Search

Sentiment analysis and opinion mining is an area that has experienced considerable growth over the last decade. This area of research attempts to determine the feelings, opinions, emotions, among other things, of people on something or someone. To do this, natural language techniques and machine learning algorithms are used. This article discusses the problem of extracting sentiment and opinions from a collection of reviews on scientific articles conducted under an international conference on computing in northern Chile. The first aim of this analysis is to automatically determine the orientation of a review and contrast this with the assessment made by the reviewer of the article. This would allow scientists to characterize and compare reviews crosswise and more objectively support the overall assessment of a scientific article. A hybrid approach that combines an unsupervised machine learning algorithm with techniques from natural language processing is proposed to analyze reviews. This method uses part-of-speech (POS) tagging to obtain the syntactic structure of a sentence. This syntactic structure, along with the use of dictionaries, allows determining the semantic orientation of the review through a scoring algorithm. A set of experiments were conducted to evaluate the capability and performance of the proposed approaches relative to a baseline, using standard metrics, such as accuracy, precision, recall, and the F 1 -score. The results show improvements in the case of binary, ternary and a 5-point scale classification in relation to classical machine learning algorithms such as SVM and NB, but they also present a challenge to improve the multiclass classification in this domain. [ABSTRACT FROM AUTHOR]

When the concentration focuses on data mining, frequent itemset mining (FIM) and high-utility itemset mining (HUIM) are commonly addressed and researched. Many related algorithms are proposed to reveal the general relationship between utility, frequency, and items in transaction databases. Although these algorithms can mine FIMs or HUIMs quickly, these algorithms merely take into account frequency or utility as a unilateral criterion for itemsets but the other factors (e.g., distance, price) could be also valuable for decision-making. A new skyline framework has been presented to mine frequent high utility patterns (SFUPs) to better support user decision-making. Several new algorithms have been proposed one after another. However, the Internet of Things (IoT), mobile Internet, and traditional Internet are generating massive amounts of data every day, and these cutting-edge standalone algorithms can not satisfy the new challenge of finding interesting patterns from this data. Big Data uses a distributed architecture in the form of cloud computing to filter and process this data to extract useful information. This paper proposes a novel parallel algorithm on Hadoop as a three-stage iterative algorithm based on MapReduce. MapReduce is used to divide the mining tasks of the whole large data set into multiple independent sub-tasks to find frequent and high utility patterns in parallel. Numerous experiments were done in this paper, and from the results, the algorithm can handle large datasets and show good performance on Hadoop clusters. [ABSTRACT FROM AUTHOR]

Network data is ubiquitous, such as telecommunication, transport systems, online social networks, protein-protein interactions, etc. Since the huge scale and the complexity of network data, former machine learning system tried to understand network data arduously. On the other hand, thought of multi-granular cognitive computation simulates the problem-solving process of human brains. It simplifies the complex problems and solves problems from the easier to harder. Therefore, the application of multi-granularity problem-solving ideas or methods to deal with network data mining is increasingly adopted by researchers either intentionally or unintentionally. This paper looks into the domain of network representation learning (NRL). It systematically combs the research work in this field in recent years. In this paper, it is discovered that in dealing with the complexity of the network and pursuing the efficiency of computing resources, the multi-granularity solution becomes an excellent path that is hard to go around. Although there are several papers about survey of NRL, to our best knowledge, we are the first to survey the NRL from the perspective of multi-granular computing. This paper proposes the challenges that NRL meets. Furthermore, the feasibility of solving the challenges of NRL with multi-granular computing methodologies is analyzed and discussed. Some potential key scientific problems are sorted out and prospected in applying multi-granular computing for NRL research. [ABSTRACT FROM AUTHOR]

An introduction is presented in which the editor discusses articles in the issue on topics including disorderliness of conventional service aggregation; extend the clustering objects in classical sets to fuzzy sets; and the model of deep learning.

