In this paper, we formulate image annotation as a semi-supervised learning problem under multi-instance learning framework. A novel graph based semi-supervised learning approach to image annotation using multiple instances is presented, which extends the conventional semi-supervised learning to multi-instance setting by introducing the adaptive geometric relationship between two bags of instances. The experiments over Corel images have shown that this approach outperforms other methods and is effective for image annotation.
{"title":"A Novel Region-based Image Annotation Using Multi-instance Learning","authors":"Xiaohong Hu, Xu Qian, Xinming Ma, Ziqiang Wang","doi":"10.1109/WKDD.2009.89","DOIUrl":"https://doi.org/10.1109/WKDD.2009.89","url":null,"abstract":"In this paper, we formulate image annotation as a semi-supervised learning problem under multi-instance learning framework. A novel graph based semi-supervised learning approach to image annotation using multiple instances is presented, which extends the conventional semi-supervised learning to multi-instance setting by introducing the adaptive geometric relationship between two bags of instances. The experiments over Corel images have shown that this approach outperforms other methods and is effective for image annotation.","PeriodicalId":143250,"journal":{"name":"2009 Second International Workshop on Knowledge Discovery and Data Mining","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129869892","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Density-based clustering and density-based outlier detection have been extensively studied in the data mining. However, Existing works address density-based clustering or density-based outlier detection solely. But for many scenarios, it is more meaningful to unify density-based clustering and outlier detection when both the clustering and outlier detection results are needed simultaneously. In this paper, a novel algorithm named DBCOD that unifies density-based clustering and outlier detection is proposed. In order to discover density-based clusters and assign to each outlier a degree of being an outlier, a novel concept called neighborhood-based local density factor (NLDF) is employed. The experimental results on different shape, large-scale, and high-dimensional databases demonstrate the effectiveness and efficiency of our method.
{"title":"Unifying Density-Based Clustering and Outlier Detection","authors":"Yunxin Tao, D. Pi","doi":"10.1109/WKDD.2009.127","DOIUrl":"https://doi.org/10.1109/WKDD.2009.127","url":null,"abstract":"Density-based clustering and density-based outlier detection have been extensively studied in the data mining. However, Existing works address density-based clustering or density-based outlier detection solely. But for many scenarios, it is more meaningful to unify density-based clustering and outlier detection when both the clustering and outlier detection results are needed simultaneously. In this paper, a novel algorithm named DBCOD that unifies density-based clustering and outlier detection is proposed. In order to discover density-based clusters and assign to each outlier a degree of being an outlier, a novel concept called neighborhood-based local density factor (NLDF) is employed. The experimental results on different shape, large-scale, and high-dimensional databases demonstrate the effectiveness and efficiency of our method.","PeriodicalId":143250,"journal":{"name":"2009 Second International Workshop on Knowledge Discovery and Data Mining","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130249442","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Privacy-preserving Data Mining aims at securely extracting knowledge from two or more parties' private data. Secure Multi-party Computation is the paramount approach to it. In this paper, we study Privacy-preserving Add and Multiply Exchanging Technology and present three new different approaches to Privacy-preserving Add to Multiply Protocol. After that, we analyze and compare the three different approaches about the communication overheads, the computation efforts and the security. In addition, we extend Privacy-preserving Add to Multiply Protocol to Privacy-preserving Adding to Scalar Product Protocol, which is more secure and more useful in the high security situations of Privacy-preserving Data Mining. Meantime, we present a solution for the new protocol.
{"title":"Three New Approaches to Privacy-preserving Add to Multiply Protocol and its Application","authors":"Youwen Zhu, Liusheng Huang, Wei Yang, Dong Li, Yonglong Luo, Fan Dong","doi":"10.1109/WKDD.2009.34","DOIUrl":"https://doi.org/10.1109/WKDD.2009.34","url":null,"abstract":"Privacy-preserving Data Mining aims at securely extracting knowledge from two or more parties' private data. Secure Multi-party Computation is the paramount approach to it. In this paper, we study Privacy-preserving Add and Multiply Exchanging Technology and present three new different approaches to Privacy-preserving Add to Multiply Protocol. After that, we analyze and compare the three different approaches about the communication overheads, the computation efforts and the security. In addition, we extend Privacy-preserving Add to Multiply Protocol to Privacy-preserving Adding to Scalar Product Protocol, which is more secure and more useful in the high security situations of Privacy-preserving Data Mining. Meantime, we present a solution for the new protocol.","PeriodicalId":143250,"journal":{"name":"2009 Second International Workshop on Knowledge Discovery and Data Mining","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128709781","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the design of Viterbi Decoder of punctured convolutional codes, the SST method is used to reduce power consumption in this paper. Moreover, the punctured vector table is optimized by hard wire logic, thus reducing the area required by the system. Effective operation of adjustment is used to reduce the word length of path metric memory. At the same time, this algorithm can also help reduce operation load by controlling the highest order of path metric memory, thereby reducing the scale of hardware.
