Search

• HPSO-SSM is proposed based on original particle swarm optimization. • A new wrapper-based feature selection approach based on HPSO-SSM is proposed. • The logistic map sequence is used to enhance the diversity in the search process. • An innovative position update model is presented to improve the position quality. • Our method outperforms seventeen extremely competitive methods in terms of accuracy. The "curse of dimensionality" is one of the largest problems that influences the quality of the optimization process in most data mining, pattern recognition, and machine learning tasks. Using high-dimensional datasets to train a classification model may reduce the generalization performance of the learned model. In addition, high dimensionality of the dataset results in high computational and memory costs. Feature selection is an important data preprocessing approach in many practical application domains that are relevant to expert and intelligent systems. Feature selection aims at selecting a subset of informative and relevant features from an original feature dataset. Therefore, using a feature selection approach to process the original data prior to the learning process is essential for enhancing the performance on the classification task. In this paper, hybrid particle swarm optimization with a spiral-shaped mechanism (HPSO-SSM) is proposed for selecting the optimal feature subset for classification via a wrapper-based approach. In HPSO-SSM, we make three improvements: First, a logistic map sequence is used to enhance the diversity in the search process. Second, two new parameters are introduced into the original position update formula, which can effectively improve the position quality of the next generation. Finally, a spiral-shaped mechanism is adopted as a local search operator around the known optimal solution region. For a complete evaluation, the proposed HPSO-SSM method is compared with six state-of-the-art meta-heuristic optimization algorithms, ten well-known wrapper-based feature selection techniques, and six classic filter-based feature selection methods. Various assessment indicators are used to properly evaluate and compare the performances of these approaches on twenty classic benchmark classification datasets from the UCI machine learning repository. According to the experimental results and statistical tests, the developed methods effectively and efficiently improve the classification accuracy compared with other wrapper-based approaches and filter-based approaches. The results demonstrate the high performance of the HPSO-SSM method in searching the feasible feature space and selecting the most informative attributes for solving classification problems. Therefore, the HPSO-SSM method has broad application prospects as a new feature selection approach. [ABSTRACT FROM AUTHOR]

Highlights • A business focused overview of the current state of marketing analytics. • An integrative review at the boundary of expert systems, marketing science, statistics, and operations research. • The topics of visualization, segmentation, and class prediction are featured. Practical implementation advice is given, along with a directory of open source R routines. Abstract Marketing analytics is a diverse field, with both academic researchers and practitioners coming from a range of backgrounds including marketing, expert systems, statistics, and operations research. This paper provides an integrative review at the boundary of these areas. The aim is to give researchers in the intelligent and expert systems community the opportunity to gain a broad view of the marketing analytics area and provide a starting point for future interdisciplinary collaboration. The topics of visualization, segmentation, and class prediction are featured. Links between the disciplines are emphasized. For each of these topics, a historical overview is given, starting with initial work in the 1960s and carrying through to the present day. Recent innovations for modern, large, and complex "big data" sets are described. Practical implementation advice is given, along with a directory of open source R routines for implementing marketing analytics techniques. [ABSTRACT FROM AUTHOR]

Highlights • A data mining method for topology optimization is proposed. • The method sequentially applies clustering and association rule analysis. • A distance measure for pairs of configurations is defined. • Confidence and lift indexes in multiobjective topology optimizations are defined. • The method is applied in 3- and 4-objective design problems in 2 and 3 dimensions. Abstract Optimum design problems, including structural optimization problems, often include multiple objectives. A multiobjective optimization problem usually provides a number of optimal solutions, called non-dominated solutions or Pareto-optimal solutions. In multiobjective topology optimization scenarios, decision makers face the challenging task of choosing the most effective solution that meets their needs; serial comparisons among a set of Pareto-optimal solution are cumbersome, as are trial-and-error attempts to find an appropriate solution among a host of alternatives. On the other hand, the recent integration of data mining techniques in multiobjective optimization methods can provide decision makers with important, highly pertinent, and useful knowledge. In this paper, we propose a data mining technique for knowledge discovery in multiobjective topology optimization. The proposed method sequentially applies clustering and association rule analysis to a Pareto-optimal solution set. First, clustering is applied in the design space and the result is then visualized in the objective space. After clustering, detailed features in each cluster are analyzed based on the concept of association rule analysis, so that characteristic substructures can be extracted from each cluster of solutions. In four numerical examples, we demonstrate that the proposed method provides pertinent knowledge that aids comprehension of the key substructures responsible for one or more desired performances, thereby giving decision makers a useful tool for discovery of particularly effective design solutions. [ABSTRACT FROM AUTHOR]

Highlights • Models for diagnosis aid may provoke distrust among practitioners and patients. • The development of models for diagnosis aid falls under data mining concerns. • The survey raises an expert-aware approach for suitable diagnosis aid. • Interpretability is a major concern for the acceptance of diagnosis aid models. • The recourse to data mining for diagnosis aid poses many technical challenges. Abstract Data mining continues to play an important role in medicine; specifically, for the development of diagnosis aid models used in expert and intelligent systems. Although we can find abundant research on this topic, clinicians remain reluctant to use decision support tools. Social pressure explains partly this lukewarm position, but concerns about reliability and credibility are also put forward. To address this reticence, we emphasize the importance of the collaboration between both data miners and clinicians. This survey lays the foundation for such an interaction, by focusing on the specifics of diagnosis aid, and the related data modeling goals. On this regard, we propose an overview on the requirements expected by the clinicians, who are both the experts and the final users. Indeed, we believe that the interaction with clinicians should take place from the very first steps of the process and throughout the development of the predictive models, thus not only at the final validation stage. Actually, against a current research approach quite blindly driven by data, we advocate the need for a new expert-aware approach. This survey paper provides guidelines to contribute to the design of daily helpful diagnosis aid systems. [ABSTRACT FROM AUTHOR]

