E-FORENSICS 2008
WKDD 2008

    WKDD

    1st International ICST Workshop on Knowledge Discovery and Data Mining

    Knowledge discovery and data mining have become areas of growing significance because of the recent increasing demand for KDD techniques, including those used in machine learning, databases, statistics, knowledge acquisition, data visualization, and high performance computing. Knowledge discovery a…

    Knowledge discovery and data mining have become areas of growing significance because of the recent increasing demand for KDD techniques, including those used in machine learning, databases, statistics, knowledge acquisition, data visualization, and high performance computing. Knowledge discovery and data mining can be extremely beneficial for the field of Artificial Intelligence in many areas, such as industry, commerce, government, education and so on. In view of this, a workshop on knowledge Discovery and Data Mining aims to provide an opportunity for the sharing of original research results, innovative ideas, state-of-the-art developments, and implementation experiences in knowledge discovery and data mining among researchers in academic and industrial organizations.

    more »
    Editor(s): Qi LuoWuhan, Mingmin Gong, Feng Xiong and Fei Yu
    Publisher
    ACM
    Conference dates
    23rd–24th Jan 2008
    Location
    Adelaide, Australia
    Appeared in EUDL
    29th Nov 2011
    Appears in
    ACM Digital Library

    Copyright © 2011–2013 ICST

    • Solving Constrained Optimization via a Modified Genetic Particle Swarm Optimization

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Liu Zhiming, Wang Cheng, Li Jiang

      Abstract
      The genetic particle swarm optimization (GPSO) was derived from the original particle swarm optimization (PSO), which is incorporated with the genetic reproduction mechanisms, namely crossover and mu…The genetic particle swarm optimization (GPSO) was derived from the original particle swarm optimization (PSO), which is incorporated with the genetic reproduction mechanisms, namely crossover and mutation. Based on which a modified genetic particle swarm optimization (MGPSO) was introduced to solve constrained optimization problems. In which the differential evolution (DE) was incorporated into GPSO to enhance search performance. At each generation GPSO and DE generated a position for each particle, respectively, and the better one was accepted to be a new position for the particle. To compare and ranking the particles, the lexicographic order ranking was introduced. Moreover, DE was incorporated to the original PSO with the same method, which was used to be compared with MGSPO. MGPSO were experimented with wellknown benchmark functions. By comparison with original PSO algorithms and the evolution strategy, the simulation results have shown its robust and consistent effectiveness.
       more »

    • Study on the Application of SVM in Supplier Primary Election

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Lili Cai, Fugeng Shong, Deling Yuan

      Abstract
      Supplier selection is one of the most important things in supply chain management, the process of choosing a suitable supplier will effect enterprise’s production and quality directly. This article p…Supplier selection is one of the most important things in supply chain management, the process of choosing a suitable supplier will effect enterprise’s production and quality directly. This article puts forward a two-stage model of supplier selection based on analysis the problem of supplier management. It plots out the process as primary election stage and well-chosen stage, and builds up seven criteria to evaluate suppliers in primary election phase. Then, this context uses support vector machine to select suppliers and pays attention to two kinds of error — treat candidate supplier as non-candidate supplier and treat non-candidate supplier as candidate supplier. At last, a numerical simulation is used to explain selection of kernel function and sample training; the result reveals that this new method is practical and realistic and could reduce selection time.
       more »

    • CBERS-02 remote sensing data mining using decision tree algorithm

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Xingping Wen, Guangdao Hu, Xiaofeng Yang

