Showing papers in "Knowledge Based Systems in 2016"
TL;DR: The SCA algorithm obtains a smooth shape for the airfoil with a very low drag, which demonstrates that this algorithm can highly be effective in solving real problems with constrained and unknown search spaces.
Abstract: This paper proposes a novel population-based optimization algorithm called Sine Cosine Algorithm (SCA) for solving optimization problems. The SCA creates multiple initial random candidate solutions and requires them to fluctuate outwards or towards the best solution using a mathematical model based on sine and cosine functions. Several random and adaptive variables also are integrated to this algorithm to emphasize exploration and exploitation of the search space in different milestones of optimization. The performance of SCA is benchmarked in three test phases. Firstly, a set of well-known test cases including unimodal, multi-modal, and composite functions are employed to test exploration, exploitation, local optima avoidance, and convergence of SCA. Secondly, several performance metrics (search history, trajectory, average fitness of solutions, and the best solution during optimization) are used to qualitatively observe and confirm the performance of SCA on shifted two-dimensional test functions. Finally, the cross-section of an aircraft's wing is optimized by SCA as a real challenging case study to verify and demonstrate the performance of this algorithm in practice. The results of test functions and performance metrics prove that the algorithm proposed is able to explore different regions of a search space, avoid local optima, converge towards the global optimum, and exploit promising regions of a search space during optimization effectively. The SCA algorithm obtains a smooth shape for the airfoil with a very low drag, which demonstrates that this algorithm can highly be effective in solving real problems with constrained and unknown search spaces. Note that the source codes of the SCA algorithm are publicly available at http://www.alimirjalili.com/SCA.html .
3,088 citations
TL;DR: This paper used a 7-layer deep convolutional neural network to tag each word in opinionated sentences as either aspect or non-aspect word, and developed a set of linguistic patterns for the same purpose and combined them with the neural network.
Abstract: In this paper, we present the first deep learning approach to aspect extraction in opinion mining. Aspect extraction is a subtask of sentiment analysis that consists in identifying opinion targets in opinionated text, i.e., in detecting the specific aspects of a product or service the opinion holder is either praising or complaining about. We used a 7-layer deep convolutional neural network to tag each word in opinionated sentences as either aspect or non-aspect word. We also developed a set of linguistic patterns for the same purpose and combined them with the neural network. The resulting ensemble classifier, coupled with a word-embedding model for sentiment analysis, allowed our approach to obtain significantly better accuracy than state-of-the-art methods.
716 citations
TL;DR: The empirical results demonstrate that the proposed FOA-SVM method can obtain much more appropriate model parameters as well as significantly reduce the computational time, which generates a high classification accuracy.
Abstract: In this paper, a new support vector machines (SVM) parameter tuning scheme that uses the fruit fly optimization algorithm (FOA) is proposed. Termed as FOA-SVM, the scheme is successfully applied to medical diagnosis. In the proposed FOA-SVM, the FOA technique effectively and efficiently addresses the parameter set in SVM. Additionally, the effectiveness and efficiency of FOA-SVM is rigorously evaluated against four well-known medical datasets, including the Wisconsin breast cancer dataset, the Pima Indians diabetes dataset, the Parkinson dataset, and the thyroid disease dataset, in terms of classification accuracy, sensitivity, specificity, AUC (the area under the receiver operating characteristic (ROC) curve) criterion, and processing time. Four competitive counterparts are employed for comparison purposes, including the particle swarm optimization algorithm-based SVM (PSO-SVM), genetic algorithm-based SVM (GA-SVM), bacterial forging optimization-based SVM (BFO-SVM), and grid search technique-based SVM (Grid-SVM). The empirical results demonstrate that the proposed FOA-SVM method can obtain much more appropriate model parameters as well as significantly reduce the computational time, which generates a high classification accuracy. Promisingly, the proposed method can be regarded as a useful clinical tool for medical decision making.
456 citations
TL;DR: This work proposes a density peaks clustering based on k nearest neighbors (DPC-KNN) which introduces the idea of k nearestNeighborhood (Knn) into DPC and has another option for the local density computation and introduces principal component analysis (PCA) into the model of DPC.
Abstract: Density peaks clustering (DPC) algorithm published in the US journal Science in 2014 is a novel clustering algorithm based on density. It needs neither iterative process nor more parameters. However, original algorithm only has taken into account the global structure of data, which leads to missing many clusters. In addition, DPC does not perform well when data sets have relatively high dimension. Especially, DPC generates wrong number of clusters of real-world data sets. In order to overcome the first problem, we propose a density peaks clustering based on k nearest neighbors (DPC-KNN) which introduces the idea of k nearest neighbors (KNN) into DPC and has another option for the local density computation. In order to overcome the second problem, we introduce principal component analysis (PCA) into the model of DPC-KNN and further bring forward a method based on PCA (DPC-KNN-PCA), which preprocesses high-dimensional data. By experiments on synthetic data sets, we demonstrate the feasibility of our algorithms. By experiments on real-world data sets, we compared this algorithm with k-means algorithm and spectral clustering (SC) algorithm in accuracy. Experimental results show that our algorithms are feasible and effective.