In knowledge discovery, experts frequently need to combine knowledge from different domains to get new insights and derive new conclusions. Intelligent systems should support the experts in the search for relationships between concepts from different domains, where huge amounts of possible combinations require the systems to be efficient but also sufficiently general, open and interactive to enable the experts to creatively guide the discovery process. The paper proposes a cross-domain literature mining methodology that achieves this functionality by combining the functionality of two complementary text mining tools: clustering and topic ontology creation tool OntoGen and cross-domain bridging terms exploration tool CrossBee. Focusing on outlier documents identified by OntoGen contributes to the efficiency, while CrossBee allows for flexible and user-friendly bridging concepts exploration and identification. The proposed approach, which is domain independent and can support cross-domain knowledge discovery in any field of science, is illustrated on a biomedical case study dealing with Alzheimer’s disease, one of the most threatening age-related diseases, deteriorating lives of numerous individuals and challenging the ageing society as a whole. By applying the proposed methodology to Alzheimer’s disease and gut microbiota PubMed articles, we have identified Nitric oxide synthase (NOS) as a potentially valuable link between these two domains. The results support the hypothesis of neuroinflammatory nature of Alzheimer’s disease, and is indicative for the quest for identifying strategies to control nitric oxide-associated pathways in the periphery and in the brain. By addressing common mediators of inflammation using literature-based discovery, we have succeeded to uncover previously unidentified molecular links between Alzheimer’s disease and gut microbiota with a multi-target therapeutic potential. [ABSTRACT FROM AUTHOR]

Abstract: After credit has been granted, lenders use behavioural scoring to assess the likelihood of default occurring during some specific outcome period. This assessment is based on customers’ repayment performance over a given fixed period. Often the outcome period and fixed performance period are arbitrarily selected, causing instability in making predictions. Behavioural scoring has failed to receive the same attention from researchers as application scoring. The bias for application scoring research can be attributed, in part, to the large volume of data required for behavioural scoring studies. Furthermore, the commercial sensitivities associated with such a large pool of customer data often prohibits the publication of work in this area. This paper focuses on behavioural scoring and evaluates the contrasting effects of altering the performance period and outcome period using 7-years worth of data from the Irish market. The results of this work indicate that a 12-month performance period yields an easier prediction task when compared with other historical payment periods of varying lengths. This article also quantifies differences in the classification performance of logistic regression arising from different outcome periods length. Our findings show that the performance of a logistic regression classifier degrades significantly when the outcome window is extended beyond 6-month. Finally we consider different approaches to how the concept of default is defined. Typically whether the customer is identified as a default risk or not is set based on either (i) whether the account is in default at the end of the outcome period or (ii) at any time during the outcome period. This paper studies both approaches and finds that the latter approach resulted in an easier classification problem, that is, it gives the highest assurance that the classification will be correct. [Copyright &y& Elsevier]

Large graphs are scale free and ubiquitous having irregular relationships. Clustering is used to find existent similar patterns in graphs and thus help in getting useful insights. In real-world, nodes may belong to more than one cluster thus, it is essential to analyze fuzzy cluster membership of nodes. Traditional centralized fuzzy clustering algorithms incur high communication cost and produce poor quality of clusters when used for large graphs. Thus, scalable solutions are obligatory to handle huge amount of data in less computational time with minimum disk access. In this paper, we proposed a parallel fuzzy clustering algorithm named ‘PGFC’ for handling scalable graph data. It will be advantageous from the viewpoint of expert systems to develop a clustering algorithm that can assure scalability along with better quality of clusters for handling large graphs.The algorithm is parallelized using bulk synchronous parallel (BSP) based Pregel model. The cluster centers are initialized using degree centrality measure, resulting in lesser number of iterations. The performance of PGFC is compared with other state of art clustering algorithms using synthetic graphs and real world networks. The experimental results reveal that the proposed PGFC scales up linearly to handle large graphs and produces better quality of clusters when compared to other graph clustering counterparts. [ABSTRACT FROM AUTHOR]

Data clustering has been proven to be an effective method for discovering structure in medical datasets. The majority of clustering algorithms produce exclusive clusters meaning that each sample can belong to one cluster only. However, most real-world medical datasets have inherently overlapping information, which could be best explained by overlapping clustering methods that allow one sample belong to more than one cluster. One of the simplest and most efficient overlapping clustering methods is known as overlapping k-means (OKM), which is an extension of the traditional k-means algorithm. Being an extension of the k-means algorithm, the OKM method also suffers from sensitivity to the initial cluster centroids. In this paper, we propose a hybrid method that combines k-harmonic means and overlapping k-means algorithms (KHM-OKM) to overcome this limitation. The main idea behind KHM-OKM method is to use the output of KHM method to initialize the cluster centers of OKM method. We have tested the proposed method using FBCubed metric, which has been shown to be the most effective measure to evaluate overlapping clustering algorithms regarding homogeneity, completeness, rag bag, and cluster size-quantity tradeoff. According to results from ten publicly available medical datasets, the KHM-OKM algorithm outperforms the original OKM algorithm and can be used as an efficient method for clustering medical datasets. [ABSTRACT FROM AUTHOR]

Abstract: In order to determine how data mining techniques (DMT) and their applications have developed, during the past decade, this paper reviews data mining techniques and their applications and development, through a survey of literature and the classification of articles, from 2000 to 2011. Keyword indices and article abstracts were used to identify 216 articles concerning DMT applications, from 159 academic journals (retrieved from five online databases), this paper surveys and classifies DMT, with respect to the following three areas: knowledge types, analysis types, and architecture types, together with their applications in different research and practical domains. A discussion deals with the direction of any future developments in DMT methodologies and applications: (1) DMT is finding increasing applications in expertise orientation and the development of applications for DMT is a problem-oriented domain. (2) It is suggested that different social science methodologies, such as psychology, cognitive science and human behavior might implement DMT, as an alternative to the methodologies already on offer. (3) The ability to continually change and acquire new understanding is a driving force for the application of DMT and this will allow many new future applications. [Copyright &y& Elsevier]

