Search

This paper introduces a novel proposal to discover the best associative classification rules through studying the influence of the attributes used in robust catalogues. Notice that a catalogue is defined as a dataset free of duplicate records. Moreover, a robust catalogue is obtained when incomplete records and those with uncertainty are eliminated from a catalogue. Therefore, a robust catalogue is a collection of association rules with 100% confidence and unitary support. In this paper we demonstrate that robust catalogues contain the same association rules as the datasets from which they were obtained, but can be managed in memory without eliminating any data from the analysis. In fact, the experiments performed show that all robust catalogues contained in a classification dataset are easily obtained, providing millions of associative classification rules with 100% confidence to the expert researcher in data mining. [ABSTRACT FROM AUTHOR]

• A model-free reinforcement learning algorithm for adaptive low-level PID control. • Double Q-learning provides excelent performance with low computational cost. • Incremental state and action spaces discretization for fast reinforcement learning. • The Double Q-PID algorithm is effectively applied to multiple robotic platforms. • The proposed algorithm effectively improves the adaptability of the PID controllers. Many expert systems have been developed for self-adaptive PID controllers of mobile robots. However, the high computational requirements of the expert systems layers, developed for the tuning of the PID controllers, still require previous expert knowledge and high efficiency in algorithmic and software execution for real-time applications. To address these problems, in this paper we propose an expert agent-based system, based on a reinforcement learning agent, for self-adapting multiple low-level PID controllers in mobile robots. For the formulation of the artificial expert agent, we develop an incremental model-free algorithm version of the double Q -Learning algorithm for fast on-line adaptation of multiple low-level PID controllers. Fast learning and high on-line adaptability of the artificial expert agent is achieved by means of a proposed incremental active-learning exploration-exploitation procedure, for a non-uniform state space exploration, along with an experience replay mechanism for multiple value functions updates in the double Q -learning algorithm. A comprehensive comparative simulation study and experiments in a real mobile robot demonstrate the high performance of the proposed algorithm for a real-time simultaneous tuning of multiple adaptive low-level PID controllers of mobile robots in real world conditions. [ABSTRACT FROM AUTHOR]

Highlights • An architecture for numerical design of experiment configuration is proposed. • A bayesian network with a "multi-net" strategy is proposed. • Models are trained from historical data and expert knowledge. • The performances of the proposed method are validated through a case study. Abstract This paper describes a specific knowledge-based system (KBS) to assist designers in configuring numerical design of experiments (NDoE) processes efficiently. NDoE processes are applied in product design to improve the quality of product, by taking into account variabilities and uncertainties. NDoE processes are defined by various and complex methodologies to achieve several objectives, as optimization, surrogate modeling or sensitivity analysis. On the other hand, NDoE processes may demand huge computing resources to execute hundreds simulations, and also advanced expert knowledge to set the best configuration amongst numerous possibilities. Designers aim to obtain most useful results with a minimal computational cost as soon as possible. Thus, the configuration step must be as fast as possible, and it must lead to an efficient combination of complex methods, algorithms and hyper-parameters, to obtain valuable information on the product. The proposed KBS and its inference engine, a bayesian network, is detailed and applied to a product developed by automotive industry. The KBS propose new efficient configurations to achieve designers' goal. This application shorten the configuration step of the NDoE process, and enables designers to use more complex methods. It also allows designers to capitalize knowledge and learn from each past NDoE process. [ABSTRACT FROM AUTHOR]

Voice activity detection (VAD) is an essential task in expert systems that rely on oral interfaces. The VAD module detects the presence of human speech and separates speech segments from silences and non-speech noises. The most popular current on-line VAD systems are based on adaptive parameters which seek to cope with varying channel and noise conditions. The main disadvantages of this approach are the need for some initialisation time to properly adjust the parameters to the incoming signal and uncertain performance in the case of poor estimation of the initial parameters. In this paper we propose a novel on-line VAD based only on previous training which does not introduce any delay. The technique is based on a strategy that we have called Multi-Normalisation Scoring (MNS). It consists of obtaining a vector of multiple observation likelihood scores from normalised mel-cepstral coefficients previously computed from different databases. A classifier is then used to label the incoming observation likelihood vector. Encouraging results have been obtained with a Multi-Layer Perceptron (MLP). This technique can generalise for unseen noise levels and types. A validation experiment with two current standard ITU-T VAD algorithms demonstrates the good performance of the method. Indeed, lower classification error rates are obtained for non-speech frames, while results for speech frames are similar. [ABSTRACT FROM AUTHOR]