321 citations
TL;DR: A novel technique for predicting the tastes of users in recommender systems based on collaborative filtering is presented, based on factorizing the rating matrix into two non negative matrices whose components lie within the range 0, 1 with an understandable probabilistic meaning.
Abstract: In this paper we present a novel technique for predicting the tastes of users in recommender systems based on collaborative filtering. Our technique is based on factorizing the rating matrix into two non negative matrices whose components lie within the range 0, 1 with an understandable probabilistic meaning. Thanks to this decomposition we can accurately predict the ratings of users, find out some groups of users with the same tastes, as well as justify and understand the recommendations our technique provides.
253 citations
TL;DR: A sequential three-way decision method for cost-sensitive face recognition and a series of image granulation methods based on two-dimensional subspace projection methods, which simulate a sequential decision strategy from rough granule to precise granule.
Abstract: Many previous studies on face recognition attempted to seek a precise classifier to achieve a low misclassification error, which is based on an assumption that all misclassification costs are the same. In many real-world scenarios, however, this assumption is not reasonable due to the imbalanced misclassification cost and insufficient high-quality facial image information. To address this issue, we propose a sequential three-way decision method for cost-sensitive face recognition. The proposed method is based on a formal description of granular computing. It develops a sequential strategy in a decision process. In each decision step, it seeks a decision which minimizes the misclassification cost rather than misclassification error, and it incorporates the boundary decision into the decision set such that a delayed decision can be made if available high-quality facial image information is insufficient for a precise decision. To describe the granular information of the facial image in three-way decision steps, we develop a series of image granulation methods based on two-dimensional subspace projection methods including 2DPCA, 2DLDA and 2DLPP. The sequential three-way decisions and granulation methods present an applicable simulation on human decisions in face recognition, which simulate a sequential decision strategy from rough granule to precise granule. The experiments were conducted on two popular facial image database, which validated the effectiveness of the proposed methods.
243 citations
TL;DR: A real world problem presented in the ECDBL’2014 Big Data competition is used to provide a thorough analysis on the application of some preprocessing techniques, their combination and their performance.
Abstract: Data preprocessing is a major and essential stage whose main goal is to obtain final data sets that can be considered correct and useful for further data mining algorithms. This paper summarizes the most influential data preprocessing algorithms according to their usage, popularity and extensions proposed in the specialized literature. For each algorithm, we provide a description, a discussion on its impact, and a review of current and further research on it. These most influential algorithms cover missing values imputation, noise filtering, dimensionality reduction (including feature selection and space transformations), instance reduction (including selection and generation), discretization and treatment of data for imbalanced preprocessing. They constitute all among the most important topics in data preprocessing research and development. This paper emphasizes on the most well-known preprocessing methods and their practical study, selected after a recent, generic book on data preprocessing that does not deepen on them. This manuscript also presents an illustrative study in two sections with different data sets that provide useful tips for the use of preprocessing algorithms. In the first place, we graphically present the effects on two benchmark data sets for the preprocessing methods. The reader may find useful insights on the different characteristics and outcomes generated by them. Secondly, we use a real world problem presented in the ECDBL’2014 Big Data competition to provide a thorough analysis on the application of some preprocessing techniques, their combination and their performance. As a result, five different cases are analyzed, providing tips that may be useful for readers.
229 citations
TL;DR: A new relation is defined to describe the similarity degree of incomplete information and interval number is utilized to acquire the loss function in a novel three-way decision model based on incomplete information system.
Abstract: We induce the incomplete information to DTRS and build a three-way decision model.We use a hybrid information table to deal with the integrated information system.We list the key steps and algorithm of the proposed three-way decision model. As a natural extension of three-way decisions with incomplete information, this paper provides a novel three-way decision model based on incomplete information system. First, we define a new relation to describe the similarity degree of incomplete information. Then, in view of the missing values presented in incomplete information system, we utilize interval number to acquire the loss function. A hybrid information table which consist both of the incomplete information and loss function, is used to deal with the new three-way decision model. The key steps and algorithm for constructing the integrated three-way decision model are also carefully investigated. An empirical study of medical diagnosis validates the reasonability and effectiveness of our proposed model.
225 citations
TL;DR: The most important approaches to serendipity in recommender systems are summarized, different definitions and formalizations of the concept are compared, state-of-the-art serendIPity-oriented recommendation algorithms and evaluation strategies to assess the algorithms are discussed, and future research directions are provided based on the reviewed literature.