Abstract: Association rule mining and classification are important tasks in data mining. Using association rules has proved to be a good approach for classification. In this paper, we propose an accurate classifier based on class association rules (CARs), called CAR-IC, which introduces a new pruning strategy for mining CARs, which allows building specific rules with high confidence. Moreover, we propose and prove three propositions that support the use of a confidence threshold for computing rules that avoids ambiguity at the classification stage. This paper also presents a new way for ordering the set of CARs based on rule size and confidence. Finally, we define a new coverage strategy, which reduces the number of non-covered unseen-transactions during the classification stage. Results over several datasets show that CAR-IC beats the best classifiers based on CARs reported in the literature. [Copyright &y& Elsevier]

The rapidly growing world energy use already has concerns over the exhaustion of energy resources and heavy environmental impacts. As a result of these concerns, a trend of green and smart cities has been increasing. To respond to this increasing trend of smart cities with buildings every time more complex, in this paper we have proposed a new method to solve energy inefficiencies detection problem in smart buildings. This solution is based on a rule-based system developed through data mining techniques and applying the knowledge of energy efficiency experts. A set of useful energy efficiency indicators is also proposed to detect anomalies. The data mining system is developed through the knowledge extracted by a full set of building sensors. So, the results of this process provide a set of rules that are used as a part of a decision support system for the optimisation of energy consumption and the detection of anomalies in smart buildings. [ABSTRACT FROM AUTHOR]

Abstract: Using a bibliometric approach, this paper analyzes research trends and forecasts of data mining from 1989 to 2009 by locating heading “data mining” in topic in the SSCI database. The bibliometric analytical technique was used to examine the topic in SSCI journals from 1989 to 2009, we found 1181 articles with data mining. This paper implemented and classified data mining articles using the following eight categories—publication year, citation, country/territory, document type, institute name, language, source title and subject area—for different distribution status in order to explore the differences and how data mining technologies have developed in this period and to analyze technology tendencies and forecasts of data mining under the above results. Also, the paper performs the K-S test to check whether the analysis follows Lotka’s law. Besides, the analysis also reviews the historical literatures to come out technology diffusions of data mining. The paper provides a roadmap for future research, abstracts technology trends and forecasts, and facilitates knowledge accumulation so that data mining researchers can save some time since core knowledge will be concentrated in core categories. This implies that the phenomenon “success breeds success” is more common in higher quality publications. [Copyright &y& Elsevier]

Abstract: This paper is a contribution to theoretical background of data mining, more precisely to fuzzy association analysis. We consider three the most commonly used confirmation measures and we study relations among found and known associations given by them. Good understanding of such relationships is necessary for creating more efficient algorithms or for subsequent work with found associations as well as for cooperation with the consumer of the data mining process. Even if our motivation to this work arose from mining of linguistic associations, found properties that coincide with semantics of mined associations are valid in general. Additionally, some examples showing how to use obtained properties are also contained in this paper. [Copyright &y& Elsevier]

Abstract: One-Versus-All (OVA) classification is a classifier construction method where a k-class prediction task is decomposed into k 2-class sub-problems. One base model is constructed for each sub-problem and the base models are then combined into one model. Aggregate model implementation is the process of constructing several base models which are then combined into a single model for prediction. In essence, OVA classification is a method of aggregate modeling. This paper reports studies that were conducted to establish whether OVA classification can provide predictive performance gains when large volumes of data are available for modeling as is commonly the case in data mining. It is demonstrated in this paper that firstly, OVA modeling can be used to increase the amount of training data while at the same time using base model training sets whose size is much smaller than the total amount of available training data. Secondly, OVA models created from large datasets provide a higher level of predictive performance compared to single k-class models. Thirdly, the use of boosted OVA base models can provide higher predictive performance compared to un-boosted OVA base models. Fourthly, when the combination algorithm for base model predictions is able to resolve tied predictions, the resulting aggregate models provide a higher level of predictive performance. [Copyright &y& Elsevier]

Abstract: In a recent paper by Toloo et al. [Toloo, M., Sohrabi, B., & Nalchigar, S. (2009). A new method for ranking discovered rules from data mining by DEA. Expert Systems with Applications, 36, 8503–8508], they proposed a new integrated data envelopment analysis model to find most efficient association rule in data mining. Then, utilizing this model, an algorithm is developed for ranking association rules by considering multiple criteria. In this paper, we show that their model only selects one efficient association rule by chance and is totally depended on the solution method or software is used for solving the problem. In addition, it is shown that their proposed algorithm can only rank efficient rules randomly and will fail to rank inefficient DMUs. We also refer to some other drawbacks in that paper and propose another approach to set up a full ranking of the association rules. A numerical example illustrates some contents of the paper. [Copyright &y& Elsevier]

Abstract: The detection of non-technical losses (NTLs), in most papers, commonly deals with the utilization of the registered consumption for each customer; besides, some researchers used the economic activity, the active/reactive ratio and the contract power. Currently, utility company databases store enormous amounts of information on both installations and customers: consumption, technical information on the measure equipment, documentation, inspections results, commentaries of inspectors, etc. In this paper, an integrated expert system (IES) for the analysis and classification of all the available useful information of the customer is presented. Customer classification identifies the presence of an NTL and the problem type. This IES include several modules: text mining module for analysis of inspector commentaries and extraction of additional information on the customer, data mining module to draw up the rules that determine the customer estimate consumption, and the Rule Based Expert System module to analyze each customer using the results of the text and data mining modules. This IES is used with real data extracted from Endesa company databases. Endesa is the most important power distribution company in Spain, and one of the most significant companies of Europe. This IES is used in the test phase by human experts in the Endesa company. In this phase, the IES is used as a Decision Support System (DSS), as it contains another module which provides a report with additional information about the customer and a summarized result that the inspectors can use to reach a decision. [Copyright &y& Elsevier]

Abstract: An efficient expert system for the power transformer condition assessment is presented in this paper. Through the application of Duval’s triangle and the method of the gas ratios a first assessment of the transformer condition is obtained in the form of a dissolved gas analysis (DGA) diagnosis according IEC 60599. As a second step, a knowledge mining procedure is performed, by conducting surveys whose results are fed into a first Type-2 Fuzzy Logic System (T2-FLS), in order to initially evaluate the condition of the equipment taking only the results of dissolved gas analysis into account. The output of this first T2-FLS is used as the input of a second T2-FLS, which additionally weighs up the condition of the paper-oil system. The output of this last T2-FLS is given in terms of words easily understandable by the maintenance personnel. The proposed assessing methodology has been validated for several cases of transformers in service. [Copyright &y& Elsevier]

