Search

Abstract: There is a large value in making Pulp and Paper mills more energy efficient. ABB has developed an energy assessment service where opportunities to save energy in the paper machine are identified. The energy assessment is done by quantifying energy flows, benchmarking energy users, data mining and steam sensitivity analysis and by experiments and additional measurements at the paper machine. Energy quantification helped in identifying main energy consumer, benchmarking was useful to assess the gap between operating performance and best performance whereas data mining and steam sensitivity analysis helped in studying the impact of key operating variables on performance of paper machines. After the assessment an action plan was presented to the mill for energy efficiency improvement together with a return on investment. [Copyright &y& Elsevier]

In this paper Principal Component Analysis (PCA) is proposed for monitoring electric consumption of building. PCA allows modeling correlations between independent variables (weather, calendar) and energy consumption at different time scales (hourly, daily, weekly monthly). Multiway principal component analysis (MPCA) is used to model time dependencies of variables as it is commonly done in batch process monitoring. This approach allows defining simple statistic indices T 2 and SPE to be used in monitoring charts. These indices are used to detect abnormal behaviours at selected time scales. After detection, contribution analysis is performed to isolate variables responsible of such misbehaviour. Exploitation of such models, obtained during normal operating conditions, can be used to detect both faults in sensors and misbehaviours in consumption patterns with respect to independent variables. The paper presents the methodology and illustrates it in a case study focused on academic buildings situated in the Campus of the University of Girona. [ABSTRACT FROM AUTHOR]

Abstract: Data source resulted from preprocessing affects directly the quality of data mining. The methods are not the same according to particular application fields and industries. This paper describes the data preprocessing of credit card in customer segmentation, association analysis and risk detection in detail. Firstly, some tables are selected from credit card database which concerned with analysis topic. Then the paper handles with problems in selected initial data such as noisy data, missing values through data preprocessing which mainly include data cleaning, integration, transformation and reduction, and obtains training sample data needed. [Copyright &y& Elsevier]

Abstract The energy system's operation strategy has a significant influence on the system efficiency. The determination of the optimal operation strategy requires the effective application of a substantial amount of data accumulated from the operation or even forecasting process. However, data processing, which is not obtained based on the physical mechanism, may fail to explain, under certain boundary conditions. Therefore, using E-CPS to combine data mining and physical model to build a more efficient and accurate model or even a real-time model is vital and has received more attention in practical applications. This paper proposes a general process for constructing E-CPS and applies it to the simulation of building energy systems. The E-CPS results in a 70% reduction in the renewable energy waste rate of the building's energy system and a 57% reduction in the purchase of electricity. Finally, it is concluded that the use of intelligent algorithm-based data mining combined with physical models to build E-CPS is practical and effective. [ABSTRACT FROM AUTHOR]

Abstract The latest revisions in April 2018 of the 2010/31/UE and 2012/27/UE Directives on Energy Efficiency and Energy Savings respectively, point out the need of the development of smart energy indexes for buildings with the aim to (i) supervise the energy consumption on the building sector -that currently represents up to one third of the total final energy consumption- and (ii) lead the appropriate actions to transform the current buildings stock to nearly Zero Energy Buildings and Positive Energy Buildings. From public managed buildings, the Health System is the first energy consumer with great difference with other government administration sectors, such as Education or General Administration. Moreover, the energy bill has great impact on the sustainability of the public health care system. However, very few real data were available to characterize the energy demand on public buildings, which are usually the most intensive energy consumers, and efficiency indexes were usually obtained from simulation results. Nevertheless, thanks to the deployment of Smart Metering systems in the last years, it is possible to access to the true energy demand profiles of hundreds of these buildings. In this paper, with three years historical monthly electrical energy consumption data from the health system of the region of Castilla y León in Spain -including hospitals, outpatient facilities, clinics and other medical institutions- and the application of data mining techniques, an end-use electrical energy analysis was conducted to cluster the building housing according to the energy consumption into several energy use intensity clusters and, then, an average value and a Reference Building Energy Index for each cluster is proposed. Thus, a true energy labeling of these buildings based on their distance to the Reference Building Energy Index is done and presented in georeferenced maps. [ABSTRACT FROM AUTHOR]