Abstract: We summarize most efforts on serendipity in recommender systems.We compare definitions of serendipity in recommender systems.We classify the state-of-the-art serendipity-oriented recommendation algorithms.We review methods to assess serendipity in recommender systems.We provide the future directions of serendipity in recommender systems. Recommender systems use past behaviors of users to suggest items. Most tend to offer items similar to the items that a target user has indicated as interesting. As a result, users become bored with obvious suggestions that they might have already discovered. To improve user satisfaction, recommender systems should offer serendipitous suggestions: items not only relevant and novel to the target user, but also significantly different from the items that the user has rated. However, the concept of serendipity is very subjective and serendipitous encounters are very rare in real-world scenarios, which makes serendipitous recommendations extremely difficult to study. To date, various definitions and evaluation metrics to measure serendipity have been proposed, and there is no wide consensus on which definition and evaluation metric to use. In this paper, we summarize most important approaches to serendipity in recommender systems, compare different definitions and formalizations of the concept, discuss serendipity-oriented recommendation algorithms and evaluation strategies to assess the algorithms, and provide future research directions based on the reviewed literature.
210 citations
TL;DR: The experimental results show that the proposed tree-based incremental overlapping clustering method can identifies clusters of arbitrary shapes and does not sacrifice the computing time, and more results of comparison experiments show thatThe performance of proposed method is better than the compared algorithms in most of cases.
Abstract: Existing clustering approaches are usually restricted to crisp clustering, where objects just belong to one cluster; meanwhile there are some applications where objects could belong to more than one cluster. In addition, existing clustering approaches usually analyze static datasets in which objects are kept unchanged after being processed; however many practical datasets are dynamically modified which means some previously learned patterns have to be updated accordingly. In this paper, we propose a new tree-based incremental overlapping clustering method using the three-way decision theory. The tree is constructed from representative points introduced by this paper, which can enhance the relevance of the search result. The overlapping cluster is represented by the three-way decision with interval sets, and the three-way decision strategies are designed to updating the clustering when the data increases. Furthermore, the proposed method can determine the number of clusters during the processing. The experimental results show that it can identifies clusters of arbitrary shapes and does not sacrifice the computing time, and more results of comparison experiments show that the performance of proposed method is better than the compared algorithms in most of cases.
188 citations
TL;DR: Experimental results on the well-known MovieLens data set show that the (α*, β*)-pair determined by three-way decision is optimal not only on the training set, but also on the testing set.
Abstract: We propose a framework integrating three-way decision and random forests.We introduce a new recommender action to consult the user for the choice.We build a random forest to predict the probability that a user likes an item.The three-way thresholds are optimal for both the training set and the testing set. Recommender systems attempt to guide users in decisions related to choosing items based on inferences about their personal opinions. Most existing systems implicitly assume the underlying classification is binary, that is, a candidate item is either recommended or not. Here we propose an alternate framework that integrates three-way decision and random forests to build recommender systems. First, we consider both misclassification cost and teacher cost. The former is paid for wrong recommender behaviors, while the latter is paid to actively consult the user for his or her preferences. With these costs, a three-way decision model is built, and rational settings for positive and negative threshold values α* and β* are computed. We next construct a random forest to compute the probability P that a user will like an item. Finally, α * , 0.35 e m 0 e x β * , and P are used to determine the recommender's behavior. The performance of the recommender is evaluated on the basis of an average cost. Experimental results on the well-known MovieLens data set show that the (α*, β*)-pair determined by three-way decision is optimal not only on the training set, but also on the testing set.
TL;DR: A state-of-the-art survey of various applications of Text mining to finance, categorized broadly into FOREX rate prediction, stock market prediction, customer relationship management (CRM) and cyber security.
Abstract: Text mining has found a variety of applications in diverse domains. Of late, prolific work is reported in using text mining techniques to solve problems in financial domain. The objective of this paper is to provide a state-of-the-art survey of various applications of Text mining to finance. These applications are categorized broadly into FOREX rate prediction, stock market prediction, customer relationship management (CRM) and cyber security. Since finance is a service industry, these problems are paramount in operational and customer growth aspects. We reviewed 89 research papers that appeared during the period 2000-2016, highlighted some of the issues, gaps, key challenges in this area and proposed some future research directions. Finally, this review can be extremely useful to budding researchers in this area, as many open problems are highlighted.
TL;DR: The necessary and sufficient conditions used to construct three-way concepts on the basis of classical concepts are proved, and the algorithms building three- way concept lattices on the based of classical concept lattice are presented.