Abstract: On-line analytical processing (OLAP) is a common solution that modern enterprises use to generate, monitor, share, and administrate their analysis reports. When daily, weekly, and/or monthly reports are generated or published by the OLAP operators, all analyses on the contents of reports are left for the report readers. To discover hidden rules, similar reports, or trend inside the potentially huge amount of reports, the report readers can only rely on their smart eyes to find out any rules of such kinds. Data mining is a well-developed field for finding hidden rules inside the data itself. However, there are few techniques focus on finding hidden rules, similarity, or trend using OLAP reports as the unit of analysis. In this paper, we explore how to use data mining techniques on OLAP reports in order to automatically and effectively find the similarity knowledge of OLAP reports. We also address the appropriate presentation of this similarity knowledge to OLAP users. We compare the difference between traditional data mining and finding similarity knowledge from OLAP reports. We then proposed three methods (called OLAP_MDS, OLAP_CLU, and OLAP_M+C in this paper) to explore the effectiveness of discovering similarity knowledge from OLAP reports. Finally, we compare the pros and cons of the proposed three methods with experiments and conclude that the OLAP_M+C method should be the best in most cases. [ABSTRACT FROM AUTHOR]

Abstract: The types of police inquiries performed these days are incredibly diverse. Often data processing architectures are not suited to cope with this diversity since most of the case data is still stored as unstructured text. In this paper Formal Concept Analysis (FCA) is showcased for its exploratory data analysis capabilities in discovering domestic violence intelligence from a dataset of unstructured police reports filed with the Amsterdam-Amstelland police in the Netherlands. From this data analysis it is shown that FCA can be a powerful instrument to operationally improve policing practice. For one, it is shown that the definition of domestic violence employed by the police is not always as clear as it should be, making it hard to use it effectively for classification purposes. In addition, this paper presents newly discovered knowledge for automatically classifying certain cases as either domestic or non-domestic violence. Moreover, it provides practical advice for detecting incorrect classifications performed by police officers. A final aspect to be discussed is the problems encountered because of the sometimes unstructured way of working of police officers. The added value of this paper resides in both using FCA for exploratory data analysis, as well as with the application of FCA for the detection of domestic violence. [ABSTRACT FROM AUTHOR]

Abstract: Electronic commerce (EC) has become a trend in the world nowadays. However, most researches neglect a fundamental issue – the user’s product-specific knowledge on which the useful intelligent systems are based. This research employs the user’s product-specific knowledge and mine his/her interior desire on appropriate target products as a part of personalization process to construct the overall EC strategy for businesses. This paper illustrates a novel web usage mining approach, based on the sequence mining technique applied to user’s navigation behaviour, to discover patterns in the navigation of websites. Three critical contributions are made in this paper: (1) using the footstep graph to visualize the user’s click-stream data and any interesting pattern can be detected more easily and quickly; (2) illustrating a novel sequence mining approach to identify pre-designated user navigation patterns automatically and integrates back-propagation network (BPN) model smoothly; and (3) applying the empirical research to indicate that the proposed approach can predict and categorize the users’ navigation behaviour with high accuracy. [Copyright &y& Elsevier]

In this paper, the feature selection problem was formulated as a multi-objective optimization problem, and new criteria were proposed to fulfill the goal. Foremost, data were pre-processed with missing value replacement scheme, re-sampling procedure, data type transformation procedure, and min-max normalization procedure. After that a wide variety of classifiers and feature selection methods were conducted and evaluated. Finally, the paper presented comprehensive experiments to show the relative performance of the classification tasks. The experimental results revealed the success of proposed methods in credit approval data. In addition, the numeric results also provide guides in selection of feature selection methods and classifiers in the knowledge discovery process. [Copyright &y& Elsevier]

Abstract: The operating status of an enterprise is disclosed periodically in a financial statement. As a result, investors usually only get information about the financial distress a company may be in after the formal financial statement has been published. If company executives intentionally package financial statements with the purpose of hiding the actual status of the company, then investors will have even less chance of obtaining the real financial information. For example, a company can manipulate its current ratio by up to 200% so that its liquidity deficiency will not show up as a financial distress in the short run. To improve the accuracy of the financial distress prediction model, this paper adopted the operating rules of the Taiwan stock exchange corporation (TSEC) which were violated by those companies that were subsequently stopped and suspended, as the range of the analysis of this research. In addition, this paper also used financial ratios, other non-financial ratios, and factor analysis to extract adaptable variables. Moreover, the artificial neural network (ANN) and data mining (DM) techniques were used to construct the financial distress prediction model. The empirical experiment with a total of 37 ratios and 68 listed companies as the initial samples obtained a satisfactory result, which testifies for the feasibility and validity of our proposed methods for the financial distress prediction of listed companies. This paper makes four critical contributions: (1) The more factor analysis we used, the less accuracy we obtained by the ANN and DM approach. (2) The closer we get to the actual occurrence of financial distress, the higher the accuracy we obtain, with an 82.14% correct percentage for two seasons prior to the occurrence of financial distress. (3) Our empirical results show that factor analysis increases the error of classifying companies that are in a financial crisis as normal companies. (4) By developing a financial distress prediction model, the ANN approach obtains better prediction accuracy than the DM clustering approach. Therefore, this paper proposes that the artificial intelligent (AI) approach could be a more suitable methodology than traditional statistics for predicting the potential financial distress of a company. [Copyright &y& Elsevier]