The paper describes the development of a system for receiving questions from users and providing answers, which is a part of a larger knowledge-based (KB) system based on Node of Knowledge (NOK) conceptual framework for KB system development. A part of NOK conceptual framework is described, as well as a new method of text-expressed knowledge representation – formalized NOK (FNOK). Furthermore, the structure of system for user interaction and NOK based KB system named Question Answering NOK (QANOK) is presented. The paper answers the question whether it is possible to create a part of QANOK over FNOK formalized knowledge which would accept user questions using natural language and provide answers in natural language. Three algorithms were developed – an algorithm which translates text-expressed knowledge in natural language into FMTEK (Formalized Method for Text Expressed Knowledge) formalized notation, an algorithm which translates a question in natural language into qFMTEK (question FMTEK) formalized knowledge notation, and an algorithm which uses qFMTEK formalized question to find an answer within FMTEK formalized knowledge. Algorithms were tested against a predefined set of sentences and questions in English language. Limitations of developed algorithms are also presented, as well as further development of QANOK system. [ABSTRACT FROM AUTHOR]

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: This paper describes an approach that uses multi-label classification methods for search tagged learning objects (LOs) by Learning Object Metadata (LOM), specifically the model offers a methodology that illustrates the task of multi-label mapping of LOs into types queries through an emergent multi-label space, and that can improve the first choice of learners or teachers. In order to build the model, the paper also proposes and preliminarily investigates the use of multi-label classification algorithm using only the LO features. As many LOs include textual material that can be indexed, and such indexes can also be used to filter the objects by matching them against user-provided keywords, we then did experiments using web classification with text features to compare the accuracy with the results from metadata (LO feature). [Copyright &y& Elsevier]

Abstract: A map building algorithm for mobile robots is introduced in this paper. The perceived environment is represented in a map containing in each cell a probability of presence of an object or part of an object. The environment is represented as a collection of modular occupancy grids which are added to the map as far as the mobile robot finds objects outside the existing grids. In this approach a time-of-flight (ToF) camera is exploited as a range sensor for mapping. Indeed, one of the areas where ToF sensors are adequate is in obstacle avoidance, because the detection region is not only horizontal but also vertical, allowing to detect obstacles with complex shapes. The main steps of the map building algorithm are extensively described in the paper. The results of testing the algorithm are considered in two different indoor environments. [Copyright &y& Elsevier]

Abstract: This paper deals with algorithms for text localization and character segmentation in images for process automation in the steel-making industry. Each character which comprises slab identification numbers may be corrupted severely before it is captured by network cameras. Therefore, proper processing is required to localize the target texts successfully. In this paper, we propose (1) a method to evaluate the closeness of an edge patch to the form of a closed contour, (2) an edge inspection method to determine character colors and estimate font thickness, and (3) three reasonable binarization methods to increase the performance of the algorithm for the detection of the left and right boundaries of the text rectangle. The experimental results show that the proposed algorithms are reliable. [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: In this paper, we set out to compare several techniques that can be used in the analysis of imbalanced credit scoring data sets. In a credit scoring context, imbalanced data sets frequently occur as the number of defaulting loans in a portfolio is usually much lower than the number of observations that do not default. As well as using traditional classification techniques such as logistic regression, neural networks and decision trees, this paper will also explore the suitability of gradient boosting, least square support vector machines and random forests for loan default prediction. Five real-world credit scoring data sets are used to build classifiers and test their performance. In our experiments, we progressively increase class imbalance in each of these data sets by randomly under-sampling the minority class of defaulters, so as to identify to what extent the predictive power of the respective techniques is adversely affected. The performance criterion chosen to measure this effect is the area under the receiver operating characteristic curve (AUC); Friedman’s statistic and Nemenyi post hoc tests are used to test for significance of AUC differences between techniques. The results from this empirical study indicate that the random forest and gradient boosting classifiers perform very well in a credit scoring context and are able to cope comparatively well with pronounced class imbalances in these data sets. We also found that, when faced with a large class imbalance, the C4.5 decision tree algorithm, quadratic discriminant analysis and k-nearest neighbours perform significantly worse than the best performing classifiers. [Copyright &y& Elsevier]

Abstract: In this paper, we extend the classical economic order quantity model to allow for the demand rate influenced by both selling price and displayed stock level. A finite period system is considered under inflation for deteriorating items. In addition, we impose a ceiling on the number of on-display stocks. We also allow for shortages and general partial backlogging. The objective is to maximize the discounted total profit by finding the optimal replenishment number, replenishment schedule and selling price changed periodically. An effective search procedure is provided to find the optimal solution by employing the properties derived in this paper and particle swarm optimization algorithm. Several numerical examples are used to illustrate the features of the proposed model. [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: To know whether a patent is registered or rejected, one should rely on the subjective judgments of patent examiners and patent attorneys. In order to overcome this drawback, we propose an algorithm which is able to automatically examine the registration of a patent depending on objective patent data. In this paper, we create the proposed algorithm as a system composed of three procedures: Weight Value Selection, Rejection Criterion Value Selection, and Prediction. In Weight Value Selection, the core words which are the main content of patent documents are extracted. The algorithm finds the average word appearance rates in the document and compares it to the average number of words from rejected patent documents. The algorithm extracts core words from the patent documents and integrates them into an integration core word database. In Rejection Criterion Value Selection, the algorithm extracts the core words from other patent documents that are not used for generating the integration core word database. It finds a relevant document‘s similarity value using extracted core words and the weights of the integration core word database. After that, the algorithm sets each document’s similarity value and accepts the result about the registration or rejection of each patent document as an input value. The algorithm sets the boundary value of a class by running pattern recognition algorithms such as K-means, Perceptron, and Regularized Discriminant Analysis. In the third procedure, Prediction, the algorithm extracts the core words from patent documents for prediction. The algorithm compares the two values created in the first and second steps, and uses a similarity value to predict acceptance or rejection. The proposed Automated Patent Examining System in this paper derives objective prediction results based on past patent data so that we do not have to rely on the subjective judgments of the existing group of patent examiners and patent attorneys to know if a patent will be registered. [Copyright &y& Elsevier]