Abstract: In order to obtain useful information from the huge amount of Automatical Identification System (AIS) data and improve the maritime safety, this paper is devoted to statistical analysis and mining of AIS data for the purpose of obtaining marine traffic knowledge. Firstly, this paper proposes a marine traffic data mining system which integrates Data Warehouse (DW) and Data Mining (DM) technologies. After pre-knowledge are introduced, the structure of the marine traffic data mining system is described in detail. Then, the protype of this system is realized. The statistical analysis method is used to access AIS DW for primary marine traffic knowledge, such as distribution by ship''s type and ship''s length in a harbor sea area. The DM tool designed by us is used to uncover the hidden knowledge. The results show that the system is steady and knowledge mined by this system is interesting to maritime safety administration and harbor administration. [Copyright &y& Elsevier]

Abstract: In data mining, outlier analysis is an important task. In recent years, many techniques have been developed for outliers mining. This paper provides an overview of the existing methods for outlier analysis. And on the base of these researches, we first propose the definition of local outlier and some formulas, and then develop a clusteringbased approach to mining local outliers. The clustering algorithm used in this paper is based on the maximal frequent itemsets in association rule mining, the properties of maximal frequent itemsets are applied to clustering. After clustering, the objects which have many common features are grouped together. By calculating the standard deviation of a cluster with respect to a certain attribute and local outlier factor of an object with respect to a certain attribute, local outliers are identified. The experimental results show that our approach is effective and appropriate to mining local outliers. [Copyright &y& Elsevier]

Abstract: The paper researched the process visualization of the data mining which is based on the latest researches and discusses in the word. The paper proposed the analyzer to change the parameter in order to participate and interact with the process of the data mining and put forward a kind of visualization model of the Apriori Algorithm depends on the comparison of a visualization methods in association rules. Therefore, The paper has played a positive role for technology research and application. [Copyright &y& Elsevier]

Abstract: Aiming at network teaching of high school, personalized learning system model is proposed in this paper. The model includes data source preprocessing, students feature modeling and data mining platform. According to major feature, gender feature, personality feature and cognitive feature, this paper proposes a student feature analysis model under network. Based on the model and applying the ID3 classification algorithm, we can make a decision to forecast whether students have the ability to pass test. Besides, it also provides the decision-making supporting for individualized teaching. [Copyright &y& Elsevier]

Abstract: The normal educational resources management has been a significant development and is more mature. However, the normal educational resources management still has many deficiencies, such as data mining, management forecasts, information regeneration. The paper focuses on the management of abnormal academic resources, such as resources manageability, network reliability, equipment availability. It is easy to cause resource wastage and equipment duplication without attention. Many studies in the past are lack of consideration in this regard, with the effective integration of various educational resources and classification management as the goal of this paper. The overall scheme of effective integration and classification management of the educational resources is designed based on Institute of Disaster Prevention. And with Microsoft Active Directory, a high efficient educational resources management system is implemented through LDAP and ADSI programming. [Copyright &y& Elsevier]

Abstract: We present a novel method for tensor completion with missing data. The problem of missing data comes up in many areas of science and engineering including data mining, machine learning, biomedical signal, and computer vision. These data are always very large and have multi-modes. Therefore, we need a well method to find the missing data of multi-way arrays (i.e., tensors). In this paper we propose an algorithm to solve the problem by tensor multilinear rank minimization. The contribution of our paper is that we convert tensor rank minimization to minimize the mode-n rank of a tensor along each mode and then formulate as low rank matrix completion for mode-n matricization of a tensor. The resulting nuclear norm related minimization problem can be efficiently solved by many recent developed methods. Our numerical results on randomly generated data demonstrate that our algorithm is quick and provides much better recoverability than a few state-of-art tensor completion algorithms. Numerical experiments on image inpainting problems demonstrate the effectiveness of our method in real world. [Copyright &y& Elsevier]