{"title":"An Efficient Viterbi Decoder for Digital Mobile Multimedia Broadcasting Receiver","authors":"Hongli Zhu, G. Gao, Gang Bi","doi":"10.1109/WKDD.2009.196","DOIUrl":"https://doi.org/10.1109/WKDD.2009.196","url":null,"abstract":"In the design of Viterbi Decoder of punctured convolutional codes, the SST method is used to reduce power consumption in this paper. Moreover, the punctured vector table is optimized by hard wire logic, thus reducing the area required by the system. Effective operation of adjustment is used to reduce the word length of path metric memory. At the same time, this algorithm can also help reduce operation load by controlling the highest order of path metric memory, thereby reducing the scale of hardware.","PeriodicalId":143250,"journal":{"name":"2009 Second International Workshop on Knowledge Discovery and Data Mining","volume":"31 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120900023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ontology mapping is a core task to achieve interoperability among the distributed and heterogeneous ontologies. The quality of ontology mapping is still not good due to the inconsideration of semantic information of the ontology. This paper proposes an ontology mapping method based on tree segmentation. It first divides the ontology into a set of sub-trees with different granularities according to the structure of the ontologies, then use Sub-tree Mapping Algorithm to map them. Preliminary experiments demonstrate that the proposed mapping method performs well in both precision and recall compared with the current mapping methods.
{"title":"Research on Tree Segmentation-based Ontology Mapping","authors":"Liansheng Li, Lihui Huang, Qinghua Guan, Dezhi Xu","doi":"10.1109/WKDD.2009.135","DOIUrl":"https://doi.org/10.1109/WKDD.2009.135","url":null,"abstract":"Ontology mapping is a core task to achieve interoperability among the distributed and heterogeneous ontologies. The quality of ontology mapping is still not good due to the inconsideration of semantic information of the ontology. This paper proposes an ontology mapping method based on tree segmentation. It first divides the ontology into a set of sub-trees with different granularities according to the structure of the ontologies, then use Sub-tree Mapping Algorithm to map them. Preliminary experiments demonstrate that the proposed mapping method performs well in both precision and recall compared with the current mapping methods.","PeriodicalId":143250,"journal":{"name":"2009 Second International Workshop on Knowledge Discovery and Data Mining","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122357817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Personal knowledge management is an increasingly important subject in the field of knowledge management, and it has its own characteristics distinguished from organizational knowledge management. The PKM systems are required to manage personal tacit knowledge efficiently and to low down its threshold for the use for the mass people, which recent knowledge management systems cannot make. Blog, the typical web2.0 application, gives an effective approach for this problem to support PKM. A blog based PKM system is deemed to be possible and to meet the urgent needs for PKM of individuals.
{"title":"A Study on Blog Based Personal Knowledge Management","authors":"Gang Li, Yi Li","doi":"10.1109/WKDD.2009.166","DOIUrl":"https://doi.org/10.1109/WKDD.2009.166","url":null,"abstract":"Personal knowledge management is an increasingly important subject in the field of knowledge management, and it has its own characteristics distinguished from organizational knowledge management. The PKM systems are required to manage personal tacit knowledge efficiently and to low down its threshold for the use for the mass people, which recent knowledge management systems cannot make. Blog, the typical web2.0 application, gives an effective approach for this problem to support PKM. A blog based PKM system is deemed to be possible and to meet the urgent needs for PKM of individuals.","PeriodicalId":143250,"journal":{"name":"2009 Second International Workshop on Knowledge Discovery and Data Mining","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127210375","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Associative classification has high classification accuracy and strong flexibility. However, it still suffers from overfitting since the classification rules satisfied both minimum support and minimum confidence are returned as strong association rules back to the classifier. In this paper, we propose a new association classification method based on compactness of rules, it extends Apriori Algorithm¿which considers the interestingness, importance, overlapping relationships among rules. At last, experimental results shows that the algorithm has better classification accuracy in comparison with CBA and CMAR are highly comprehensible and scalable.
{"title":"Association Classification Based on Compactness of Rules","authors":"Q. Niu, Shixiong Xia, Lei Zhang","doi":"10.1109/WKDD.2009.160","DOIUrl":"https://doi.org/10.1109/WKDD.2009.160","url":null,"abstract":"Associative classification has high classification accuracy and strong flexibility. However, it still suffers from overfitting since the classification rules satisfied both minimum support and minimum confidence are returned as strong association rules back to the classifier. In this paper, we propose a new association classification method based on compactness of rules, it extends Apriori Algorithm¿which considers the interestingness, importance, overlapping relationships among rules. At last, experimental results shows that the algorithm has better classification accuracy in comparison with CBA and CMAR are highly comprehensible and scalable.","PeriodicalId":143250,"journal":{"name":"2009 Second International Workshop on Knowledge Discovery and Data Mining","volume":"08 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127224600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The premise of the enterprise that implements the product family architecture is the analysis to the sailed products history data, carrying on accuracy demarcation of the customer sets and products sets. But, the traditional cluster methods consist in simultaneous one partitioning of the set of products or the set of customers. In this paper we review the most widely used and successful biclustering techniques and use bicluster consist in simultaneous partitioning of the set of products and the set of customers. The bicluster algorithm is discussed. The results of the improved bicluster algorithm used in the product family architecture analysis show that, compared with traditional cluster methods, the quality of clustering results is enhanced obviously, the mining expression models are better, and the data possess a strong consistency with fluctuation on the condition.