      Abstract
      In recent years, decision tree algorithms have been successfully used for land cover classification from remote sensing data. In this paper, CART (classification and regression trees) and C5.0 decisi…In recent years, decision tree algorithms have been successfully used for land cover classification from remote sensing data. In this paper, CART (classification and regression trees) and C5.0 decision tree algorithms were used to CBERS-02 remote sensing data. Firstly, the remote sensing data was transformed using the Principal Component Analysis (PCA) and multiple-band algorithm. Then, the training data was collected from the combining total 20 processed bands. Finally, the decision tree was constructed by CART and C5.0 algorithm respectively. Comparing two results, the most important variables are clearly band3,4, band1,4 and band2,4. The depth of the CART tree is only two with the relative high accuracy. The classification outcome was calculated by CART tree. In order to validate the classification accuracy of CART tree, the Confusion Matrices was generated by the ground truth data collected using visual interpretation and the field survey and the kappa coefficient is 0.95.
       more »

    • The method engineering process for multi-agent system development

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Xue Xiao, Zhang Xue Yan

      Abstract
      The agent-oriented (AO) methodology is an effective means for constructing distributed systems. Despite a great deal of research, a number of challenges still exist before making agent-based computin…The agent-oriented (AO) methodology is an effective means for constructing distributed systems. Despite a great deal of research, a number of challenges still exist before making agent-based computing a widely accepted paradigm in software engineering practice. In order to solve the problem of “difficult to implement”, the paper presents a method engineering development process for facilitating building up new agent system by reusing method meta models and design patterns. As the supporting tool, the hierarchical development architecture(HDA) plays a key role in the process. To exemplify its feasibility and effectiveness, the construction of C4I system is presented as a case study.
       more »

    • An Empirical Study on Improving the Manufacturing Informatization Index System of China

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Wei Guo, Ke Chen, Jia Wang

      Abstract
      The manufacturing informatization index system (MIIS) is an indispensable tool to measure the informatization level of Chinese manufacturing industry and evaluate the implementation effect of the man…The manufacturing informatization index system (MIIS) is an indispensable tool to measure the informatization level of Chinese manufacturing industry and evaluate the implementation effect of the manufacturing informatization engineering (MIE) conducted by the government of China. Thus far, the constructs of MIIS has not been validated. This study fills this void by employing structural equation modeling (SEM) to test the MIIS model. The samples in this study come from the standard database of Chinese manufacturing informatization established by the data survey of MIE during the “Tenth Five-Year Plan” period, including 12896 enterprises samples from 11 manufacturing industries and 3472 support samples from 29 provinces of China. Based on the results of SEM analysis, some indexes of MIIS are adjusted and an improved MIIS is got at last. This empirical study proves that combining SEM technology and standard data resources would be an ideal method to improve MIIS.
       more »

    • An Empirical Research of Multi-Classifier Fusion Methods and Diversity Measure in Remote Sensing Classification

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Hongchao Ma, Wei Zhou, Xinyi Dong, Honggen Xu

      Abstract
      In this paper, Multi-Classifier System (MCS) is applied to the automatic classification of remote sensing images, and some effective multi-classifier fusion methods with relatively high accuracy are …In this paper, Multi-Classifier System (MCS) is applied to the automatic classification of remote sensing images, and some effective multi-classifier fusion methods with relatively high accuracy are proposed based on substantive experiments. The classification accuracy of MCS has been remarkably improved compared to single classifier with an average increment of 5%. In addition, a diversity measure named EPD is presented, and the paper proves that its ability in predicting the performance of classifiers combining can be used to assist the construction of multiple classifier systems.
       more »

    • Ontology-based Research on Wind Power Plant Information Interaction

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Yong-li Zhu, Xin-ying Wang, Dong-ling Cheng

      Abstract
      At present, wind power industry has developed rapidly in our country, and has also promoted the step of wind power plant information construction. But in this process, it’s impossible to set up a uni…At present, wind power industry has developed rapidly in our country, and has also promoted the step of wind power plant information construction. But in this process, it’s impossible to set up a uniform wind power plant management information system using conventional technology and method, because the wind power plant’s devices span a long time, have many categories and the information model and communication protocol are also diverse in different wind turbines. Thus, this paper imports knowledgesharing- based ontology semantic technology, builds the ontology on the basis of wind power plant information model and describes it using OWL. It achieves the wind power plant information sharing, interaction and procession at the semantic level, and provides a new method to build a uniform wind power plant management information system.
       more »