The advance in information technology has enabled the real-time monitoring and controls over building operations. Massive amounts of building operational data are being collected and available for knowledge discovery. Advanced data analytics are urgently needed to fully realize the potential of big building operational data in enhancing building energy efficiency. Data mining (DM) technology, which is renowned for its excellence in discovering hidden knowledge from massive datasets, has attracted increasing attention from the building industry. The rapid development in DM has provided powerful mining methods for extracting insights in various knowledge representations. Gradual pattern mining is a promising technique for identifying interesting patterns in big data. The knowledge discovered is represented as gradual rules, i.e., 'the more/less A, the more/less B'. It can bring special interests to building energy management by highlighting co-variations among building variables. This paper investigates the usefulness of gradual pattern mining in analysing massive building operational data. Together with the use of decision trees, motif discovery and association rule mining, a comprehensive mining method is developed to ensure the quality and applicability of the knowledge discovered. The method is validated through a case study, using the real-world data retrieved from an educational building in Hong Kong. It shows that novel and valuable insights on building operation characteristics can be obtained, based on which fault detection and optimal control strategies can be developed to enhance building operational performance. [ABSTRACT FROM AUTHOR]

While online optimal control is regarded as an efficient tool to improve the operating efficiency of air conditioning, traditional optimal control strategies utilize the so-called time-driven optimization (TDO) scheme which triggers actions by "time". Although it works well for simple air conditioning systems, several limitations are encountered when systems become more and more complex. Since TDO is a periodic scheme, it may not be suitable or efficient to react to stochastic operational changes. Recently, in order to solve those limitations, the event-driven optimization (EDO) scheme has been proposed, in which actions are triggered by "event". However, previous studies only used prior knowledge to discover important events, which could only find events for general systems, and might not comprehensive because human prior knowledge is limited after all. Moreover, prior-knowledge-based method is able to discover new knowledge. Thus, this paper presents an effective data mining approach to discover the hidden knowledge in massive data set for EDO in building air conditioning systems. Results shown that data-mining-based EDO achieves a higher energy saving with reduced computation load, in comparison with the traditional TDO. Since the data mining approach can help to automate the process of finding critical events and event threshold, it also improves the practicability of EDO. [ABSTRACT FROM AUTHOR]

This paper presents a method to automatically cluster typical days of energy consumption in one or several buildings. The method is based on an optimized version of the Symbolic Aggregate approXimation (SAX) method. SAX is a data mining technique for clustering time series with recent applications in building fault detection and building performance assessment. The number of clusters and accuracy of SAX highly depends on two highly sensitive input variables, i.e., the word size and the alphabet size. We propose the use of the genetic algorithm NSGA-II to optimize the number of words and alphabet size of SAX subjected to three fitness objectives, i.e., maximize data accuracy and compression and minimize complexity. In addition, we propose the use of MAVT as selection method of the optimal solution. The methodology is applied to measured energy consumption data of three representative buildings on a university campus in Singapore. Potential future uses of the approach include advanced studies in fault detection and calibration of urban building performance models. [ABSTRACT FROM AUTHOR]

This paper presents an intelligent scheme for high impedance fault detection using mathematical morphology and decision tree. The current signals are pre-processed using mathematical morphology and estimation of the signal features is used to generate a decision tree model. The final relaying operation based on generated data mining decision tree model. The proposed method is tested on a standard test system with a wide range of power system operating conditions. Simulation results show that the proposed method can be highly reliable in detecting high impedance fault for harmless and secured operations. [ABSTRACT FROM AUTHOR]

Today's buildings are not only energy intensive, but also information intensive. Massive amounts of operational data are available for knowledge discovery. Data mining (DM) has excellent ability in extracting insights from massive data. This paper performs a case study on the assessment of building operational performance using DM techniques. Typical DM techniques are compared and considerations for choosing specific DM techniques for the case study are presented. The methodology developed has been applied to analyze the data retrieved from a university building in Hong Kong. Useful insights have been obtained to identify typical operation patterns and energy conservation opportunities. [ABSTRACT FROM AUTHOR]

