parameters of the tested algorithms were maintained at the level of default. During the research several populations were generated with number of clusters equal 9 for data gathered from the area of 100m 2 . In the subsequent step statistics were calculated and outcomes were shown as spatial visualization and in tabular form.
The amount of unstructured text data available is growing exponentially due to the proliferation of digital information such as emails, text messages, blogs, social media posts, and product reviews. For users of e-com...
详细信息
ISBN:
(纸本)9781509006632
The amount of unstructured text data available is growing exponentially due to the proliferation of digital information such as emails, text messages, blogs, social media posts, and product reviews. For users of e-commerce websites such as Amazon, navigating thousands of reviews before buying a product can be a daunting task. Unsupervised machine learning techniques can be used to automatically analyze preprocessed data from these websites in order to provide consumers with an improved user experience before purchasing a product. In this work, we leverage two flat clustering algorithms on Amazon review data: K-means and Peak-searching to perform clustering of product reviews based on topic. The experimental results show that K-means clustering performs better than Peak-searching clustering in terms of grouping similar reviews based on topics.
Fuzzy clustering is an alternative method to conventional or hard clustering algorithms, which makes partitions of data containing similar subjects. The tendency of adopting machine learning, big data science, cloud c...
详细信息
ISBN:
(纸本)9781509052646
Fuzzy clustering is an alternative method to conventional or hard clustering algorithms, which makes partitions of data containing similar subjects. The tendency of adopting machine learning, big data science, cloud computation in various industries depends on unsupervised learning on data structures to tell the story about consumers' behavior, fraud detection, and market segmentation. Fuzzy clustering contrasts with hard clustering by its nonlinear nature and discipline of flexibility in grouping massive data. It provides more accurate and close-to-nature solutions for partitions and herein implies more possibility of solutions for decision-making. In the specific matter of computation, fuzzy clustering has its roots in fuzzy logic and indicates the likelihood or degrees of one data point belonging to more than one group. This paper focuses on the study of models of fuzzy clustering in various cases. Uniquely designed algorithms enhance the accuracy of outcomes and are worth studying to assist future work. In some case scenarios, modeling processes are data-driven and place emphasis on the distances between points and new centers of clusters. In some other cases, which aim at market segmentation or evaluation of patients by healthcare records, membership degree is a key element in the algorithm. This paper surveys a wide-range of research that has well-designed mathematic models for fuzzy clustering, some of which include genetic algorithms and neural networks. The last section introduces open sources of Python and displays sample results from hands-on practice with these packages.
In this study, several possibilistic clustering methods are proposed based on noise clustering. The proposed methods are motivated by the fact that conventional possibilistic clustering methods do not correspond with ...
详细信息
ISBN:
(纸本)9781509026791
In this study, several possibilistic clustering methods are proposed based on noise clustering. The proposed methods are motivated by the fact that conventional possibilistic clustering methods do not correspond with noise clustering methods in entropy-regularized situations, whereas these methods do correspond in Bezdek-type fuzzified situation.
The aim of writing this paper is to provide a detailed, in order description and analysis of the often used and important algorithms of clustering with focus on the recent advances, and to provide an extensive compari...
详细信息
ISBN:
(纸本)9781509002115
The aim of writing this paper is to provide a detailed, in order description and analysis of the often used and important algorithms of clustering with focus on the recent advances, and to provide an extensive comparison of these algorithms in terms of their complexities and applications.
clustering algorithms has been widely used in Ad hoc network with its ability to construct network quickly,conveniently and flexibly and without the need of default network *** this paper,Firstly,some shortcomings of ...
详细信息
ISBN:
(纸本)9781510823808
clustering algorithms has been widely used in Ad hoc network with its ability to construct network quickly,conveniently and flexibly and without the need of default network *** this paper,Firstly,some shortcomings of the typical algorithm AOW(adaptive on-demand weighted algorithm) are introduced and ***,we discusses the calculating method of nodes weights with perceptron algorithm and the cluster heads selecting process with modified algorithms based on original AOW to meet system ***,a adaptive on-demand weighted algorithm based perceptron( Per AOW) is proposed to select cluster heads in Ad hoc *** to AOW,simulation results proved that the proposed algorithm are improving network topological structure and giving 5.2% better load balance factor( LBF).
In data mining, clustering is the most popular, powerful and commonly used unsupervised learning technique. It is a way of locating similar data objects into clusters based on some similarity. clustering algorithms ca...
详细信息
ISBN:
(纸本)9781467385497
In data mining, clustering is the most popular, powerful and commonly used unsupervised learning technique. It is a way of locating similar data objects into clusters based on some similarity. clustering algorithms can be categorized into seven groups, namely Hierarchical clustering algorithm, Density-based clustering algorithm, Partitioning clustering algorithm, Graph-based algorithm, Grid-based algorithm, Model-based clustering algorithm and Combinational clustering algorithm. These clustering algorithms give different result according to the conditions. Some clustering techniques are better for large data set and some gives good result for finding cluster with arbitrary shapes. This paper is planned to learn and relates various data mining clustering algorithms. algorithms which are under exploration as follows: K-Means algorithm, K-Medoids, Distributed K-Means clustering algorithm, Hierarchical clustering algorithm, Grid-based Algorithm and Density based clustering algorithm. This paper compared all these clustering algorithms according to the many factors. After comparison of these clustering algorithms I describe that which clustering algorithms should be used in different conditions for getting the best result.
暂无评论