This paper presents a hybrid approach of case-based reasoning and rule-based reasoning, as an alternative to the purely rule-based method, to build a clinical decision support system for ICU. This enables the system to tackle problems like high complexity, low experienced new staff and changing medical conditions. The purely rule-based method has its limitations since it requires explicit knowledge of the details of each domain of ICU, such as cardiac domain hence takes years to build knowledge base. Case-based reasoning uses knowledge in the form of specific cases to solve a new problem, and the solution is based on the similarities between the new problem and the available cases. This paper presents a case-based reasoning and rule-based reasoning based model which can provide clinical decision support for all domains of ICU unlike rule-based inference models which are highly domain knowledge specific. Experiments with real ICU data as well as simulated data clearly demonstrate the efficacy of the proposed method. [Copyright &y& Elsevier]

Abstract: Clustering is an important function in data mining. Its typical application includes the analysis of consumer’s materials. Adaptive resonance theory network (ART) is very popular in the unsupervised neural network. Type I adaptive resonance theory network (ART1) deals with the binary numerical data, whereas type II adaptive resonance theory network (ART2) deals with the general numerical data. Several information systems collect the mixing type attitudes, which included numeric attributes and categorical attributes. However, ART1 and ART2 do not deal with mixed data. If the categorical data attributes are transferred to the binary data format, the binary data do not reflect the similar degree. It influences the clustering quality. Therefore, this paper proposes a modified adaptive resonance theory network (M-ART) and the conceptual hierarchy tree to solve similar degrees of mixed data. This paper utilizes artificial simulation materials and collects a piece of actual data about the family income to do experiments. The results show that the M-ART algorithm can process the mixed data and has a great effect on clustering. [Copyright &y& Elsevier]

Recently, associative Petri nets (APNs) have been attracted in knowledge representation and modeling of expert systems. However, it is a challenge in transferring from a complex system to an APNs model. Moreover, the domain knowledge and relationship finding between processes has increased in difficulty and complexity in expert and intelligence systems. This paper addresses this issue by proposing a novel systematic procedure to guide the analyst in creating decision support models. The proposed construction algorithm can not only reduce the effort of construction but also can easily implement for knowledge representation. It can give a summary to describe all states but also simplify the reasoning process. An example of malicious email detection is presented to provide empirical evidence in the utility of the systematic procedure of associative Petri net model construction. Experimental result shows that our constructed APN model outperforms than other methods. [ABSTRACT FROM AUTHOR]

Mining high utility episode rules in complex event sequences has emerged as an important topic in data mining because the utility-based episode rules generated may provide important insights that facilitate decision making for expert and intelligent systems. Although one may employ previous methods in this research area to indirectly construct utility-based episode rules, they typically lack efficiency and effectiveness for real-world applications. In this paper, we develop a novel methodology to directly generate high utility episode rules during the mining process, which is the first work addressing the issue of utility-based episode rule mining. Our goal is to simultaneously resolve the difficulty of the previous reported methods for frequent episode mining and utility-based episode mining. An algorithm called UBER-Mine (Utility-Based Episode Rules) and a structure named UR-Tree (Utility Rule Tree) are proposed to mine efficiently the complete set of high utility episode rules in complex event sequences. In short, UBER-Mine is based on an extended downward closure property, which can efficiently discover utility-based episode rules. On the other hand, UR-Tree can maintain important event information without producing candidate episodes to further accelerate the mining process. Results on both real and synthetic datasets show that UBER-Mine with UR-Tree has good scalability on large datasets and runs faster than the basic UBER-Mine and the current best high utility episode mining algorithm over 100 times. Furthermore, by proposing a high-utility episode-rule model called IV-UBER (InVestment by Utility-Based Episode Rules), we further demonstrate the effectiveness of our method for mining high utility-based episode rules on a real-world application for stock investment. The experimental results show that our proposed IV-UBER method outperforms several state-of-the-art algorithms in terms of both precision and annualized return for investment. [ABSTRACT FROM AUTHOR]

Highlights: [•] This paper reviews different methods and techniques to determine which variables could be the most important to financial institutions in order to predict the likely levels of trust among electronic banking users including socio-demographic, economic, financial and behavioural strategic variables that entities have in their database. [•] In order to validate the results, five experts in the subject with experience in different national and international companies were consulted. [•] The best method to perform variable selection using the expert’s criterion is the MGA using Mutual Information computed with the k-NN algorithm. [ABSTRACT FROM AUTHOR]

Abstract: In this paper we introduce an automatic system to perform authority control in digital libraries based on data mining techniques. This system is able to find the different representations for an author name as well as to distinguish between different authors sharing the same name. Using that information, the system shows the user the results of a search over a digital library properly grouped according to their authorship. To accomplish this task, it only uses information that can be directly obtained from the digital library itself without any kind of external data. The system has been tested using different digital libraries on the web. [Copyright &y& Elsevier]

Abstract: The marketspaces of the “New Economy” and the eServices revolution have enabled the formation of new types of partnerships which are electronically mediated. Web-based electronic commerce has also brought a tremendous increase in the volume of data that can be mined for valuable managerial knowledge. The data mining procedures used in this process can be enhanced by employing intelligent agents. This paper describes emerging electronic partnerships between players in developing electronic marketspaces and identifies typical data flows between such players, with an analysis of the potential role of data mining and intelligent agent technology. By identifying the complex nature of information flows between the vast numbers of economic entities, we identify opportunities for applying data mining techniques that can lead to knowledge discovery. In particular, we show how a Generic Agent-based data Mining Architecture (GAMA) can be customized to support managerial decision-making and problem solving in a networked economy. A prototype implementation of GAMA is presented, along with a demonstration of the some of the capabilities of the system. Finally, we explore the role of agents in promoting and maintaining strong automated relationships between various strategic partners. [Copyright &y& Elsevier]