Abstract: This paper investigates the effects of preventive-maintenance cost functions on the optimal preventive-maintenance policy for a leased product with Weibull life-time distribution. During the lease period, any product failures are rectified by minimal repairs and may incur a penalty to the lessor, if the time duration for performing a minimal repair exceeds a pre-specified time limit. To reduce repair costs and possible penalty, preventive-maintenance actions are scheduled in the lease contract. The objective of this paper is to derive the optimal preventive-maintenance schedule and maintenance degrees such that the expected total cost is minimized. Some structural properties of the optimal policy are obtained and an efficient algorithm is provided to search for the optimal policy. For the cases where the preventive maintenance cost is a constant or a linearly increasing function, the effects of the preventive-maintenance cost function on the optimal policy are investigated in detail both theoretically and numerically. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, a structural truss mass optimization on size and shape is performed taking into account frequency constraints. It is well-known that structural optimizations on shape and size are highly non-linear dynamic optimization problems since this mass reduction conflicts with the frequency constraints especially when they are lower bounded. Besides, vibration modes may switch easily due to shape modifications. This paper intends to investigate the use of a particle swarm optimization (PSO) algorithm as an optimization engine in this type of problem. This choice is based on reported well-behavior of such algorithm as global optimizer in other areas of knowledge. Another feature of the algorithm is taken into account for this choice, like the fact that it is not gradient based, but just based on simple objective function evaluation. The algorithm is briefly revised highlighting its most important features. It is presented four examples regarding the optimization of trusses on shape and size with frequency constraints. The examples are widely reported and used in the related literature as benchmarks. The results show that the algorithm performed similar to other methods and even better in some cases. [Copyright &y& Elsevier]

Abstract: Novelty detection aims at reducing redundant information from a chronologically ordered list of documents or sentences. Other studies of novelty detection have been conducted on the English language, but few papers have addressed the problem of multilingual novelty detection. Likewise, research in multilingual information retrieval have rarely been applied to novelty detection. This paper attempts to bridge the two disciplines by first describing the preprocessing steps for English, Malay and Chinese, then applying document and sentence-level novelty detection for the three languages on APWSJ and TREC 2004 Novelty Track data. Experiments on sentence-level novelty detection show similar results for all three languages, which indicates that our algorithm is suitable for multilingual novelty detection at the sentence level. However, results for document-level novelty detection show a disparity across the different languages, with English and Malay outperforming Chinese. After applying sentence-level novelty detection to detect novel documents, we observe substantial improvements on all three languages. This demonstrates that segmenting documents into sentences improves document-level novelty detection in multiple languages, and has practical benefits for a real-time multilingual novelty detection system. [Copyright &y& Elsevier]

Abstract: According to the peer-to-peer (P2P) sociality, an expert-driven multi-semantic self-organized P2P semantic community model is built in this paper to resolve the two principle problems of P2P resource management, which are how to possess enormous number of high quality shared resource, how to find the resource quickly and accurately. The main contribution of this paper is: ① the autonomous peer (AP) model and Autonomic Semantic Community (ASC)model are build, ② one local domain-related trust and trusted evaluation is projected that bases on semantic similarity, history evaluation and time effect and it takes the requirement and owned resource two aspects into account to present one AP. ③ an algorithm of fusing APs’local trust degree and self-evaluation is designed.④ a procedure to collaboratively elect expert AP e is proposed. ⑤ Self-organized community formation algorithm SoFA and its communication mechanism are built. In SoFA, APs gather its communication information and according to it to evaluate itself and other APs locally, build and dynamically adjust local topology, elect the domain expert AP e and then an expert-driven ASC is established. Experiment shows our domain trust evaluation and self-evaluation mechanism and SoFA could not only adaptively optimize network to improve the efficiency of service (resource) discovery but also base on the synthesized trust degree to find the free-riders, malice peers, low quality service peers and to keep them within limits. [Copyright &y& Elsevier]