Abstract: The model of three-way concept lattices, a novel model for widely used three-way decisions, is an extension of classical concept lattices in formal concept analysis. This paper systematically analyses the connections between two types of three-way concept lattices (object-induced and attribute-induced three-way concept lattices) and classical concept lattices. The relationships are discussed from the viewpoints of elements, sets and orders, respectively. Furthermore, the necessary and sufficient conditions used to construct three-way concepts on the basis of classical concepts are proved, the algorithms building three-way concept lattices on the basis of classical concept lattices are presented. The obtained results are finally demonstrated and verified by examples.
TL;DR: This paper proposes a novel method of automated detection and localization of MI by using ECG signal analysis that can aid the physicians and clinicians in accurate and faster location of MIs, and thereby providing adequate time available for the requisite treatment decision.
Abstract: Identification and timely interpretation of changes occurring in the 12 electrocardiogram (ECG) leads is crucial to identify the types of myocardial infarction (MI). However, manual annotation of this complex nonlinear ECG signal is not only cumbersome and time consuming but also inaccurate. Hence, there is a need of computer aided techniques to be applied for the ECG signal analysis process. Going further, there is a need for incorporating this computerized software into the ECG equipment, so as to enable automated detection of MIs in clinics. Therefore, this paper proposes a novel method of automated detection and localization of MI by using ECG signal analysis. In our study, a total of 200 twelve lead ECG subjects (52 normal and 148 with MI) involving 611,405 beats (125,652 normal beats and 485,753 beats of MI ECG) are segmented from the 12 lead ECG signals. Firstly, ECG signal obtained from 12 ECG leads are subjected to discrete wavelet transform (DWT) up to four levels of decomposition. Then, 12 nonlinear features namely, approximate entropy ( E a x ), signal energy (?x), fuzzy entropy ( E f x ), Kolmogorov-Sinai entropy ( E k s x ), permutation entropy ( E p x ), Renyi entropy ( E r x ), Shannon entropy ( E s h x ), Tsallis entropy ( E t s x ), wavelet entropy ( E w x ), fractal dimension ( F D x ), Kolmogorov complexity ( C k x ), and largest Lyapunov exponent ( E L L E x ) are extracted from these DWT coefficients. The extracted features are then ranked based on the t value. Then these features are fed into the k-nearest neighbor (KNN) classifier one by one to get the highest classification performance by using minimum number of features. Our proposed method has achieved the highest average accuracy of 98.80%, sensitivity of 99.45% and specificity of 96.27% in classifying normal and MI ECG (two classes), by using 47 features obtained from lead 11 (V5). We have also obtained the highest average accuracy of 98.74%, sensitivity of 99.55% and specificity of 99.16% in differentiating the 10 types of MI and normal ECG beats (11 class), by using 25 features obtained from lead 9 (V3). In addition, our study results achieved an accuracy of 99.97% in locating inferior posterior infarction by using only lead 9 (V3) ECG signal. Our proposed method can be used as an automated diagnostic tool for (i) the detection of different (10 types of) MI by using 12 lead ECG signal, and also (ii) to locate the MI by analyzing only one lead without the need to analyze other leads. Thus, our proposed algorithm and computerized system software (incorporated into the ECG equipment) can aid the physicians and clinicians in accurate and faster location of MIs, and thereby providing adequate time available for the requisite treatment decision.
TL;DR: This paper constructs a novel rough set model for feature subset selection, and defines the dependency between fuzzy decision and condition attributes and employ the dependency to evaluate the significance of a candidate feature, using which a greedyfeature subset selection algorithm is designed.
Abstract: Rough set theory has been extensively discussed in machine learning and pattern recognition. It provides us another important theoretical tool for feature selection. In this paper, we construct a novel rough set model for feature subset selection. First, we define the fuzzy decision of a sample by using the concept of fuzzy neighborhood. A parameterized fuzzy relation is introduced to characterize fuzzy information granules for analysis of real-valued data. Then, we use the relationship between fuzzy neighborhood and fuzzy decision to construct a new rough set model: fuzzy neighborhood rough set model. Based on this model, the definitions of upper and lower approximation, boundary region and positive region are given, and the effects of parameters on these concepts are discussed. To make the new model tolerate noises in data, we introduce a variable-precision fuzzy neighborhood rough set model. This model can decrease the possibility that a sample is classified into a wrong category. Finally, we define the dependency between fuzzy decision and condition attributes and employ the dependency to evaluate the significance of a candidate feature, using which a greedy feature subset selection algorithm is designed. The proposed algorithm is compared with some classical algorithms. The experiments show that the proposed algorithm gets higher classification performance and the numbers of selected features are relatively small.
TL;DR: This study defines a complex and dynamic MAGDM problem, and proposes its resolution framework, a selection process in the context of heterogeneous attributes is proposed that obtains the ranking of individual alternatives and a collective solution.