    • Research of a virtual 3D study pattern based on constructive theory in e-learning

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Xinyu Duan, Ping Jiang

      Abstract
      E-learning is most important by its role in distance teaching, and as supplementary learning material, it’s mainly used in higher education area. Paper first introduced the significance of e-Learning…E-learning is most important by its role in distance teaching, and as supplementary learning material, it’s mainly used in higher education area. Paper first introduced the significance of e-Learning and application of constructive theory. In contrast with traditional education, its application possesses broad prospects. Then we made a systematic exposition of interrelated theory basis, discussed a new 3D virtual study pattern. Lastly, it presented two application systems that we developed. One is a virtual audio-video multimedia teaching center which has good immersion feelings; another is a virtual music appreciation environment which has dynamic interaction performance.
       more »

    • A Customer Satisfaction Degree Evaluation Model Based on Support Vector Machine

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Wang Ting, Hua Zhiwu

      Abstract
      An efficient classification algorithm is proposed for evaluating the customer satisfaction degree. The algorithm is based on the RBF-Kernel support vector machine and multilevel binary tree classifie…An efficient classification algorithm is proposed for evaluating the customer satisfaction degree. The algorithm is based on the RBF-Kernel support vector machine and multilevel binary tree classifier. Fuzzy membership function was used to quantify the evaluation indices. The evaluation indices and the SVM algorithm were used to design a customer satisfaction degree evaluation model. The novel evaluation method has higher accuracy in comparison with the traditional fuzzy comprehensive evaluation method and BP evaluation method.
       more »

    • Towards Self-tuning of Dynamic Resources for Workloads

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      FU Duan, Yongjie Han, Qiuyong Zhao, Keming Xie

      Abstract
      In self-tuning of database performance research area, the optimization of dynamic resources is important. This paper focused on the self-tuning of dynamic resources on the base of feedback mechanism,…In self-tuning of database performance research area, the optimization of dynamic resources is important. This paper focused on the self-tuning of dynamic resources on the base of feedback mechanism, designed and achieved a simple model combined with Oracle database. This model gave an evaluating principle of system performance, and could adjust system parameters automatically to achieve the goal of system performance. And an evaluation model was designed to evaluate the performance state of the running system. The result after the self-tuning of some parameters was illustrated by simulating various workloads.
       more »

    • Coupling Analysis of Regional Economic Structure System

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      ZHANG Yi-xin

      Abstract
      Regional economic structure system is an exoteric self-organization system with dissipation structure and high-step, multi-variables, multi-loops and nonlinear feedback structure. On the basis of the…Regional economic structure system is an exoteric self-organization system with dissipation structure and high-step, multi-variables, multi-loops and nonlinear feedback structure. On the basis of the qualitative analysis of coupling relations of subsystems in regional economic structure system, considering Heilongjiang province as the example, the paper establishes the model applying system dynamics method. By means of this model, the operating mechanism of regional economic structure system is studied. Some results have been concluded: the error is little between simulative value and true value, which means that the model could represent the performance of regional economic structure system and the obtained strategic conclusions can give reference to the future regulation.
       more »

    • Mining High Utility Itemsets in Large High Dimensional Data

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Guangzhu Yu, Keqing Li, Shihuang Shao

      Abstract
      Existing algorithms for utility mining are inadequate on datasets with high dimensions or long patterns. This paper proposes a hybrid method, which is composed of a row enumeration algorithm (i.e., I…Existing algorithms for utility mining are inadequate on datasets with high dimensions or long patterns. This paper proposes a hybrid method, which is composed of a row enumeration algorithm (i.e., Inter-transaction) and a column enumeration algorithm (i.e., Two-phase), to discover high utility itemsets from two directions: Two-phase seeks short high utility itemsets from the bottom, while Intertransaction seeks long high utility itemsets from the top. In addition, optimization technique is adopted to improve the performance of computing the intersection of transactions. Experiments on synthetic data show that the hybrid method achieves high performance in large high dimensional datasets.
       more »