In this paper we present a collaborative data science project bringing together experts from the fields of physical and information sciences to tackle to challenges and opportunities of analysing large and rich datasets obtained from the monitoring of building integrated PV systems operating in the built environment. We present data mining analysis techniques to classify data according to environmental conditions and system performance, to distinguish between nominal and anomalous behaviour, and to identify instrumentation faults. These methods were implemented using data from the RESSOURCES project to construct evaluate the performance of a PV envelope, and to validate a simplified physical model to predict thermal and aerodynamic behaviour. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, we propose a transmission control protocol, TCP, for data mining in WSNs. By using collision notification, when the collision occurs, TCP reduces the energy consumption in data mining. Performance of data mining application improves since the remainder of sensory mined data is not sent unnecessarily. Instead, this channel is released for more productive transmissions of mined data. Therefore, TCP can efficiently handle the collision during the data mining, and conserve more energy to extend the lifetime of WSNs for data mining. Most importantly, TCP is a distributed scheme without requirement of time synchronization, so it is practical to realize TCP in real time data mining application. Moreover, TCP is evaluated in both simulation and testbed experiments, and the results indicate that TCP significantly improves mined data transmission amount and mined data delivery energy consumption. [Copyright &y& Elsevier]

Abstract: Inventory management is a significant activity in enterprise manufacture. It plays a vital role in not only the turnover of circulating fund but also the level of customer service. An increasing number of companies pay more emphasis on tradeoff between lowering inventory cost and advancing customer service level as well as effective and efficient inventory management. This paper attempts to introduce the relative theories of inventory management and conduct a case study through rationally analyzing the certain company''s industry background, business characteristics, material category and inventory features. Finally it concludes on proposing a suitable improved method about inventory management according to its own situation. [Copyright &y& Elsevier]

Abstract: Starting with the practical significance of search engine, this paper introduced the classification of search engine and analyzed its advantages and disadvantages, focused on the basic system structure and key technologies in its individual modules, finally predicted the trends for it. [Copyright &y& Elsevier]

Abstract: This article studies the personalized modeling techniques, presents a personalized client-based model and the user''s access history page as mining object without the involvement of too many users. Automatic feedback is from the user who derived implicit in user''s interest. This paper focuses on the user interest mining-related technology to the user''s visitation page content as interest in the historical sources of information modeling, which are the use''s structural features of HTML page content to extract the important part of the sub-string matching that is based on words and the statistical method of combining segmentation. The word content of the page segmentation is used to express the thematic content of the page, and then it removes the segmentation results to stop words, use this theme in the page vector space model and represent the characteristics of words. With word frequency, position, and the weighted combination of the nonlinear function, it calculates the weights. [Copyright &y& Elsevier]

Abstract: FP-Growth is a recursive mining algorithm for frequent patterns. With each recursion, a new FP-tree and the corresponding header table are built. This paper proposes another approach to improve the mining efficiency: only one FP-tree is constructed, and instead of building new trees in every recursion, builds new header tables; frequent patterns are generated along the way, so we call this algorithm “header table recursion (HTR)” method. Tests show that it will get better performance for relative small datasets because the compress of the FP-tree has no significant impact on the performance yet inflicts extra time cost in tree creation operations. However on larger datasets, the traversal of the original FP-tree will become more time-consuming and its overall performance needs further testing. [Copyright &y& Elsevier]

Abstract: Flip links are widely used in web pages in some popular scenarios, which are very useful for many applications, such as focused crawling and query result extraction. However, automatically extracting flip link blocks from web pages is not a trivial problem due to the diversity on both DOM tree structure and visual presentation. In this paper, we propose a practical solution for automatically extracting the flip link block from any web page. This solution consists of three steps: first segment each web page into a set of blocks based on the visual information, then identifies the block(s) that contain the flip link block using a machine learning approach, and finally extract the hyperlinks pointing to next pages from the identified block(s) by comparing the patterns in multiple blocks. Experimental results indicate that this technique is highly accurate. [Copyright &y& Elsevier]