{"title":"Bicluster Algorithm and Used in Market Analysis","authors":"Shuyong Liu, Yan Chen, Mingyuan Yang, Rui Ding","doi":"10.1109/WKDD.2009.224","DOIUrl":"https://doi.org/10.1109/WKDD.2009.224","url":null,"abstract":"The premise of the enterprise that implements the product family architecture is the analysis to the sailed products history data, carrying on accuracy demarcation of the customer sets and products sets. But, the traditional cluster methods consist in simultaneous one partitioning of the set of products or the set of customers. In this paper we review the most widely used and successful biclustering techniques and use bicluster consist in simultaneous partitioning of the set of products and the set of customers. The bicluster algorithm is discussed. The results of the improved bicluster algorithm used in the product family architecture analysis show that, compared with traditional cluster methods, the quality of clustering results is enhanced obviously, the mining expression models are better, and the data possess a strong consistency with fluctuation on the condition.","PeriodicalId":143250,"journal":{"name":"2009 Second International Workshop on Knowledge Discovery and Data Mining","volume":"250 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124059623","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Since the inception of the Internet in the mid-1990s, e-commerce has grown at a very rapid rate. Firms had realized that they could be pushed to the world stage of the knowledge-based economic by e-commerce. So, most of firms are stepping up to the adoption of those technologies for upgrading their businesses. Regarding the significant role of customers, the study of their behaviors influence on the adoption of e-commerce by the firms becomes a main stream in scholastic path. This paper aims to study the four customer behaviors which influence the adoption of e-commerce technologies by firms. The study results showed that the customer behaviors influence the adoption of e-commerce by firms.
{"title":"Research on the Adoption of Electronic Commerce Based on Customer's Behavior","authors":"Yuantao Jiang, Yuqin Sun","doi":"10.1109/WKDD.2009.129","DOIUrl":"https://doi.org/10.1109/WKDD.2009.129","url":null,"abstract":"Since the inception of the Internet in the mid-1990s, e-commerce has grown at a very rapid rate. Firms had realized that they could be pushed to the world stage of the knowledge-based economic by e-commerce. So, most of firms are stepping up to the adoption of those technologies for upgrading their businesses. Regarding the significant role of customers, the study of their behaviors influence on the adoption of e-commerce by the firms becomes a main stream in scholastic path. This paper aims to study the four customer behaviors which influence the adoption of e-commerce technologies by firms. The study results showed that the customer behaviors influence the adoption of e-commerce by firms.","PeriodicalId":143250,"journal":{"name":"2009 Second International Workshop on Knowledge Discovery and Data Mining","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124418743","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
For different kinds of scale networks, the different weight ciphers should be designed for adapting various operation environments. In wireless networks or RFID environment, the lightweight ciphers or ultra-lightweight ciphers have been paid more and more attention. At Internet environment, there exist some block cipher algorithms used in top layer but not published block ciphers announce to only design for the other layers. The standardized security protocols of block cipher are oriented to the public level which is seldom considered based on different secure situation in openness Internet environment. And in trusted platform, some modules still need heavy ciphers. So this paper’s approach is that it needs precise cryptographic system according to different networks and different functions. The scalable and flexible block cipher application is another approach to currently secure standard and the lightweight block cipher design should be paid more attention. In order to expending networks member number easily, some intelligent functions are combined in the design.
{"title":"A Comment to the Intelligent Functions of Different Weight Ciphers","authors":"Lan Luo, Zhiguang Qin, Shijie Zhou","doi":"10.1109/WKDD.2009.18","DOIUrl":"https://doi.org/10.1109/WKDD.2009.18","url":null,"abstract":"For different kinds of scale networks, the different weight ciphers should be designed for adapting various operation environments. In wireless networks or RFID environment, the lightweight ciphers or ultra-lightweight ciphers have been paid more and more attention. At Internet environment, there exist some block cipher algorithms used in top layer but not published block ciphers announce to only design for the other layers. The standardized security protocols of block cipher are oriented to the public level which is seldom considered based on different secure situation in openness Internet environment. And in trusted platform, some modules still need heavy ciphers. So this paper’s approach is that it needs precise cryptographic system according to different networks and different functions. The scalable and flexible block cipher application is another approach to currently secure standard and the lightweight block cipher design should be paid more attention. In order to expending networks member number easily, some intelligent functions are combined in the design.","PeriodicalId":143250,"journal":{"name":"2009 Second International Workshop on Knowledge Discovery and Data Mining","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116840279","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：