Abstract: In classical multiple attribute group decision making (MAGDM), decision makers evaluate predefined alternatives based on predefined attributes. In other words, the set of alternatives and the set of attributes are fixed throughout the decision process. However, real-world MAGDM problems (e.g., the decision processes of the United Nations Security Council) frequently have the following features. (1) Decision makers have different interests, and they thus use individual sets of attributes to evaluate the individual alternatives. In some situations, the individual sets of attributes may be heterogeneous. (2) In the decision process, decision makers do not have to reach a consensus regarding the use of the set of attributes. Instead, decision makers hope to find an alternative that is approved by all or most of them. (3) Finally, both the individual sets of attributes and the individual sets of alternatives can change dynamically in the decision process. By incorporating the above practical features into MAGDM, this study defines a complex and dynamic MAGDM problem, and proposes its resolution framework. In the resolution framework, a selection process in the context of heterogeneous attributes is proposed that obtains the ranking of individual alternatives and a collective solution. In addition, a consensus process is developed that generates adjustment suggestions for individual sets of attributes, individual sets of alternatives and individual preferences, thus helping decision makers reach consensus. Compared with existing MAGDM models, this study provides a flexible framework to form an approximate decision model to real-world MAGDM problems.
TL;DR: This new method uses natural language processing essential techniques, a sentiment lexicon enhanced with the assistance of SentiWordNet, and fuzzy sets to estimate the semantic orientation polarity and its intensity for sentences, which provides a foundation for computing with sentiments.
Abstract: The objective of this article is to present a hybrid approach to the Sentiment Analysis problem at the sentence level. This new method uses natural language processing (NLP) essential techniques, a sentiment lexicon enhanced with the assistance of SentiWordNet, and fuzzy sets to estimate the semantic orientation polarity and its intensity for sentences, which provides a foundation for computing with sentiments. The proposed hybrid method is applied to three different data-sets and the results achieved are compared to those obtained using Naive Bayes and Maximum Entropy techniques. It is demonstrated that the presented hybrid approach is more accurate and precise than both Naive Bayes and Maximum Entropy techniques, when the latter are utilised in isolation. In addition, it is shown that when applied to datasets containing snippets, the proposed method performs similarly to state of the art techniques.
TL;DR: A new total uncertainty measure in evidence theory is proposed directly in the framework of belief functions theory without changing the theoretical frameworks and has no drawbacks in traditional ones and has desired properties.
Abstract: A new total uncertainty measure in evidence theory is proposed.The new measure is directly defined in the evidential framework.The new measure is not a generalization of those in the probabilistic framework.The belief intervals and distance metric are used for the new measure's design.The new measure has no drawbacks in traditional ones and has desired properties. The theory of belief functions is a very important and effective tool for uncertainty modeling and reasoning, where measures of uncertainty are very crucial for evaluating the degree of uncertainty in a body of evidence. Several uncertainty measures in the theory of belief functions have been proposed. However, existing measures are generalizations of measures in the probabilistic framework. The inconsistency between different frameworks causes limitations to existing measures. To avoid these limitations, in this paper, a new total uncertainty measure is proposed directly in the framework of belief functions theory without changing the theoretical frameworks. The average distance between the belief interval of each singleton and the most uncertain case is used to represent the total uncertainty degree of the given body of evidence. Numerical examples, simulations, applications and related analyses are provided to verify the rationality of our new measure.
TL;DR: This work presents a discriminative path-based method for fact checking in knowledge graphs that incorporates connectivity, type information, and predicate interactions, and finds that the discrim inative predicate path model is easily interpretable and provides sensible reasons for the final determination.
Abstract: Traditional fact checking by experts and analysts cannot keep pace with the volume of newly created information. It is important and necessary, therefore, to enhance our ability to computationally determine whether some statement of fact is true or false. We view this problem as a link-prediction task in a knowledge graph, and present a discriminative path-based method for fact checking in knowledge graphs that incorporates connectivity, type information, and predicate interactions. Given a statement S of the form (subject, predicate, object), for example, (Chicago, capitalOf, Illinois), our approach mines discriminative paths that alternatively define the generalized statement (U.S. city, predicate, U.S. state) and uses the mined rules to evaluate the veracity of statement S . We evaluate our approach by examining thousands of claims related to history, geography, biology, and politics using a public, million node knowledge graph extracted from Wikipedia and PubMedDB. Not only does our approach significantly outperform related models, we also find that the discriminative predicate path model is easily interpretable and provides sensible reasons for the final determination.
TL;DR: The experimental results, analysis and statistical tests demonstrate the ability of the proposed combination method to improve prediction performance against all base classifiers, namely, LR, MARS and seven traditional combination methods, in terms of average accuracy, area under the curve, the H-measure and Brier score.