Abstract: With the help of data mining software SPSS Clementine12.0, taking Apriori and GRI model as the way of data mining, Ph.D borrow records of Textile Department of Donghua University was analyzed in this paper. After data mining,some related knowledge in different disciplines can be found. It can give some guides on teaching and optimizing the discipline construction. [Copyright &y& Elsevier]

Abstract: Electroencephalogram EEG and MEG research are the non-invasive brain imaging techniques. The main research measured by the scalp EEG or MEG brain source signals can obtain three-dimensional quantitative information. Structure to match with the actual situation is the first model to solve EEG forward, the inverse problem of the clinical effectiveness of pre-conditions and key. Model of the head and the head are on the premise for the brain magnetic resonance image contour extraction and segmentation. Since the first complex internal structure, the access to the brain volume data segmentation is more difficult. In this paper, we propose a simulated annealing-based deformable model method and realize the head T2 weighted magnetic resonance images of the cross-sectional area of the extraction of the brain. To avoid falling into the local minimum in the deformation of the deformable model, we use a simulated annealing process of thinking, making the algorithm with the global search capability. The experimental results show that the improved algorithm is effective and robust. [Copyright &y& Elsevier]

Abstract: The development of computer network technology and distributed database technology have promoted data distributed storage and the new technical generation of distributed data mining. Distributed data mining uses distributed computing and find the needed knowledge of users from the distributed database, it has wide application prospect. This paper studies M - Agent technology application mechanism in data mining. It has summarized the development of Agent technology and data mining technology, and analyzed the advantage of M - Agent combining with distributed data mining, then has designed a distributed data mining modeling based on M - Agent technology, and studied the mining method of this model on association rules in distributed database. [Copyright &y& Elsevier]

Abstract: To improve the response speed of landslide hazard and consummate disaster warning mechanism, adapt to the disaster-affected trends, based on analysis of the parallel coordinate visual data mining and disaster warning, this paper presents a method from the main to each based on parallel coordinates to solve the problems existing in landslide early warning, and does some research on early warning mechanisms and processes of this method, applying the method to test landslide early warning induced by rainfall. The experiment results shows this method is also applicable to the cause of earthquakes and human activity disaster analysis, which can simplify the warning process, improve the power of landslide early warning, and build a solid base for the future disaster warning. [Copyright &y& Elsevier]

Abstract: This paper presents a proposal for an architecture that integrates knowledge discovery systems (automatic acquisition) and knowledge based systems (experts systems). This work formulates considerations over the viability of the implementation of this architecture according to the advance of the technologies involved.. [Copyright &y& Elsevier]

Abstract: As e-commerce is becoming more and more popular and interactive, Merchants selling products on the Web 2.0 often ask their customers to review the products which they have purchased. The number of reviews can be in hundreds or even thousands for a popular product. This makes it difficult for a potential customer to read them or to make an informed decision on whether to purchase the product. It also makes it difficult for the manufacturer of the product to keep track of customer opinions. In this research, we aim to mine and to analyze all the customer reviews of a product. The algorithm proposed in this research is more reliable on opinion identification because it is unsupervised and the accuracy of the result improves as the number of reviews increases. Our research is performed in four steps: (1) mining the frequent aspects of a product that have been commented on by customers; (2) mining the infrequent aspects of a product which have been commented by customers (3) identifying opinion words in each review and deciding whether each opinion word is positive, negative or neutral; (4) summarizing the results. This paper proposes several novel techniques to perform these tasks. Our experimental results using reviews of a number of products sold online demonstrate the effectiveness of the techniques. [Copyright &y& Elsevier]