Abstract: The aim of process mining is to discover the process model from the event log which is recorded by the information system. Typical steps of process mining algorithm can be described as: (1) generating event traces from event log, (2) analyzing event traces and obtaining ordering relations of tasks, (3) generating process model with ordering relations of tasks. The first two steps could be very time consuming involving millions of events and thousands of event traces. This paper presents a novel algorithm (λ-algorithm) which almost eliminates these two steps in generating event traces from event log and analyzing event traces so as to reduce the performance of process mining algorithm. Firstly, we retrieve the event multiset (input data of algorithm marked as MS) which records the frequency of each event but ignores their orders when extracted from event logs. The event in event multiset contains the information of post-activities. Secondly, we obtain ordering relations from event multiset. The ordering relations contain causal dependency, potential parallelism and non-potential parallelism. Finally, we discover a process models with ordering relations. The complexity of λ-algorithm is only bound up with the event classes (the set of events in event logs) that has significantly improved the performance of existing process mining algorithms and is expected to be more practical in real-world process mining based on event logs, as well as being able to detect SWF-nets, short-loops and most of implicit dependency (generated by non-free choice constructions). [Copyright &y& Elsevier]

Abstract: The knowledge acquisition (KA) process is not “mining from the expert’s head” and writing rules for building knowledge-based systems (KBS), as it was 20years ago when KA was often confused with knowledge elicitation activity, and modern engineering tools did not exist. The KA process has definitely changed. Today knowledge acquisition is considered a cognitive process that involves both dynamic modeling and knowledge generation activities. KA should be seen as a spiral of epistemological and ontological content that grows upward by transforming tacit knowledge into explicit knowledge, which in turn becomes the basis for a new spiral of knowledge generation. This paper presents some of our attempts to build a new knowledge acquisition methodology that brings together and includes all of these ideas. KAMET II, the evolution of KAMET (), represents a modern approach to creating diagnosis-specialized knowledge models that can be run by Protégé 2000, the open source ontology editor and knowledge-based framework. [Copyright &y& Elsevier]

Abstract: Association-rule mining, which is based on frequency values of items, is the most common topic in data mining. In real-world applications, customers may, however, buy many copies of products and each product may have different factors, such as profits and prices. Only mining frequent itemsets in binary databases is thus not suitable for some applications. Utility mining is thus presented to consider additional measures, such as profits or costs according to user preference. In the past, a two-phase mining algorithm was designed for fast discovering high utility itemsets from databases. When data come intermittently, the approach needs to process all the transactions in a batch way. In this paper, an incremental mining algorithm for efficiently mining high utility itemsets is proposed to handle the above situation. It is based on the concept of the fast-update (FUP) approach, which was originally designed for association mining. The proposed approach first partitions itemsets into four parts according to whether they are high transaction-weighted utilization itemsets in the original database and in the newly inserted transactions. Each part is then executed by its own procedure. Experimental results also show that the proposed algorithm executes faster than the two-phase batch mining algorithm in the intermittent data environment [Copyright &y& Elsevier]

Abstract: Frequent closed itemsets (FCI) play an important role in pruning redundant rules fast. Therefore, a lot of algorithms for mining FCI have been developed. Algorithms based on vertical data formats have some advantages in that they require scan databases once and compute the support of itemsets fast. Recent years, BitTable () and IndexBitTable () approaches have been applied for mining frequent itemsets and results are significant. However, they always use a fixed size of Bit-Vector for each item (equal to number of transactions in a database). It leads to consume more memory for storage Bit-Vectors and the time for computing the intersection among Bit-Vectors. Besides, they only apply for mining frequent itemsets, algorithm for mining FCI based on BitTable is not proposed. This paper introduces a new method for mining FCI from transaction databases. Firstly, Dynamic Bit-Vector (DBV) approach will be presented and algorithms for fast computing the intersection between two DBVs are also proposed. Lookup table is used for fast computing the support (number of bits 1 in a DBV) of itemsets. Next, subsumption concept for memory and computing time saving will be discussed. Finally, an algorithm based on DBV and subsumption concept for mining frequent closed itemsets fast is proposed. We compare our method with CHARM, and recognize that the proposed algorithm is more efficient than CHARM in both the mining time and the memory usage. [Copyright &y& Elsevier]

Abstract: Among a collaborative team, members usually come from diverse disciplines, and their demands for knowledge are also different from each other. Information flow is a type of collaborative process, which exists behind every collaborative team. This paper is concerned with how to obtain team members’ knowledge demands from the information flow. Firstly, the knowledge demands model is defined. Based on the model of knowledge demands and information filtering technologies, some approaches for mining demands from information flow are proposed. This study on the knowledge demand mining can pave the way for developing knowledge recommender systems, which can recommend proper knowledge to proper team members with a collaborative team. [Copyright &y& Elsevier]

Mining erasable itemsets first introduced in 2009 is one of new emerging data mining tasks. In this paper, we present a new data representation called NC_set, which keeps track of the complete information used for mining erasable itemsets. Based on NC_set, we propose a new algorithm called MERIT for mining erasable itemsets efficiently. The efficiency of MERIT is achieved with three techniques as follows. First, the NC_set is a compact structure, which prunes irrelevant data automatically. Second, the computation of the gain of an itemset is transformed into the combination of NC_sets, which can be completed in linear time complexity by an ingenious strategy. Third, MERIT can directly find erasable itemsets without generating candidate itemsets in some cases. For evaluating MERIT, we have conducted extensive experiments on a lot of synthetic product databases. Our performance study shows that the MERIT is efficient and is on average about two orders of magnitude faster than the META, the first algorithm for mining erasable itemsets. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, we elaborate on the systematic design of approaches that combine quantum clustering with intelligent models for knowledge extraction, learning, and representation. Clustering techniques, which acquire certain characteristics of input data, are efficient methods of extracting knowledge from numerical data sets. They can obtain information in the form of cluster centers or relevant structural parameters. The structure and parameters are easily transformed into the initial knowledge of intelligent models. In particular, quantum clustering does not depend on conventional probability approaches but infers the centers of clusters on the basis of the Schrödinger wave equation from quantum mechanics. When used for knowledge extraction, quantum clustering can determine the cluster centers by searching for minima of the potential functions in quantum mechanics. We apply the characteristics of quantum clustering to well-known intelligent models such as the Takagi–Sugeno–Kang (TSK) fuzzy model, the zero-order fuzzy model, and the radial basis function network (RBFN) to facilitate knowledge representation. To show the usefulness of the proposed approaches in knowledge management (or extraction and representation), we use benchmark data sets and compare our results with those of previous work. [Copyright &y& Elsevier]