Abstract: Banks take great care when dealing with customer loans to avoid any improper decisions that can lead to loss of opportunity or financial losses. Regarding this, researchers have developed complex credit scoring models using statistical and artificial intelligence (AI) techniques to help banks and financial institutions to support their financial decisions. Various models, from easy to advanced approaches, have been developed in this domain. However, during the last few years there has been marked attention towards development of ensemble or multiple classifier systems, which have proved their ability to be more accurate than single classifier models. However, among the multiple classifier systems models developed in the literature, there has been little consideration given to: 1) combining classifiers of different algorithms (as most have focused on building classifiers of the same algorithm); or 2) exploring different classifier output combination techniques other than the traditional ones, such as majority voting and weighted average. In this paper, the aim is to present a new combination approach based on classifier consensus to combine multiple classifier systems (MCS) of different classification algorithms. Specifically, six of the main well-known base classifiers in this domain are used, namely, logistic regression (LR), neural networks (NN), support vector machines (SVM), random forests (RF), decision trees (DT) and naive Bayes (NB). Two benchmark classifiers are considered as a reference point for comparison with the proposed method and the other classifiers. These are used in combination with LR, which is still considered the industry-standard model for credit scoring models, and multivariate adaptive regression splines (MARS), a widely adopted technique in credit scoring studies. The experimental results, analysis and statistical tests demonstrate the ability of the proposed combination method to improve prediction performance against all base classifiers, namely, LR, MARS and seven traditional combination methods, in terms of average accuracy, area under the curve (AUC), the H-measure and Brier score (BS). The model was validated over five real-world credit scoring datasets.
TL;DR: This paper proposes a generalized attribute reduct which not only considers the data but also user preference, and several reduction approaches are summarized to help users to design their appropriate reducts.
Abstract: Attribute reduction plays an important role in the areas of rough sets and granular computing. Many kinds of attribute reducts have been defined in previous studies. However, most of them concentrate on data only, which result in the difficulties of choosing appropriate attribute reducts for specific applications. It would be ideal if we could combine properties of data and user preference in the definition of attribute reduct. In this paper, based on reviewing existing definitions of attribute reducts, we propose a generalized attribute reduct which not only considers the data but also user preference. The generalized attribute reduct is the minimal subset which satisfies a specific condition defined by users. The condition is represented by a group of measures and a group of thresholds, which are relevant to user requirements or real applications. For the same data, different users can define different reducts and obtain their interested results according to their applications. Most current attribute reducts can be derived from the generalized reduct. Several reduction approaches are also summarized to help users to design their appropriate reducts.
TL;DR: This study investigates the relationship between multigranulation rough sets and concept lattices via rule acquisition and algorithm complexity analysis is made for the acquisition of "AND" decision rules, "OR" decisionrules, granular rules and disjunctive rules.
Abstract: Transforming decision systems into formal decision contexts is studied.Relationship between "AND" decision rules and granular rules is discussed.Relationship between "OR" decision rules and disjunctive rules is investigated.Support and certainty factors of different rules are compared.Algorithm complexity of rule acquisition is analyzed. Recently, by combining rough set theory with granular computing, pessimistic and optimistic multigranulation rough sets have been proposed to derive "AND" and "OR" decision rules from decision systems. At the same time, by integrating granular computing and formal concept analysis, Wille's concept lattice and object-oriented concept lattice were used to obtain granular rules and disjunctive rules from formal decision contexts. So, the problem of rule acquisition can bring rough set theory, granular computing and formal concept analysis together. In this study, to shed some light on the comparison and combination of rough set theory, granular computing and formal concept analysis, we investigate the relationship between multigranulation rough sets and concept lattices via rule acquisition. Some interesting results are obtained in this paper: (1) "AND" decision rules in pessimistic multigranulation rough sets are proved to be granular rules in concept lattices, but the inverse may not be true; (2) the combination of the truth parts of an "OR" decision rule in optimistic multigranulation rough sets is an item of the decomposition of a disjunctive rule in concept lattices; (3) a non-redundant disjunctive rule in concept lattices is shown to be the multi-combination of the truth parts of "OR" decision rules in optimistic multigranulation rough sets; and (4) the same rule is defined with a same certainty factor but a different support factor in multigranulation rough sets and concept lattices. Moreover, algorithm complexity analysis is made for the acquisition of "AND" decision rules, "OR" decision rules, granular rules and disjunctive rules.
TL;DR: A new type of fuzzy preference structure, called incomplete HFPRs, is introduced to describe hesitant and incomplete evaluation information in the group decision making (GDM) process and two goal programming models are proposed to derive the priority weights from an incompleteHFPR based on multiplicative consistency and additive consistency respectively.