Abstract: Unstructured information such as text, images, audio contains a great deal of knowledge, how to efficiently acquire these knowledge from the massive unstructured information automatically is a urgent problem. This paper presents an approach of Large-scale text analysis which is based on UIMA and Hadoop, reusing lots of analysis engines to fulfill information extraction with annotations. This approach is scalable, efficient, secure and easy to realize. The prototypical implementation has been tested and verified in real scenarios, and experimental results shows it is feasible and effective. [Copyright &y& Elsevier]

Abstract: The method of multiple neural network Fusion using Fuzzy Integral (MNNF) presented by this paper is to improve the detection performance of data mining-based intrusion detection system. The basic idea of MNNF is to mine on distinct feature training dataset by neural networks separately, and detect TCP/IP data by different neural networks, and then nonlinearly combine the results from multiple neural networks by fuzzy integral. The experiment resultsshow that this technique is superior to single neural networks for intrusion detection in terms of classification accuracy. Compared with other combination methods such as Majority, Average, Borda count, fuzzy integral is better than one of them. [Copyright &y& Elsevier]

Abstract: With the rapid development of Internet, how to accurately understand the behavior of users on visiting the Web site is a pressing issue, it belongs to Web data mining. To find the user traversal interest path is an important research field, in this paper, a new algorithm based on optimized ACA is proposed. The main idea of this algorithm is: The Web users are regarded as artificial ants, and user''s navigation patterns can be found by optimized ACA. First, a Web site model is built, then, their preferred navigation paths are found on use of optimized ACA. Experimental results show that the method can accurately reflect the user''s traversal interest paths. [Copyright &y& Elsevier]

Abstract: Through in-depth studies of the existing intrusion detection systems, for the shortcomings of low precision, instability and high false positive existed in the current intrusion detection, this paper applies data mining technique to intrusion detection system, and according to common intrusion detection framework, designs an intrusion detection system based on data mining. Experimental results show that this detection scheme can greatly improve the detection accuracy of intrusion detection system. [Copyright &y& Elsevier]

Abstract: The same content data transferred repeatedly between two network nodes in the Internet. It is defined as redundant network traffic. In this paper, we conducted network packets analysis on application layer, analyzed network resource location and data size accurately, and studied the traffic characters of network redundancy. Thus it can provide better analysis of network users’ behaviors and personalized web services. Additionally, we hope it can provide reference for the development of network technology, especially for the development of new standards and protocols in China Next Generation Internet (CNGI). [Copyright &y& Elsevier]

Abstract: Entity relation extraction is the task of finding semantic relations between two entities from text. In this paper, wepropose a novel Web-mining-based Chinese entity relation extraction approach that explicitly defines and explores the Web mining process and relation feature words. The results of the research indicate this method is promising. [Copyright &y& Elsevier]

Abstract: In this paper, a novel method, KFP-SVM, of generating synthetic samples with kernel model is proposed to deal with imbalanced data sets. In this method, the majority samples are selected by removing those points which invade the minority samples, and the minority samples are mapped into a high-dimension space with a kernel function which is called kernel space. Then the distance of the minority samples in kernel space is calculated. The nearest neighbors of every minority samples are found out and linear combination is used to generate synthetic samples. Finally these synthetic samples are mapped inverse into input space, so that the primary synthetic samples and a SVM model will be obtained. A group of experiments in UCI data sets have shown that this novel method with kernel model to generate the synthetic samples could improve the performance of the SVM under the criterion of F-Measure and AUC, compared with SVM,SMOTE-SVM,DEC-SVM. [Copyright &y& Elsevier]

Abstract: Clustering is a method of unsupervised learning, and a common technique for intelligent data analysis used in many fields. The enlarging volumes of information produced by the progress of technology, makes clustering of very large scale data a challenging task. Therefore, parallel processing of very large scale data set is a good solution to that problem. CLARANS is approved to be an efficient and effective clustering method for spatial data mining. It is insensitive to outliers and has the advantage of processing category data directly compared to other partition method like K-means. However, it is a serial algorithm and cannot process large scale data set efficiently. In this paper, we propose a parallel CLARANS clustering algorithm based on MapReduce, which is a current and powerful parallel programming framework. Experimental results demonstrate that the parallel CLARANS algorithm can handle large scale data set efficiently and effectively. Furthermore, the evaluation criterions of scaleup, speedup and sizeup validate its efficiency. [Copyright &y& Elsevier]