Abstract: Although has been widely used to study geographical problems, case-based reasoning (CBR) method is far less than perfect and research is in great need of to improve CBR-based geographic data representation modeling, as well as spatial similarity computation and reasoning algorithm. This paper reports an improved CBR-based method for studying the spatially complex land use change. Based on a brief summary of advantages and challenges of current existing quantitative methods, the paper first proposes to introduce the CBR approach for land use change study. A three-component model (“problem”, “geographic environment”, and “outcome”) was proposed to represent the land use change cases among which there are complicated and inherent spatial relationships. This paper then presents an algorithm to retrieve the inherent spatial relationships, which are then introduced into the CBR similarity reasoning algorithm to predict land use change. The method was tested by examining the land use change in Pearl River Mouth area in China and yields a similar prediction accuracy of 80% as that derived by applying the Bayesian networks approach to the exact same data. As a result, the CBR-based method proposed in this study provides an effective and explicit solution to represent and solve the complicated geographic problems. [Copyright &y& Elsevier]

Abstract: A belief rule base inference methodology using the evidential reasoning (RIMER) approach has been developed recently. A belief rule base (BRB), which can be treated as a more generalized expert system, extends traditional IF-THEN rules, but requires the assignment of some system parameters including rule weights, attribute weights, and belief degrees. These parameters need to be determined with care for reliable system simulation and prediction. Some off-line optimization models have been proposed, but it is expensive to train and re-train these models in particular for large-scale systems. Moreover, the recursive algorithms are also proposed to fine tune a BRB online, which require less calculation time and satisfy the real-time requirement. However, the earlier mentioned learning algorithms are all based on a predetermined structure of the BRB. For a complex system, prior knowledge may not be perfect, which leads to the construction of an incomplete or even inappropriate initial BRB structure. Also, too many rules in an initial BRB may lead to over fitting, whilst too few rules may result in under fitting. Consequently, such a BRB system may not be capable of achieving overall optimal performance. In this paper, we consider one realistic and important case where both a preliminary BRB structure and system parameters assigned to given rules can be adjusted online. Based on the definition of a new statistical utility for a belief rule as investigated in this paper, a sequential learning algorithm for online constructing more compact BRB systems is proposed. Compared with the other learning algorithms, a belief rule can be automatically added into the BRB or pruned from the BRB, and our algorithm can also satisfy the real-time requirement. In addition, our algorithm inherits the feature of RIMER, i.e., only partial input and output information is required, which could be either incomplete or vague, either numerical or judgmental, or mixed. In order to verify the effectiveness of the proposed algorithm, a practical case study about oil pipeline leak detection is studied and examined to demonstrate how the algorithm can be implemented. [Copyright &y& Elsevier]

Abstract: In this paper, a novel fuzzy classifier for multi-classification problems, based on support vector data description (SVDD) and improved PCM, is proposed. The proposed method is the robust version of SVDD by assigning a weight to each data point, which represents fuzzy membership degree of the cluster computed by the improved PCM method. Accordingly, this paper presents the multi-classification algorithm based on the robust weighted SVDD, and gives the simple classification rule. Experimental results show that the proposed method can reduce the effect of outliers and yield higher classification rate. [Copyright &y& Elsevier]

Abstract: Objective: This paper presents an algorithm for the solution of the motif discovery problem (MDP). Methods and materials: Motif discovery problem can be considered in two cases: motifs with insertions/deletions, and motifs without insertions/deletions. The first group motifs can be found by stochastic and approximated methods. The second group can be found by using stochastic and approximated methods, but also deterministic method. We proved that the second group motifs can be found with a deterministic algorithm, and so, it can be said that the second motifs finding is a P-type problem as proved in this paper. Results and conclusions: An algorithm was proposed in this paper for motif discovery problem. The proposed algorithm finds all motifs which are occurred in the sequence at least two times, and it also finds motifs of various sizes. Due to this case, this algorithm is regarded as Automatic Exact Motif Discovery Algorithm. All motifs of different sizes can be found with this algorithm, and this case was proven in this paper. It shown that automatic exact motif discovery is a P-type problem in this paper. The application of the proposed algorithm has been shown that this algorithm is superior to MEME, MEME3, Motif Sampler, WEEDER, CONSENSUS, AlignACE. [Copyright &y& Elsevier]

This paper provides a systematic review of previous software fault prediction studies with a specific focus on metrics, methods, and datasets. The review uses 74 software fault prediction papers in 11 journals and several conference proceedings. According to the review results, the usage percentage of public datasets increased significantly and the usage percentage of machine learning algorithms increased slightly since 2005. In addition, method-level metrics are still the most dominant metrics in fault prediction research area and machine learning algorithms are still the most popular methods for fault prediction. Researchers working on software fault prediction area should continue to use public datasets and machine learning algorithms to build better fault predictors. The usage percentage of class-level is beyond acceptable levels and they should be used much more than they are now in order to predict the faults earlier in design phase of software life cycle. [Copyright &y& Elsevier]