Abstract: The concept of hesitant fuzzy preference relation (HFPR) has been recently introduced to allow the decision makers (DMs) to provide several possible preference values over two alternatives. This paper introduces a new type of fuzzy preference structure, called incomplete HFPRs, to describe hesitant and incomplete evaluation information in the group decision making (GDM) process. Furthermore, we define the concept of multiplicative consistency incomplete HFPR and additive consistency incomplete HFPR, and then propose two goal programming models to derive the priority weights from an incomplete HFPR based on multiplicative consistency and additive consistency respectively. These two goal programming models are also extended to obtain the collective priority vector of several incomplete HFPRs. Finally, a numerical example and a practical application in strategy initiatives are provided to illustrate the validity and applicability of the proposed models.
TL;DR: An analysis of the tweets in the dataset to investigate the open research issue of how separated figurative linguistic phenomena irony and sarcasm are, with a special focus on the role of features related to the multi-faceted affective information expressed in such texts.
Abstract: The use of irony and sarcasm has been proven to be a pervasive phenomenon in social media posing a challenge to sentiment analysis systems. Such devices, in fact, can influence and twist the polarity of an utterance in different ways. A new dataset of over 10,000 tweets including a high variety of figurative language types, manually annotated with sentiment scores, has been released in the context of the task 11 of SemEval-2015. In this paper, we propose an analysis of the tweets in the dataset to investigate the open research issue of how separated figurative linguistic phenomena irony and sarcasm are, with a special focus on the role of features related to the multi-faceted affective information expressed in such texts. We considered for our analysis tweets tagged with #irony and #sarcasm, and also the tag #not, which has not been studied in depth before. A distribution and correlation analysis over a set of features, including a wide variety of psycholinguistic and emotional features, suggests arguments for the separation between irony and sarcasm. The outcome is a novel set of sentiment, structural and psycholinguistic features evaluated in binary classification experiments. We report about classification experiments carried out on a previously used corpus for #irony vs #sarcasm. We outperform in terms of F-measure the state-of-the-art results on this dataset. Overall, our results confirm the difficulty of the task, but introduce new data-driven arguments for the separation between #irony and #sarcasm. Interestingly, #not emerges as a distinct phenomenon.
TL;DR: The novelty of the proposed SN-GDM is that it can use indirect trust relationship via trusted third partners (TTPs) as a reliable resource to determine experts’ weights.
Abstract: A novel social network based group decision making (SN-GDM) model with experts’ weights not provided beforehand and with the following four tuple information: trust; distrust; hesitancy; and inconsistency, is introduced. The concepts of trust score (TS) and knowledge degree (KD) are defined and combined into a trust order space. Then, a strict trust ranking order relation of trust function values (TFs) is built in which TS and KD play a similar role to the mean and the variance in statistics. After the operational laws of TFs for uninorm operators are built, the uninorm propagation operator is investigated. It can propagate through a network both trust and distrust information simultaneously and therefore it prevents the loss of trust information in the propagating process. When an indirect trust relationship is built, the uninorm trust weighted average (UTWA) operator and the uninorm trust ordered weighted average (UTOWA) operator are defined and used to aggregate individual trust relationship and to obtain their associated ranking order relation. Hence, the most trusted expert is distinguished from the group, and the weights of experts are determined in a reasonable way: the higher an expert is trusted the more importance value is assigned to the expert. Therefore, the novelty of the proposed SN-GDM is that it can use indirect trust relationship via trusted third partners (TTPs) as a reliable resource to determine experts’ weights. Finally, the individual trust decision making matrices are aggregated into a collective one and the alternative with the highest trust order relation is selected as the best one.
TL;DR: The interdisciplinary works in which TAD is reported are surveyed and characterized to characterize the learning strategies, methods and applications; extract the important open issues in TAD and provide the corresponding existing solutions according to the state-of-the-art.
Abstract: Traditional spectral-based methods such as PCA are popular for anomaly detection in a variety of problems and domains. However, if data includes tensor (multiway) structure (e.g. space-time-measurements), some meaningful anomalies may remain invisible with these methods. Although tensor-based anomaly detection (TAD) has been applied within a variety of disciplines over the last twenty years, it is not yet recognized as a formal category in anomaly detection. This survey aims to highlight the potential of tensor-based techniques as a novel approach for detection and identification of abnormalities and failures. We survey the interdisciplinary works in which TAD is reported and characterize the learning strategies, methods and applications; extract the important open issues in TAD and provide the corresponding existing solutions according to the state-of-the-art.
TL;DR: A neuron model based on dendritic mechanisms and a phase space reconstruction (PSR) to analyze the Shanghai Stock Exchange Composite Index, Deutscher Aktienindex, N225, and DJI Average and shows that the attractors obtained can be observed intuitively in a three-dimensional search space, thereby allowing the characteristics of dynamic systems to be analyzed.