Abstract: There have been many kinds of association rule mining (ARM) algorithms, e.g., Apriori and FP-tree, to discover meaningful frequent patterns from a large dataset. Particularly, it is more difficult for such ARM algorithms to be applied for temporal databases which are continuously changing over time. Such algorithms are generally based on repeating time-consuming tasks, e.g., scanning databases. To deal with this problem, in this paper, we propose a constraint graph-based method for maintaining frequent patterns (FP) discovered from the temporal databases. Particularly, the constraint graph, which is represented as a set of constraint between two items, can be established by temporal persistency of the patterns. It means that some patterns can be used to build the constraint graph, when the patterns have been shown in a set of the FP. Two types of constraints can be generated by users and adaptation. Based on our scheme, we find that a large number of dataset has been efficiently reduced during mining process and the gathering information while updating. [Copyright &y& Elsevier]

Abstract: Learning predictors for student retention is very difficult. After reviewing the literature, it is evident that there is considerable room for improvement in the current state of the art. As shown in this paper, improvements are possible if we (a) explore a wide range of learning methods; (b) take care when selecting attributes; (c) assess the efficacy of the learned theory not just by its median performance, but also by the variance in that performance; (d) study the delta of student factors between those who stay and those who are retained. Using these techniques, for the goal of predicting if students will remain for the first three years of an undergraduate degree, the following factors were found to be informative: family background and family’s social-economic status, high school GPA and test scores. [Copyright &y& Elsevier]

Abstract: Mining interesting and useful frequent patterns from large databases attracts much attention in recent years. Among the mining approaches, finding temporal patterns and regularities is very important due to its practicality. In the past, Hong et al. proposed the up-to-date patterns, which were frequent within their up-to-date lifetime. Formally, an up-to-date pattern is a pair with the itemset and its valid corresponding lifetime in which the user-defined minimum support threshold must be satisfied. They also proposed an Apriori-like approach to find the up-to-date patterns. This paper thus proposes the up-to-date pattern tree (UDP tree) to keep the up-to-date 1-patterns in a tree structure for reducing database scan. It is similar to the FP-tree structure but more complex due to the requirement of up-to-date patterns. The UDP-growth mining approach is also designed to find the up-to-date patterns from the UDP tree. The experimental results show that the proposed approach has a better performance than the level-wise mining algorithm. [Copyright &y& Elsevier]

Abstract: In this paper, we generalize the k-modes clustering algorithm by weighting attribute value in the dissimilarity computation. Such a generalization generates clusters with stronger intra-similarities, leading to better clustering performance. Experimental results on real life data show that the new k-modes algorithm is superior to the standard k-modes algorithm with respect to clustering accuracy. [Copyright &y& Elsevier]

Abstract: With the continuously increasing needs of location information for users around the world, applications of geospatial information have gained a lot of attention in both research and commercial organizations. Extraction of semantics from the image content for geospatial information seeking and knowledge discovery has been thus becoming a critical process. Unfortunately, the available geographic images may be blurred, either too light or too dark. It is therefore often hard to extract geographic features directly from images. In this paper, we describe our developed methods in applying local scale-invariant features and bag-of-keypoints techniques to annotating images, in order to carry out image categorization and geographic knowledge discovery tasks. First, local scale-invariant features are extracted from geographic images as representative geographic features. Subsequently, the bag-of-keypoints methods are used to construct a visual vocabulary and generate feature vectors to support image categorization and annotation. The annotated images are classified by using geographic nouns. The experimental results show that the proposed approach is sensible and can effectively enhance the tasks of geographic knowledge discovery. [Copyright &y& Elsevier]

Abstract: Obtaining new information in a short time is becoming crucial in today’s economy. A lot of information both offline or online is easily acquired, exacerbating the problem of information overload. Novelty mining detects documents/sentences that contain novel or new information and presents those results directly to users (). Many methods and algorithms for novelty mining have previously been studied, but none have compared and discussed the impact of term weighting on the evaluation measures. This paper performed experiments to recommend the best term weighting function for both document and sentence-level novelty mining. [Copyright &y& Elsevier]

Abstract: Many studies have tried to employ data mining methods to discover useful patterns and knowledge from data streams on sensor networks. However, it is difficult to apply such data mining methods to the sensor streams intermixed from heterogeneous sensor networks. In this paper, to improve the performance of conventional data mining methods, we propose an ontology-based data preprocessing scheme, which is composed of two main phases: (i) session identification and (ii) error detection. The ontology can provide and describe semantics of data measured by each sensor. Thus, by comparing the semantics, we can find out not only relationships between sensor streams but also temporal dynamics of a data stream. To evaluate the proposed method, we have collected sensor streams from in our building during 30days. By using two well-known data mining methods (i.e., co-occurrence pattern and sequential pattern), the results from raw sensor streams and ones from sensor streams with preprocessing were compared with respect to two measurements recall and precision. [ABSTRACT FROM AUTHOR]

Abstract: With the emergence of Web 2.0, user-generated metadata or so-called folksonomies facilitates management of digital content in the World Wide Web. The folksonomies have increasingly become a viable alternative for knowledge seekers to classify their knowledge (resources) and navigate the resources. Traditionally, in expert systems, taxonomy has been fulfilling these roles. However, in a taxonomy, the taxonomic structures are defined by the taxonomists or professional experts of the domain which often do not reflect user vocabulary. In addition, the terms of the taxonomy become outdated very fast which in turn compromises the efficiency and effectiveness of knowledge classification and navigation. On the other hand, maintenance of the taxonomy is time consuming and exhausted. This paper presents taxonomy and folksonomy integration algorithm, namely TaxoFolk to integrate the folksonomy into a taxonomy to enhance knowledge classification and navigation. The output is a hybrid taxonomy–folksonomy structure. The TaxoFolk algorithm comprises of data mining techniques, namely formal concept analysis (FCA), ID3 classification and simple matching coefficients (SMC). Three different taxonomy domains and its folksonomy are used to setup the experiments to identify and to evaluate the threshold values for filtering infrequent tags and invalid tags to automate the TaxoFolk algorithm to integrate a filtered folksonomy with a pre-defined taxonomy for enhancing knowledge navigation. The conducted experiments have ascertained the most appropriate thresholds values for effective building of TaxoFolk structure. [ABSTRACT FROM AUTHOR]