Abstract: The purpose of this paper is to develop a model selection method for time constrained decision problems. Usually there are many models in a decision support system (DSS), thus the model management system (MMS) is an important component of a DSS. In this paper, we study a DSS that offers competitive models with different levels of solution accuracy. The goodness of a model will be judged by its payoff value that involves the model solution accuracy and information costs. A Genetic Algorithm (GA) is used to find the maximum payoff value of a model, and the model with the best payoff value is selected for the target problem. We also propose a model selection system that is easy to implement with the help of web services technology. [Copyright &y& Elsevier]

The growing technology industry has led to the steady enhancement of expert systems, often at the cost of increased complexity for the systems’ end users. Efforts to improve the prescriptive elements of systems, however, often prove unsuccessful, since the nature of complex and high-dimensional decision problems is difficult to capture precisely by models and algorithms. To rectify this deficiency, complementary softwares may be used to accept decision-making input from users. In this paper, we introduce a graphical interface-based multi-criteria decision support system for designing radiation therapy treatment plans. While many automated strategies for treatment plan generation exist in the literature, they often require a large amount of iteration and a priori decision-making in practice, so much of the planning is done manually. Our interface, morDiRECT (the Medical Operations Research Laboratory’s Display for Ranking and Evaluating Customized Treatments) uses the variability associated with the planning parameters to generate diverse plan sets automatically, creating a comprehensive and visible decision space for users. We demonstrate morDiRECT’s generation process, built-in analytical tooling and graphical display using four clinical case studies. In three cases, we find plans that fully dominate the benchmark forward plans, as well as additional plans that possess potentially desirable tradeoffs for all cases. Our results demonstrate that with relatively little upfront effort, users can pre-generate and choose from a diverse set of clinically acceptable plans, leading to reliable treatments for head-and-neck patients. [ABSTRACT FROM AUTHOR]

• A novel weighted ensemble algorithm is proposed for indoor localization. • Obtained around 95% accuracy even when train and test conditions are different. • Incorporating mean and variance of RSSIs improve accuracy of the ensemble to 98%. • The algorithm works for varying granularity: room level grid and 1m × 1m grid. Indoor localization systems have the capability to change the way of providing location-based services in a closed environment. Though there is no agreed-upon technology that works best in indoor, WiFi signal is an important alternative as most of such places are covered by WiFi Access Points (APs). In this paper, the problem of indoor localization is investigated from the perspective of expert systems through applying machine learning techniques. The significant variation of WiFi signal strength with ambient conditions as well as device configuration badly affects the localization accuracy. Thus, the fingerprinting effort required to train a localization system subject to context heterogeneity is huge. The uncertainty in localization performance due to varying contexts is hardly investigated in the literature. Consequently, the main contribution of this paper is to propose a weighted ensemble classifier based on Dempster–Shafer belief theory to efficiently handle context heterogeneity. Here, the context is defined in terms of different smartphone configurations used for training and testing the system as well as temporal variation of signals. The method presented here utilizes the Dempster–Shafer theory of belief functions to calculate the weights of the base learners in the decision of the ensemble. Belief theory is applied here to handle the inherent uncertainty in WiFi signal variations due to heterogeneous context. Real life experiments are conducted for two datasets, JUIndoorLoc and UJIIndoorLoc at different granularity levels. For JUIndoorLoc, with state-of-the-art classifiers, 86–97% accuracy can be achieved for 10-fold cross-validation. However, when the training context differs from the test conditions, accuracy drops to 62–87%. In such a scenario, the proposed weighted ensemble technique is found to achieve almost 98% localization accuracy when RSSIs, mean and variance of RSSIs are considered as features. The technique can lead to an effective expert system for indoor localization at varying granularity levels. Such systems would be beneficial for pervasive indoor positioning applications as no dedicated infrastructure is needed for positioning. [ABSTRACT FROM AUTHOR]

Abstract: Some of the real-world problems are represented with just one label but many of today’s issues are currently being defined with multiple labels. This second group is important because multi-label classes provide a more global picture of the problem. From the study of the characteristics of the most influential systems in this area, MlKnn and RAkEL, we can observe that the main drawback of these specific systems is the time required. Therefore, the aim of the current paper is to develop a more efficient system in terms of computation without incurring accuracy loss. To meet this objective we propose MlCBR, a system for multi-label classification based on Case-Based Reasoning. The results obtained highlight the strong performance of our algorithm in comparison with previous benchmark methods in terms of accuracy rates and computational time reduction. [Copyright &y& Elsevier]

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: Datasets with missing values are frequent in real-world classification problems. It seems obvious that imputation of missing values can be considered as a series of secondary tasks, while classification is the main purpose of any machine dealing with these datasets. Consequently, Multi-Task Learning (MTL) schemes offer an interesting alternative approach to solve missing data problems. In this paper, we propose an MTL-based method for training and operating a modified Multi-Layer Perceptron (MLP) architecture to work in incomplete data contexts. The proposed approach achieves a balance between both classification and imputation by exploiting the advantages of MTL. Extensive experimental comparisons with well-known imputation algorithms show that this approach provides excellent results. The method is never worse than the traditional algorithms – an important robustness property – and, also, it clearly outperforms them in several problems. [Copyright &y& Elsevier]