Abstract: As a complicated dynamic system, financial time series calls for an appropriate forecasting model. In this study, we propose a neuron model based on dendritic mechanisms and a phase space reconstruction (PSR) to analyze the Shanghai Stock Exchange Composite Index, Deutscher Aktienindex, N225, and DJI Average. The PSR allows us to reconstruct the financial time series, so we can prove that attractors exist for the systems constructed. Thus, the attractors obtained can be observed intuitively in a three-dimensional search space, thereby allowing us to analyze the characteristics of dynamic systems. In addition, using the reconstructed phase space, we confirmed the chaotic properties and the reciprocal to determine the limit of prediction through the maximum Lyapunov exponent. We also made short-term predictions based on the nonlinear approximating dendritic neuron model, where the experimental results showed that the proposed methodology which hybridizes PSR and the dendritic model performed better than traditional multi-layered perceptron, the Elman neural network, the single multiplicative neuron model and the neuro-fuzzy inference system in terms of prediction accuracy and training time. Hopefully, this hybrid technology is capable to advance the research for financial time series and provide an effective solution to risk management.
TL;DR: This paper refine several existing concepts to improve the generality and clarity of former definitions of association rules, and proposes a number of new notions, such as transactional data soft sets, parameter-taxonomic soft sets and M-realizations of parameter sets, to facilitate soft set based association rule mining.
Abstract: Detailed insight into soft set based association rule mining is provided.Drawbacks of some existing definitions are pointed out and rectified.New concepts and algorithms for soft association rule mining are developed.Essentials of new concepts and algorithms are highlighted by two case studies. Association rules, one of the most useful constructs in data mining, can be exerted to capture interesting dependencies between variables in large datasets. Herawan and Deris initiated the investigation of mining association rules from transactional datasets using soft set theory. Unfortunately, some existing concepts in the literature were unable to realize properly Herawan and Deris's initial idea. This paper aims to offer further detailed insights into soft set based association rule mining. With regard to regular association rule mining using soft sets, we refine several existing concepts to improve the generality and clarity of former definitions. Regarding maximal association rule mining based on soft sets, we point out the drawbacks of some existing definitions and offer some way to rectify the problem. A number of new notions, such as transactional data soft sets, parameter-taxonomic soft sets, parameter cosets, realizations and M-realizations of parameter sets are proposed to facilitate soft set based association rule mining. Several algorithms are designed to find M-realizations of parameter sets or extract ź-M-strong and γ-M-reliable maximal association rules in parameter-taxonomic soft sets. We also present an example to illustrate potential applications of our method in clinical diagnosis. Moreover, two case studies are conducted to highlight the essentials of soft set based association rule mining approach.
TL;DR: Experimental results show that the proposed multi-label learning approaches can not only reduce the dimensionality of label-specific features when compared with LIFT, but also achieve satisfactory performance among some popular multi- label learning approaches.
Abstract: We propose two multi-label learning approaches with LIFT reduction.The idea of fuzzy rough set attribute reduction is adopted in our approaches.Sample selection improves the efficiency in feature dimension reduction. In multi-label learning, since different labels may have some distinct characteristics of their own, multi-label learning approach with label-specific features named LIFT has been proposed. However, the construction of label-specific features may encounter the increasing of feature dimensionalities and a large amount of redundant information exists in feature space. To alleviate this problem, a multi-label learning approach FRS-LIFT is proposed, which can implement label-specific feature reduction with fuzzy rough set. Furthermore, with the idea of sample selection, another multi-label learning approach FRS-SS-LIFT is also presented, which effectively reduces the computational complexity in label-specific feature reduction. Experimental results on 10 real-world multi-label data sets show that, our methods can not only reduce the dimensionality of label-specific features when compared with LIFT, but also achieve satisfactory performance among some popular multi-label learning approaches.
TL;DR: SmartSA is introduced, a lexicon-based sentiment classification system for social media genres which integrates strategies to capture contextual polarity from two perspectives: the interaction of terms with their textual neighbourhood (local context) and text genre (global context).
Abstract: The lexicon-based approaches to opinion mining involve the extraction of term polarities from sentiment lexicons and the aggregation of such scores to predict the overall sentiment of a piece of text. It is typically preferred where sentiment labelled data is difficult to obtain or algorithm robustness across different domains is essential. A major challenge for this approach is accounting for the semantic gap between prior polarities of terms captured by a lexicon and the terms' polarities in a specific context (contextual polarity). This is further exacerbated by the fact that a term's contextual polarity also depends on domains or genres in which it appears. In this paper, we introduce SmartSA, a lexicon-based sentiment classification system for social media genres which integrates strategies to capture contextual polarity from two perspectives: the interaction of terms with their textual neighbourhood (local context) and text genre (global context). We introduce an approach to hybridise a general purpose lexicon, SentiWordNet, with genre-specific vocabulary and sentiment. Evaluation results from diverse social media show that our strategies to account for local and global contexts significantly improve sentiment classification, and are complementary in combination. Our system also performed significantly better than a state-of-the-art sentiment classification system for social media, SentiStrength.