Abstract: One of the most challenging problems in data mining is to develop scalable algorithms capable of mining massive data sets. A novel decision forest learning algorithm named FDF is proposed in this paper to represent multi-level semantic knowledge of the relationship between the data and information implicated. FDF provides their users with just a single set of rules by redefining information gain of information theory, then each tree in the decision forest is constructed in the down-top learning framework, and the number of trees and stopping criteria can be set automatically. When no existing tree match test samples, FDF will build new logical rules for this and thus realize scalable construction process. Empirical studies on a set of natural domains show that decision forest has clear advantages with respect to probabilistic performance. [ABSTRACT FROM AUTHOR]

Abstract: Utility mining has recently been an emerging topic in the field of data mining. It finds out high utility itemsets by considering both the profits and quantities of items in transactions. It may have a bias if items are not always on shelf. In this paper, we thus design a new kind of patterns, named high on-shelf utility itemsets, which considers not only individual profit and quantity of each item in a transaction but also common on-shelf time periods of a product combination. We also propose a two-phased mining algorithm to effectively and efficiently discover high on-shelf utility itemsets. In the first phase, the possible candidate on-shelf utility itemsets within each time period are found level by level. In the second phase, the candidate on-shelf utility itemsets are further checked for their actual utility values by an additional database scan. At last, the experimental results on synthetic datasets also show the proposed approach has a good performance. [ABSTRACT FROM AUTHOR]

Abstract: Many algorithms have been proposed to efficiently mine association rules. One of the most important approaches is FP-growth. Without candidate generation, FP-growth proposes an algorithm to compress information needed for mining frequent itemsets in FP-tree and recursively constructs FP-trees to find all frequent itemsets. Performance results have demonstrated that the FP-growth method performs extremely well. In this paper, we propose the IFP-growth (improved FP-growth) algorithm to improve the performance of FP-growth. There are three major features of IFP-growth. First, it employs an address-table structure to lower the complexity of forming the entire FP-tree. Second, it uses a new structure called FP-tree+ to reduce the need for building conditional FP-trees recursively. Third, by using address-table and FP-tree+ the proposed algorithm has less memory requirement and better performance in comparison with FP-tree based algorithms. The experimental results show that the IFP-growth requires relatively little memory space during the mining process. Even when the minimum support is low, the space needed by IFP-growth is about one half of that of FP-growth and about one fourth of that of nonordfp algorithm. As to the execution time, our method outperforms FP-growth by one to 300 times under different minimum supports. The proposed algorithm also outperforms nonordfp algorithm in most cases. As a result, IFP-growth is very suitable for high performance applications. [ABSTRACT FROM AUTHOR]

Abstract: Internet social networks have arisen in the last years as powerful tools where people exchange knowledge and multimedia content. They help to share interests between groups of people with common features. Undoubtedly, there is an inherent social network in any e-learning system, where the main actors are teachers, learners and learning resources. Most e-learning software are mainly focused in content dissemination and group work, but the possibilities that Internet LMSs could offer go further. Recently, there has been research work focused on Web Communities for learning and their formulation as Social Networks. Thus, social network analysis may be applied to infer group structures and to make intelligent recommendation systems and data mining. This paper proposes a method for the formulation and interpretation of learning management platforms as social networks. In order to achieve a major generalization, we develop an ontology to integrate the information from different Learning Management Systems. After that, a personalized social network is extracted from the ontology. This change in the point of view of a LMS could be a challenge to make further studies about learners, teachers and learning resources to obtain a better understanding of their social structure, and therefore to make or improve decisions about the learning process. [ABSTRACT FROM AUTHOR]

Abstract: Sitemaps designed by webmasters are not only presenting the main usage flows for users, but also organizing the hierarchical concept of the website. However, websites seldom provide sitemap pages to facilitate users to browse pages easily. Even provided, these sitemaps are not for machine-understanding, although few websites provide sitemaps with the XML format. In this paper, we develop a system, SiteMap Generator (SMG), to automatically generate the hierarchical sitemap for a website. SMG consists of five components. Sequence Translator translates a page’s HTML source into a long sequence and then Page Partitioner splits the page into blocks based on analyzing the sequence complexity. Block Identifier categorizes each block into one of three block types: content, structure or redundant. Using the popular sequence searching tool, BLAST, Block Cluster calculates similarities between blocks so that blocks with similar functionalities are grouped and considered as candidate blocks for the sitemap. Finally, Hyperlink Analyzer transforms page-to-page links into block-to-block links and applies Kleinberg’s HITS algorithm to estimate authority and hub values of each block. Block entropy value derived from features entropies is also used to improve the HITS. Several experiments on three websites: Mozilla, CNN and Yahoo! News, show that SMG is useful to partition a page into blocks (F1=86%), identify the block type (F1=85%), and generate the sitemap for the website (F1=63%). [ABSTRACT FROM AUTHOR]

Abstract: In semiconductor manufacturing processes, run-to-run (R2R) control is used to improve productivity by adjusting process inputs run by run. A process will be controlled based on information obtained during or after a process, including metrology values of wafers. Those metrology values, however, are usually available for only a small fraction of sampled wafers. In order to overcome the limitation, one can use virtual metrology (VM) to predict metrology values of all wafers, based on sensor data from production equipments and actual metrology values of sampled wafers. In this paper, we develop VM prediction models using various data mining techniques. We also develop a VM embedded R2R control system using the exponentially weighted moving average (EWMA) scheme. The experiments consist of two parts: (1) verifying VM prediction models with actual production equipments data, and (2) conducting simulations of the R2R control system. Our VM prediction models are found to be accurate enough to be directly implemented in actual manufacturing processes. The simulation results show that the VM embedded R2R control system improves productivity. [ABSTRACT FROM AUTHOR]