Abstract: This paper presents a novel solution, data cutting and sorting method (DCSM), to improve the efficiency of Apriorialgorithm. By means of excising and sorting processes, DCSM has reduced the redundant data and simplified the search procedure. The calculation time of DCSM is less 40 times than Apriori algorithm. In addition, the empirical hospital utilizes DCSM to analyze the data of health examination and the records of out-patient department (OPD). According to the relationships between the condition of health examination and the diagnosis department of OPD, the empirical hospital constructs the disease forecast system to provide the health information of the patients to physician for fellow-up diagnosis. [Copyright &y& Elsevier]

Abstract: Issues about health care of the pilot are the most important part in civil aviation medicine area. In this paper, we propose a method which uses the data analysis techniques to study issues associated with medical certification decisions and their effects on the Chinese pilot population to ensure safety of flight. First of all, based on the data collected by the civil aviation medicine certification management information system, we build a civil aviation medicine data warehouse. Then we discuss the problems occurred during the process of data cleaning. Finally, we adopt a decision tree method based on the association function value to complete the data mining and analyze the results. [Copyright &y& Elsevier]

Abstract: Web search engines have been helping users to locate information they need on the web. But they are far from optimal in that they only return general search results for different people who issue the same query to them, regardless of the differences in interest among individuals. In this paper we try to explore how different people perceive search results for the same query by mining clicks, which acts as a proxy for relevance, from a large body of American Online (AOL) query log data. In an attempt to investigate the variability in what people are searching for when they issue the same query, we only select queries from query log for which we have clicks from at least eight individuals. Extensive analysis of clicks on search results for the same query reveals that there are variations in judgments across individuals as shown in the potential for personalization curve. It shows that there is an observable gap between individual preferences and best group preferences for the results of the same query and this gap increases as the number of people in the group increases. This tells us how much room there is to improve the search result through personalization. [Copyright &y& Elsevier]

Abstract: Data warehouse and Data Mining Techniques can be very useful to retail industry deciders. A retail industry in zaozhuang city was studied in this paper. The star model of the retail data warehouse is designed and developed. By analyzing and mining various kinds of information of retail industry using the Data Mining Techniques, can provide useful information to improve the retail industry''s competitive ability. [Copyright &y& Elsevier]

Abstract: In this paper, secure audio forensic mark algorithm to trace the illegal distribution is proposed. The proposed algorithm uses off-axis hologram when make the forensic mark for embedding into the audio. Embedding and extracting is performed in the DWT domain, and the Arnold transform of the forensic mark is used for improving the security. The hologram generator makes the hologram of the Arnold transformed mark. Because of the characteristics of the off-axis hologram, the proposed algorithm can embed high capacity information and extract without the original signal. [Copyright &y& Elsevier]

Abstract: The real estate information system in China has improved dramatically since 2005. The housing transactions databases in most cities not only keep comprehensive records of sales contracts, but also store many physical and social attributes of the sold housing units. In this paper, a useful data mining application is introduced, which can discover the relationship between house price and its attributes, and find out obviously higher or lower contract prices. Local governments can use this application to look up typographical errors as well as detect fraud contracts. The software has been deployed in Beijing and runs fairly well in real practice. [Copyright &y& Elsevier]

Abstract: No a detection method can detect all of intrusion behaviors, a perfect IDS should be a unity of multiple detection means. This paper''s object is to establish an intrusion detection system model based on data mining technology, which can detect known intrusion effectively and has a good capacity to recognize unknown data pattern which can’t be detected effectively in traditional IDS. The experiments proved that the method is effective to detect intrusion such as scanning and DoS. [Copyright &y& Elsevier]

