A method of shape encoding and retrieval is proposed in this letter, which uses centripetal code to encode shape and extracts shape's convex for retrieval. For the rotation invariance and translation invariance of...A method of shape encoding and retrieval is proposed in this letter, which uses centripetal code to encode shape and extracts shape's convex for retrieval. For the rotation invariance and translation invariance of the centripetal code and the normalization of convex,the proposed retrieval method is similarity transform resistant, Experimental results confirm this capability.展开更多
Electronic Commerce or E-Commerce is the concept of making trade in electronic environment which has emerged by the use of the internet in commerce and increased after 1995. Organizations started to use electronic com...Electronic Commerce or E-Commerce is the concept of making trade in electronic environment which has emerged by the use of the internet in commerce and increased after 1995. Organizations started to use electronic commerce by internet facilities nearby their traditional marketing methods to increase their sales and profits. E-Commerce has many advantages when compared to traditional trade. This paper aims to take out these advantages for the companies which want to make use of electronic commerce and includes important advices for these companies. The paper also includes the basic version of an e-commerce business, by analysing one of the most successful firms in Turkey, yemeksepeti.com in terms of an Entity-Relationship Diagram （ERD） and Event Driven Process （EDP） to explain the organizations how an e-commerce business model should be constructed systematically. An ERD is a data modeling technique which graphically illustrates an information system＇s entities and the relationships between those entities. On the other hand, EDP is a chain for the modeling language which is used for describing business processes and workflows of an organization. This study examines yemeksepeti.com for providing an example of these two diagrams, which should be a milestone for the newcomers of an e-commerce. Organizing an e-commerce should start with the software designs and these examples of ERD and EDP which are the base for the software of an e-commerce business. The limitations of the paper include the examination of an only one company which operates only in Turkey, thus the further research should examine a global company and/or an e-commerce business that operates in several industries.展开更多
A hybrid feature selection and classification strategy was proposed based on the simulated annealing genetic algonthrn and multiple instance learning （MIL）. The band selection method was proposed from subspace decom...A hybrid feature selection and classification strategy was proposed based on the simulated annealing genetic algonthrn and multiple instance learning （MIL）. The band selection method was proposed from subspace decomposition, which combines the simulated annealing algorithm with the genetic algorithm in choosing different cross-over and mutation probabilities, as well as mutation individuals. Then MIL was combined with image segmentation, clustering and support vector machine algorithms to classify hyperspectral image. The experimental results show that this proposed method can get high classification accuracy of 93.13% at small training samples and the weaknesses of the conventional methods are overcome.展开更多
基金National Natural Science Foundation of China(No. 60172045)863-306 Project (863-306-ZT03-09)
文摘A method of shape encoding and retrieval is proposed in this letter, which uses centripetal code to encode shape and extracts shape's convex for retrieval. For the rotation invariance and translation invariance of the centripetal code and the normalization of convex,the proposed retrieval method is similarity transform resistant, Experimental results confirm this capability.
文摘Electronic Commerce or E-Commerce is the concept of making trade in electronic environment which has emerged by the use of the internet in commerce and increased after 1995. Organizations started to use electronic commerce by internet facilities nearby their traditional marketing methods to increase their sales and profits. E-Commerce has many advantages when compared to traditional trade. This paper aims to take out these advantages for the companies which want to make use of electronic commerce and includes important advices for these companies. The paper also includes the basic version of an e-commerce business, by analysing one of the most successful firms in Turkey, yemeksepeti.com in terms of an Entity-Relationship Diagram （ERD） and Event Driven Process （EDP） to explain the organizations how an e-commerce business model should be constructed systematically. An ERD is a data modeling technique which graphically illustrates an information system＇s entities and the relationships between those entities. On the other hand, EDP is a chain for the modeling language which is used for describing business processes and workflows of an organization. This study examines yemeksepeti.com for providing an example of these two diagrams, which should be a milestone for the newcomers of an e-commerce. Organizing an e-commerce should start with the software designs and these examples of ERD and EDP which are the base for the software of an e-commerce business. The limitations of the paper include the examination of an only one company which operates only in Turkey, thus the further research should examine a global company and/or an e-commerce business that operates in several industries.
文摘A hybrid feature selection and classification strategy was proposed based on the simulated annealing genetic algonthrn and multiple instance learning （MIL）. The band selection method was proposed from subspace decomposition, which combines the simulated annealing algorithm with the genetic algorithm in choosing different cross-over and mutation probabilities, as well as mutation individuals. Then MIL was combined with image segmentation, clustering and support vector machine algorithms to classify hyperspectral image. The experimental results show that this proposed method can get high classification accuracy of 93.13% at small training samples and the weaknesses of the conventional methods are overcome.