Abstract: Many techniques have been proposed for credit risk prediction, from statistical models to artificial intelligence methods. However, very few research efforts have been devoted to deal with the presence of noise and outliers in the training set, which may strongly affect the performance of the prediction model. Accordingly, the aim of the present paper is to systematically investigate whether the application of filtering algorithms leads to an increase in accuracy of instance-based classifiers in the context of credit risk assessment. The experimental results with 20 different algorithms and 8 credit databases show that the filtered sets perform significantly better than the non-preprocessed training sets when using the nearest neighbour decision rule. The experiments also allow to identify which techniques are most robust and accurate when confronted with noisy credit data. [Copyright &y& Elsevier]

Abstract: Market segmentation is a core marketing concept that is conceptually simple to define and understand, but inherently a multi-criteria problem that is hard to measure and computationally difficult in many aspects. This paper reviews the development of market segmentation techniques and identifies the computational issues of the applications of market segmentation. A multidimensional unified framework for market segmentation is proposed based on the relationship among segmentation variables, data measures, and the multi-objective optimization techniques implemented. We conduct an empirical comparison of two prominent methods: a concomitant finite mixture model and a multi-objective evolutionary algorithm. The result shows that the proposed framework helps to understand different segmentation models and solutions and to guide the development of new market segmentation solution techniques. [Copyright &y& Elsevier]

Abstract: An adaptive clustering procedure specifically designed for process monitoring, fault detection and isolation is presented in this paper. The key feature of the proposed procedure can be identified as its underlying capability to detect novelties in the system’s mode of operation and, thus, to identify previously unseen functioning modes of the process. Once a novelty is detected, relevant informations are used to enrich the knowledge-base of the algorithm and as a result the proposed clustering procedure evolves and learns the new features of the monitored process in accordance with the available process data. The suggested clustering procedure is theoretically illustrated and its effectiveness has been investigated experimentally. Particularly, the on-line implementation of the algorithm and its integration with a fault detection expert system have been considered by making reference to a pneumatic process. [Copyright &y& Elsevier]

Abstract: Transaction data record various information about individuals, including their purchases and diagnoses, and are increasingly published to support large-scale and low-cost studies in domains such as marketing and medicine. However, the dissemination of transaction data may lead to privacy breaches, as it allows an attacker to link an individual’s record to their identity. Approaches that anonymize data by eliminating certain values in an individual’s record or by replacing them with more general values have been proposed recently, but they often produce data of limited usefulness. This is because these approaches adopt value transformation strategies that do not guarantee data utility in intended applications and objective measures that may lead to excessive data distortion. In this paper, we propose a novel approach for anonymizing data in a way that satisfies data publishers’ utility requirements and incurs low information loss. To achieve this, we introduce an accurate information loss measure and an effective anonymization algorithm that explores a large part of the problem space. An extensive experimental study, using click-stream and medical data, demonstrates that our approach permits many times more accurate query answering than the state-of-the-art methods, while it is comparable to them in terms of efficiency. [Copyright &y& Elsevier]

Abstract: Cascade of unsupervised and supervised learning algorithms are suitable in all those problems where there are large unlabelled input datasets and the underlying data structure is hidden and not clearly defined. In petroleum geology the understanding and characterization of reservoirs needs integration of different subsurface data in order to create reliable reservoir models. The large amount of data for each well and the presence of different wells to be simultaneously analysed make this task both complex and time consuming. In this scenario, the development of reliable characterization methods is of prime importance in order to help the geologist and reduce the subjectivity of data interpretation. In this paper, we propose a novel interpretation system based on the use of unsupervised and supervised learning techniques in cascade. Using unsupervised algorithm the domain expert identifies relevant clusters that will be used as classes in the following step, in order to learn a classifier to be applied to new instances and wells. We test the approach over five real well dataset using different evaluating techniques. Main advantages of this approach are the ability to manage and use a large amount of data simultaneously and the reduction in interpretation time of a group of wells. [Copyright &y& Elsevier]

Abstract: Steganography is the process of hiding information on a host signal. Transparency is referred to the ability to avoid suspicion about the existence of a secret message. The most popular mechanisms for hiding data in audio signals are the Least Significant Bit (LSB) substitution, Frequency Masking (FM), Spread Spectrum (SS), and Shift Spectrum Algorithm (SSA). In this paper, we adapt the Frequency Masking concept using an efficient sorting of the wavelet coefficients of the secret messages and use an indirect LSB substitution for hiding speech signals into speech signals. The experimental results show that the proposed model, the Efficient Wavelet Masking (EWM) scheme, has a hiding capacity significantly higher than the Spread and Shift Spectrum Algorithms and additionally a statistical transparency higher than all of the above mentioned mechanisms. Moreover, the transparency is not dependent of the host signal chosen because the wavelet sorting guarantees the adaptation of the secret message to the host signal. [Copyright &y& Elsevier]