    • Grasping Related Words of Unknown Word for Automatic Extension of Lexical Dictionary

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Myunggwon Hwang, Jongan Park, Sunkyoung Baek, Pankoo Kim, Junho Choi

      Abstract
      An aim of this research is to grasp related words of unknown word. Currently, several lexical dictionaries have been developed for semantic retrieval such as WordNet and FrameNet. However, more new w…An aim of this research is to grasp related words of unknown word. Currently, several lexical dictionaries have been developed for semantic retrieval such as WordNet and FrameNet. However, more new words are created in every day because of new trends, new paradigm, new technology, etc. And, it is impossible to contain all of these new words. The existing methods, which grasp the meaning of unknown word, have a limitation that is not exact. To solve this limitation, we have studied the way how to make relations between known words and unknown word. As a result, we found a noble method using co-occurrence, WordNet and Bayesian probability. The method could find what words are related with unknown word and how much weight other words relate with unknown word.
       more »

    • The Data Mining Technology Based on CIMS and Its Application on Automotive Remanufacturing

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Ke Chen, Jia Wang

      Abstract
      Nowadays, with the development of computer technology, data mining has been widely used in various fields. This paper describes a CIMSMINER that combines the data mining with CIMS (Computer Integrate…Nowadays, with the development of computer technology, data mining has been widely used in various fields. This paper describes a CIMSMINER that combines the data mining with CIMS (Computer Integrated Manufacturing System) and instructs its objectives, model, physical architecture and methods. Considering the characteristics of remanufacturing of automotive products in China, the CIMSMINER is used to get the information concourse together and obtain the data mining results to help the improvement of products. The application in automotive remanufacturing is a reform to make the automotive product information chain be not only an information carrier, but also an information miner. Currently, the government strongly emphasizes energy saving and emission reducing, which is closely related to the sustainable development of China. Obviously, CIMSMINER is an effective tool to support the implementation of this policy.
       more »

    • Centrality Research on the Traditional Chinese Medicine Network*

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Zhang Dezheng, Gao Lixin, Zhang Huansheng, Liu Jianming

      Abstract
      Aiming at the complex data in the Traditional Chinese Medicine, a new way is proposed in this paper that data mining of complex relations to find out the potential information among different medicin…Aiming at the complex data in the Traditional Chinese Medicine, a new way is proposed in this paper that data mining of complex relations to find out the potential information among different medicine objects. We turned the Traditional Chinese Medicine knowledge network into graph by using information from ontology, then adopted centrality algorithm to analyze and process this graph, and finally mined valuable medicine knowledge. As the result of the verification test, this algorithm shows very good practicability.
       more »

    • A Discussion Information-Structuring Model Based on the Toulmin Formalism

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Caiquan Xiong, Ying Pan, Dehua Li

      Abstract
      Abstract In the course of discussion, the group will produce a lot of discussion information, in which the consensus is hided. This paper proposes a discussion information-structuring model for conse…Abstract In the course of discussion, the group will produce a lot of discussion information, in which the consensus is hided. This paper proposes a discussion information-structuring model for consensus building. The model divided a statement into several parts such as premise, warrant, modality, claim etc., according to the Toulmin argument formalism. “Modelity” is designed as a quantitative scale that reflects the expert’s attitude to the claim, while premise and warrant are used to provide demonstration for the expert views. We can reach a consensus by computing the value of modality. The model is a comprehensive description of the discussion information, and can effectively manage the discussion information and do consensus building.
       more »

    • A Remote Sensing Image Fusion Algorithm Based on Ordinal Fast Independent Component Analysis

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Zhongni Wang, Xianchuan Yu, Libao Zhang

