Decision tree and fuzzy rough set are two distinct but complementary *** tree is a simple and easy-understandable rule-based classifier,whereas the tool of fuzzy rough sets are effective on attribute and sample *** is...
详细信息
Decision tree and fuzzy rough set are two distinct but complementary *** tree is a simple and easy-understandable rule-based classifier,whereas the tool of fuzzy rough sets are effective on attribute and sample *** is promising to propose an approach to integrate these two rule based classification tools to construct a novel decision tree based on fuzzy rough *** this paper,based on the basic concept of fuzzy rough sets,i.e.,consistence degree,we propose a fuzzy rough decision tree which is completely different from the existing classification *** three key basic elements of decision tree,i.e.,node,branch and leaf,are designed in a new way by using the notions in fuzzy rough *** then one algorithm to build fuzzy rough tree classifier is ***,experimental results show that the proposed algorithm is readily comprehensible and effective.
This paper aims to identify knowledge management (KM) methods that have been used in three KM Standards, i.e. KM Standards of Britain, European Union(EU) and Australia, with comparative analysis of three KM issues cov...
详细信息
This paper aims to identify knowledge management (KM) methods that have been used in three KM Standards, i.e. KM Standards of Britain, European Union(EU) and Australia, with comparative analysis of three KM issues covering: basic definitions, KM work processes, essential factors and management frameworks;with a chronological analysis of their characteristics and evolution. The studies find that along with the practical development of KM practices, concepts of KM are getting understood more and more broadly, ranges of KM activities are extending, but management factors are getting highly integrated, and the management frameworks are going towards ecosystem. The paper provides valuable insight on theoretical and practical development of knowledge management methods and standards.
Clustering XML search results is an effective way to improve performance. However, the key problem is how to measure similarity between XML documents. This paper studies XML search results clustering based on element ...
详细信息
In this paper,we first analyze how the characteristics of the GPU affect the performance of link-based similarity *** on the analysis,we describe our strategies to improve the performance of link-based similarity meas...
详细信息
In this paper,we first analyze how the characteristics of the GPU affect the performance of link-based similarity *** on the analysis,we describe our strategies to improve the performance of link-based similarity measures on the GPU in ***,through extensive experiments,we evaluate the effectiveness of our strategies.
In this paper, we address the problem of data Compression which is critical in wireless sensor networks. We proposed a novel Topology-based data Compression (TDC) algorithm for wireless sensor networks. We utilize the...
详细信息
Consumer online shopping behaviors are well attended in the IS and marketing literature. Yet, there is another group of individuals who spend a lot of time online but do not purchase anything. This online window shopp...
详细信息
Consumer online shopping behaviors are well attended in the IS and marketing literature. Yet, there is another group of individuals who spend a lot of time online but do not purchase anything. This online window shopping phenomenon is intriguing to both scholars and marketers yet it is less studied and little understood. Questions such as what the online window shopping consumers do during their visits, how to differentiate their activities and how to design marketing strategies to stimulate them to buy are all essential and beg for investigation. To address this gap, we propose a typology of online window shopping consumers based on the Consumer Information Processing Model, then empirically validate and refine the typology using a set of clickstream data. The final typology contains four main types of online window shopper consumers: 1) promotion finders, 2) social & hedonic experience seekers, 3) information gatherers, and 4) learners & novices. This study extends consumer online behavior research in both e-commerce and social commerce by focusing on the specific group of consumers who only do online window shopping. Besides theoretical contributions, the findings also provide marketers and businesses with valuable references for designing targeted marketing strategies or promotional activities for online window shopping consumers.
Along with the development of Internet and Web2.0, online social networks (OSNs) are becoming an important information propagation platform. Therefore, it is of great significance to study the information propagation ...
详细信息
In this paper, the author defines Generalized Unique Game Problem (GUGP), where weights of the edges are allowed to be negative. Two special types of GUGP are illuminated, GUGP-NWA, where the weights of all edges are ...
详细信息
As the sharable and reusable domain knowledge, domain ontology increasingly serves as a foundation for semantic Web. Personalized management of domain ontologies is to provide personalized views of domain ontologies t...
详细信息
As the sharable and reusable domain knowledge, domain ontology increasingly serves as a foundation for semantic Web. Personalized management of domain ontologies is to provide personalized views of domain ontologies to users during run time according to user preferences. It helps that users can focus on their interested parts, instead of the whole. It increases the efficiency of ontology-based application. This paper proposes a framework for personalized management of domain ontologies. In this framework, a user model firstly is introduced to describe user preferences. Secondly, domain ontologies are decomposed into moderate-scale modules with high cohesion and low coupling. During run time, modules that a user is interested in are selected out based on the user's preferences. Finally, the selected modules are combined to construct personalized views of domain ontologies for the user.
A novel method is proposed to automatically extract foreground objects from Martian surface *** characteristics of Mars images are distinct,*** illumination,low contrast between foreground and background,much noise in...
详细信息
A novel method is proposed to automatically extract foreground objects from Martian surface *** characteristics of Mars images are distinct,*** illumination,low contrast between foreground and background,much noise in the background,and foreground objects with irregular *** the context of these characteristics,an image is divided into foreground objects and background *** filtering is first applied to rectify ***,wavelet transformation enhances contrast and denoises the ***,edge detection and active contour are combined to extract contours regardless of the shape of the *** results show that the method can extract foreground objects from Mars images automatically and accurately,and has many potential applications.
暂无评论