Abstract: The aim of forming collaborative learning teams is that participating students acquire new knowledge and skills through the interaction with their peers. To reach this aim, teachers usually utilize a grouping criterion based on the students’ roles and on forming well-balanced teams according to the roles of their members. However, the implementation of this criterion requires a considerable amount of time, effort and knowledge on the part of the teachers. In this paper, we propose a deterministic crowding evolutionary algorithm with the aim of assisting teachers when forming well-balanced collaborative learning teams. Considering a given number of students who must be divided into a given number of teams, the algorithm both designs different alternatives to divide students into teams and evaluates each alternative as regards the grouping criterion previously mentioned. This evaluation is carried out on the basis of knowledge of the students’ roles. To analyze the performance of the proposed algorithm, we present the computational experiments developed on ten data sets with different levels of complexity. The obtained results are really promising since the algorithm has reached optimal solutions for the first four data sets and near-optimal solutions for the remaining six data sets. [Copyright &y& Elsevier]

Abstract: Web prefetching techniques are an attractive solution to reduce the user-perceived latency. These techniques are driven by a prediction engine or algorithm that guesses following actions of web users. A large amount of prediction algorithms has been proposed since the first prefetching approach was published, although it is only over the last two or three years when they have begun to be successfully implemented in commercial products. These algorithms can be implemented in any element of the web architecture and can use a wide variety of information as input. This affects their structure, data system, computational resources and accuracy. The knowledge of the input information and the understanding of how it can be handled to make predictions can help to improve the design of current prediction engines, and consequently prefetching techniques. This paper analyzes fifty of the most relevant algorithms proposed along 15years of prefetching research and proposes a taxonomy where the algorithms are classified according to the input data they use. For each group, the main advantages and shortcomings are highlighted. [Copyright &y& Elsevier]

Abstract: The leading partitional clustering technique, k-modes, is one of the most computationally efficient clustering methods for categorical data. However, the performance of the k-modes clustering algorithm which converges to numerous local minima strongly depends on initial cluster centers. Currently, most methods of initialization cluster centers are mainly for numerical data. Due to lack of geometry for the categorical data, these methods used in cluster centers initialization for numerical data are not applicable to categorical data. This paper proposes a novel initialization method for categorical data which is implemented to the k-modes algorithm. The method integrates the distance and the density together to select initial cluster centers and overcomes shortcomings of the existing initialization methods for categorical data. Experimental results illustrate the proposed initialization method is effective and can be applied to large data sets for its linear time complexity with respect to the number of data objects. [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: The paper studies computation models for tasks performed by autonomous mobile robots. Such tasks can be accomplished by reactive control algorithms. Reactive control systems can be described using different models of computation which have as distinguishing feature the abstraction level of time. Thus, three computation models are defined: the untimed model, the synchronous model and the timed model. It is shown that the clocked-synchronous model of computation is more appropriate for describing the controller for a parallel parking task. [Copyright &y& Elsevier]

Abstract: To accurately extrapolate the breast region from a mammogram is a crucial stage of breast mass analysis. It significantly influences the overall analysis accuracy and processing speed of the whole breast mass analysis. In this paper, a novel edge map adjusting gradient vector flow snake (EMA GVF snake) algorithm for extrapolation of breast region from mammograms is proposed. In the proposed algorithm, the median filter is used to filter out the noise in a mammogram, the scale down stage is used to resize down the mammogram size (hence speeding up the extrapolation). The binarization processing stage and the morphological erosion processing stage are used to find a rough breast border. Then a novel gradient adjusting stage is applied to get a modified edge map and the gradient vector flow snake (GVF snake) is used to get the accurate breast border from the rough breast border. The proposed algorithm is tested on 322 digital mammograms from the Mammogram Image Analysis Society database. The mean error function, misclassification error function and the relative foreground area error function are conducted to evaluate the results of the detected breast border and the extracted breast region. Experimental results show that the breast border extrapolated by the proposed algorithm approximately follows the breast border extrapolated by an expert radiologist. Experimental results also show that the proposed algorithm is more robust and precise than the traditional GVF snake scheme for the breast extrapolation on mammograms. [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: Text retrieval has received a lot of attention in computer science. In the text retrieval field, the most widely-adopted similarity technique is using vector space models (VSM) to evaluate the weight of terms and using Cosine, Jaccard or Dice to measure the similarity between the query and the texts. However, these similarity techniques do not consider the effect of the sequence of the information. In this paper, we propose an integrated text retrieval (ITR) mechanism that takes the advantage of both VSM and longest common subsequence (LCS) algorithm. The key idea of the ITR mechanism is to use LCS to re-evaluate the weight of terms, so that the sequence and weight relationships between the query and the texts can be considered simultaneously. The results of mathematical analysis show that the ITR mechanism can increase the similarity on Jaccard and Dice similarity measurements when a sequential relationship exists between the query and the texts. [Copyright &y& Elsevier]