      Abstract
      Data fusion on remote sensing is hot in current image processing. The key of a successful image fusion is to find an effective and practical image fusion algorithm. A new approach , that is the ordin…Data fusion on remote sensing is hot in current image processing. The key of a successful image fusion is to find an effective and practical image fusion algorithm. A new approach , that is the ordinal fast independent component analysis for remote image fusion between Landsat ETM+ panchromatic and CBERS multi-spectral images, is proposed to eliminate high-order image data redundancy for two different Remote sensing images. The independent components are done factor analysis, and then the fused image is obtained by applying image fusion rule. Visual and statistical analysis proves that the concept of fusion based on the ordinal fast independent component analysis is promising, and it significantly increases the signal-to-noise ratio and improves the fusion quality.
       more »

    • Cloud Model-based Data Attributes Reduction for Clustering

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      XU Ru-zhi, NIE Pei-yao, LIN Pei-guang, CHU Dong-sheng

      Abstract
      Data reduction, which can simplify large scale data and not lose useful information, is an important topic of knowleadge discorvery, data clustering and classification. Aiming to solve the current pr…Data reduction, which can simplify large scale data and not lose useful information, is an important topic of knowleadge discorvery, data clustering and classification. Aiming to solve the current problem that continuous attribute in algorithm of clustering or classification has to be discrete, a new algorithm of data reduction based on cloud model is put forward. By use of cloud model, this algorithm calculates each conditional attribute’s importance to decision-making attribute( s), and obtains the reduction attributes by virtue of greedy algorithm. This new data reduction algorithm was verified by some experiments and was proved to be stable and efficient.
       more »

    • Temporal and Spatial Analysis on Labor Productivity of Radio and TV Industry in China

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Dong Chun, ZHANG Hongmei, Wang Tao, Wang Yong, QIU Agen

      Abstract
      In order to find out the industrial characteristics and development trend of radio and TV industry in China, the developing course of this industry has been analysis from labor productivity, based on…In order to find out the industrial characteristics and development trend of radio and TV industry in China, the developing course of this industry has been analysis from labor productivity, based on contrastive analysis and diagrammatic analysis. By comparison to three major industries, the developing rapidly and hopeful road has been exposed, and before it has not received adequate recognition to itself. In order to study the spatial distribution rule, using the spatial analysis function of GIS and spatial statistical models, quantitative analysis and qualitative evaluation on Radio and TV industry in China have been done. Then from the spatial scale, the spatial modes of this industry both on province and county scale have been studied deeply. The results show that there are certain spatial agglomerations in general for labor productivity in China. In the local region, obvious spatial associations exit too.
       more »

    • Evaluating Organization External Knowledge Acquisition

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      BAO Zhengqiang, GAO Kaizhou, LI Xiangqing, GUO Lei, ZHONG Hui

      Abstract
      Knowledge can be acquired from organization inside and exterior. The knowledge inside organization is limited and the organization external knowledge is important to advance organization competition …Knowledge can be acquired from organization inside and exterior. The knowledge inside organization is limited and the organization external knowledge is important to advance organization competition ability. Knowledge acquisition out of organization plays an extremely important role in organization knowledge management. This study proposed a knowledge acquisition evaluating method based on knowledge scenario. Scenario tree is defined and is used to describe the scenario of origination external knowledge. Scenario tree’s dimensions and items are confirmed and their fuzzy weight is afforded. State interval of evaluating method is plot out. Evaluating matrix is proposed for computing the feasibility of organization external knowledge acquisition. The evaluating method is a recursion process among layers. A case is offered to demonstrate the availability of this approach in knowledge acquisition form organization exterior.
       more »

    • Research and Application on Typical Process Knowledge Discovery in Mechanical Manufacturing Enterprise

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Xiaoliang Jia, Zhenming Zhang, Xitian Tian