Abstract: The paper presents a new intrusion detection model combining misuse detection and anomaly detection mode, and makes a research into the key technology of the model based on data mining theory. In the model, the association rules data mining algorithm is applied to establish abnormal behavior rule set for misuse detection to detect known intrusion rapidly. And the minimum dissimilarity clustering analysis algorithm is used to establish normal behavior rule set for anomaly detection to detect new unknown intrusion. Research of the model of intrusion detection based on data mining is made on KDD99 dataset. The experiment shows that the new model can improve true positives, decrease false positives and detect new intrusions. [Copyright &y& Elsevier]

Abstract: The paper analyzed the present situation of modern distance education and then described the necessity that carried out distance education personalized services firstly. It constructed a personalized modern distance education model based on web data mining and described the basic design idea and functions in detail. For the distance education organization, this model will provide a certain theoretical basic and reference value to build a personalized modern distance education system. [Copyright &y& Elsevier]

Abstract: Most of the current detection models for software behavior focus on the integrity detecting of its running path, but they rarely consider detecting the function realization integrity in each stage of the software running path. Therefore, dynamic credibility detection model based on scene mining for software behavior called SCDM is proposed in this paper. First of all, this model describes the integrity of various stages’ function implementation by introducing the concept of scene. Secondly, the model used association rules and sequence pattern algorithm in scene mining. Both of the algorithms can extract the hidden and valuable association information from the captured scene information set. And the mining result is used for credibility detection of software behavior. Finally, the results of simulation experiments indicate that CDMBSM model has a good detection performance. [Copyright &y& Elsevier]

Abstract: Intrusion detection using sequential pattern mining is a research focusing on the field of information security. This paper first introduces several common sequential pattern mining algorithms, and then expounds its current development with comparisons about the merits and shortcomings with the current mainstream technologies, at the same time, presenting some further discussion upon the future direction of the sequential pattern mining algorithms based intrusion detection. [Copyright &y& Elsevier]

Abstract: During the last decades, spatio-temporal data sets have drawn great attention in research areas and industrial applications. For the purpose of building better predictive models for spatio-temporal data analysis tasks, the data sets are mainly analyzed by traditional Data Mining (DM) tools and Geographical Information Systems (GIS) cooperatively. Usually the geospatial attributes of data need to be generated manually by analyst from GIS. When build/train a model and use the trained model to do prediction, the lack of automation of the geospatial attributes generation makes the task inefficient and hardly repeatable. Still, there are very few tools that can perform spatiotemporal data analysis tasks alone and automate the above procedure. In this paper, we propose a service-based spatio-temporal data analysis framework, which can leverage existing data mining tools and geographical information systems, and perform spatio-temporal data analysis tasks in an integrative manner. In this framework, a data analysis task is divided into traditional data mining tasks and geospatial related tasks, each of which is executed on an existing tool/software. The entire procedure is automated by a bundle of web services. User can analyze data efficiently and repeatably by simply accessing the service. We also develop a discription model based on the Predictive Model Markup Language (PMML) and Keyhole Markup Language (KML) to describe the geospatial related data mining models in our framework. This enables information exchanging capability with other data mining tools and geospatial rendering services. A case study on spatial association rule mining demonstrates the usability of the framework. [Copyright &y& Elsevier]

Abstract: This paper proposes a new approach for frequent itemsets mining which utilizes evolutionary autonomous agents. The optimality of frequent itemsets extraction is to find all the frequent itemsets in the itemset space whose support values are greater than the support threshold. In the presented approach, the autonomous agents, being distributed computational entities, operate in N-D lattice of the itemset space and exhibit a number of reactive behaviors. In order to effectively locate the frequent itemsets, the agents were designed to sense the local stimuli from their environment by means of evaluating the support value of local points, and accordingly activate their behaviors. The experiment results have shown that the proposed algorithm is efficient and effective. [Copyright &y& Elsevier]