Abstract: Motion planning for multiple mobile robots must ensure the optimality of the path of each and every robot, as well as overall path optimality, which requires cooperation amongst robots. The paper proposes a solution to the problem, considering different source and goal of each robot. Each robot uses a grammar based genetic programming for figuring the optimal path in a maze-like map, while a master evolutionary algorithm caters to the needs of overall path optimality. Co-operation amongst the individual robots’ evolutionary algorithms ensures generation of overall optimal paths. The other feature of the algorithm includes local optimization using memory based lookup where optimal paths between various crosses in map are stored and regularly updated. Feature called wait for robot is used in place of conventionally used priority based techniques. Experiments are carried out with a number of maps, scenarios, and different robotic speeds. Experimental results confirm the usefulness of the algorithm in a variety of scenarios. [Copyright &y& Elsevier]

Abstract: In this paper, an evolutionary algorithm of embedding evolutionary strategy (ES) in ordinal optimization (OO), abbreviated as ESOO, is proposed to solve for a good enough schedule of stochastic job shop scheduling problem (SJSSP) with the objective of minimizing the expected sum of storage expenses and tardiness penalties using limited computation time. First, a rough model using stochastic simulation with short simulation length will be used as a fitness approximation in ES to select N roughly good schedules from search space. Next, starting from the selected N roughly good schedules we proceed with goal softening procedure to search for a good enough schedule. Finally, the proposed ESOO algorithm is applied to a SJSSP comprising 8 jobs on 8 machines with random processing time in truncated normal, uniform, and exponential distributions. The simulation test results obtained by the proposed approach were compared with five typical dispatching rules, and the results demonstrated that the obtaining good enough schedule is successful in the aspects of solution quality and computational efficiency. [Copyright &y& Elsevier]

Abstract: In this paper, we propose an expert-system based parallel multi-1-dimensional block matching algorithm (ESPM-1D-BMA) for motion estimation (ME). Instead of the conventional 2D block matching, we employ the parallel multi-1D blocks matching to improve the computing speed. To improve the ME accuracy, we design a knowledge base and inference engine to determine the true motion vector (MV) from the results of parallel multi-1D blocks matching. To speed up the computing speed further, we present a hardware architecture for implementing the ESPM-1D-BMA. We have demonstrated that the MV estimation accuracy achieved by the proposed ESPM-1D-BMA is much better than the comparing fast block matching algorithms and is close to the 2 dimensional full search block matching algorithm (2D-FSBMA). We also demonstrate that the computing speed of the proposed ESPM-1D-BMA is about two times as fast as the mixed-signal 2D-FSBMA (MS-2D-FSBMA). [Copyright &y& Elsevier]

Abstract: Automatic container-code recognition is of great importance to the modern container management system. Similar techniques have been proposed for vehicle license plate recognition in past decades. Compared with license plate recognition, automatic container-code recognition faces more challenges due to the severity of nonuniform illumination and invalidation of color information. In this paper, a computer vision based container-code recognition technique is proposed. The system consists of three function modules, namely location, isolation, and character recognition. In location module, we propose a text-line region location algorithm, which takes into account the characteristics of single character as well as the spatial relationship between successive characters. This module locates the text-line regions by using a horizontal high-pass filter and scanline analysis. To resolve nonuniform illumination, a two-step procedure is applied to segment container-code characters, and a projection process is adopted to isolate characters in the isolation module. In character recognition module, the character recognition is achieved by classifying the extracted features, which represent the character image, with trained support vector machines (SVMs). The experimental results demonstrate the efficiency and effectiveness of the proposed technique for practical usage. [Copyright &y& Elsevier]

Abstract: The proper generation of fuzzy membership function is of fundamental importance in fuzzy applications. The effectiveness of the membership functions in pattern classifications can be objectively measured in terms of interpretability and classification accuracy in the conformity of the decision boundaries to the inherent probabilistic decision boundaries of the training data. This paper presents the Supervised Pseudo Self-Evolving Cerebellar (SPSEC) algorithm that is bio-inspired from the two-stage development process of the human nervous system whereby the basic architecture are first laid out without any activity-dependent processes and then refined in activity-dependent ways. SPSEC first constructs a cerebellar-like structure in which neurons with high trophic factors evolves to form membership functions that relate intimately to the probability distributions of the data and concomitantly reconcile with defined semantic properties of linguistic variables. The experimental result of using SPSEC to generate fuzzy membership functions is reported and compared with a selection of algorithms using a publicly available UCI Sonar dataset to illustrate its effectiveness. [Copyright &y& Elsevier]