      Abstract
      The source and composing of process planning knowledge is analyzed based on the state of art in discrete mechanical manufacturing enterprise. On the basis of the widely application of computer aided …The source and composing of process planning knowledge is analyzed based on the state of art in discrete mechanical manufacturing enterprise. On the basis of the widely application of computer aided process planning system (CAPP) in mechanical manufacturing enterprise, the concept of process planning knowledge discovery (PPKD) is proposed for product process planning database. CAPPFramework (a CAPP development platform that is developed by Northwestern Polytechnical University supported by China) is taken as a basic development platform, the technology architecture of process planning knowledge discovery is founded based on object-oriented model driven technology, and the process planning knowledge discovery script is designed. Elementary application research in typical process summarization is described in detail. The technology of PPKD has been used in mechanical manufacturing enterprise to support the automatically knowledge acquisition in CAPP system, and it shows good application effect.
       more »

    • Cooperation Forensic Computing Research

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Youdong ZHANG

      Abstract
      The network forensic computing is faced with the question of the complex network intrusion analyses. So a new concept of cooperation forensic computing is defined. Through to extend the theory of fun…The network forensic computing is faced with the question of the complex network intrusion analyses. So a new concept of cooperation forensic computing is defined. Through to extend the theory of function dependency, a new method called probability function dependency relationships is proposed. Combined it with the Bayesian network and K2 algorithm, the network forensic computing algorithm called CFA is proposed. For the complex network attack, CFA is able to synthesize the various forensic data resource to reappearance the crime scenario intuitionally and realize the network forensic analysis effectively.
       more »

    • Knowledge Management in the Ubiquitous Software Development

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      José Miguel Rubio L.

      Abstract
      The continuous technical advances have lead to the proliferation of very small and very cheap microprocessors, equipped with sensors and capacity of wireless communication. The information processing…The continuous technical advances have lead to the proliferation of very small and very cheap microprocessors, equipped with sensors and capacity of wireless communication. The information processing is becoming ubiquitous and it is being impregnated in all type of objects. In this article the general delineations set out towards a methodology of securing of the quality of software ubiquitous based their main characteristics: centered in the user and highly interactive. Moreover, it considered to the usability as the quality characteristic of more relevant in the development of this type of highly interactive software systems.
       more »

    • Application of tetrahedral mesh model based on neural network in solid mineral reserve estimation

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Junfang Gong, Xincai Wu, Xiuguo Liu, Shengwen Li

      Abstract
      Mineral reserve estimation involves large amounts of geological data, and the traditional manual computation method is a heavy and fussy work. Computing mineral reserve with 3D orebody modeling can i…Mineral reserve estimation involves large amounts of geological data, and the traditional manual computation method is a heavy and fussy work. Computing mineral reserve with 3D orebody modeling can increase the efficiency heavily of reserve estimation and management. On the basis of analyzing several 3D orebody modeling methods, this paper choose tetrahedral mesh model to construct orebody, and introduces a mineral reserve estimation method based on neural network. The main advantages of this technique are an according management for orebody boundary and inner grade distribution, as well as its precision. Therefore, this technique has an academic and practical value.
       more »

    • A new method: multi-factor trend regression and its application to economy forecast in Jiangxi

      Research Article in 1st International ICST Workshop on Knowledge Discovery and Data Mining

      Ding Yuechao

      Abstract
      The principle of a new method called Trend Regression is introduced and applied to the economy forecast of Jiangxi Province. The method improved previous time series forecasting method in which only …The principle of a new method called Trend Regression is introduced and applied to the economy forecast of Jiangxi Province. The method improved previous time series forecasting method in which only self-extension is done and multiple factors (variables) are not taken into consideration. Also, it got over the weakness of forecasting by general regression analysis that relies on simultaneous independent variables. A time series is the function of multiple factors. The values (independent variables) in a period may affect the value (dependent variable) to be predicated in the next period. The nearer the sample time to the predicted time, the more important the sample to the predict value. By shifting the dependent variable to establish models, sequential regression and prediction can be realized. In this way the trend of information can be mined.
       more »