Data classification is a data mining task that consists of an algorithm adjusted by a training dataset that is used to predict an object's class (unclassified) on analysis. A significant part of the performance of the classification algorithm depends on the dataset's complexity and quality. Data Complexity involves the investigation of the effects of dimensionality, the overlap of descriptive attributes, and the classes' separability. Data Quality focuses on the aspects such as noise data (outlier) and missing values. The factors Data Complexity and Data Quality are fundamental for the performance of classification. However, the literature has very few studies on the relationship between these factors and to highlight their significance. This paper applies Structural Equation Modeling and the Partial Least Squares Structural Equation Modeling (PLS-SEM) algorithm and, in an innovative manner, associates Data Complexity and Data Quality contributions to Classification Quality. Experimental analysis with 178 datasets obtained from the OpenML repository showed that the control of complexity improves the classification results more than data quality does. Additionally paper also presents a visual tool of datasets analysis about the classification performance perspective in the dimensions proposed to represent the structural model. [ABSTRACT FROM AUTHOR]

Large deployment of wireless sensor networks in various fields bring great benefits. With the increasing volume of sensor data, traditional data collection and processing schemes gradually become unable to meet the requirements in actual scenarios. As data quality is vital to data mining and value extraction, this paper presents a distributed anomaly detection framework which combines cloud computing and edge computing. The framework consists of three major components: k-nearest neighbors, locality sensitive hashing, and cosine similarity. The traditional k-nearest neighbors algorithm is improved by locality sensitive hashing in terms of computation cost and processing time. An initial anomaly detection result is given by the combination of k-nearest neighbors and locality sensitive hashing. To further improve the accuracy of anomaly detection, a second test for anomaly is provided based on cosine similarity. Extensive experiments are conducted to evaluate the performance of our proposal. Six popular methods are used for comparison. Experimental results show that our model has advantages in the aspects of accuracy, delay, and energy consumption. [ABSTRACT FROM AUTHOR]

Named entity recognition (NER) as a crucial technology is widely used in many application scenarios, including information extraction, information retrieval, text summarization, and machine translation assisted in AI-based smart communication and networking systems. As people pay more and more attention to NER, it has gradually become an independent and important research field. Currently, most of the NER models need to manually adjust their hyper-parameters, which is not only time-consuming and laborious, but also easy to fall into a local optimal situation. To deal with such problem, this paper proposes a machine learning-guided model to achieve NER, where the hyper-parameters of model are automatically adjusted to improve the computational performance. Specifically, the proposed model is implemented by using bi-directional encoder representation from transformers (BERT) and conditional random field (CRF). Meanwhile, the collaborative computing paradigm is also fused in the model, while utilizing the particle swarm optimization (PSO) to automatically search for the best value of hyper-parameters in a collaborative way. The experimental results demonstrate the satisfactory performance of our proposed model. [ABSTRACT FROM AUTHOR]

A co-location pattern is a set of spatial features that are strongly correlated in space. However, some of these patterns could be neglected if the prevalence metrics are based solely on the clique (or star) relationship. Hence, the l -reachability co-location pattern is proposed by introducing the l -reachability clique where the members of each instance pair can be reachable to each other in a given step length l. Because the average size of l -reachability co-location patterns tends to be longer, maximal l -reachability co-location pattern mining is researched in this paper. First, some sparsification strategies are introduced to shorten star neighborhood lists of instances in an updated graph called the l -reachability neighbor relationship graph, and then, they are grouped by their corresponding patterns. Second, candidate maximal l -reachability co-location patterns are iteratively detected in a size-independent way on bi-graphs that contain group keys and their intersection sets. Third, the prevalence of each candidate maximal l -reachability co-location pattern is checked in a binary search way with a natural l -reachability clique called the ⌊ l / 2 ⌋ -reachability neighborhood list. Finally, the effectiveness and efficiency of our model and algorithms are analyzed by extensive comparison experiments on synthetic and real-world spatial data sets. [ABSTRACT FROM AUTHOR]

Time series similarity search is an essential operation in time series data mining and has received much higher interest along with the growing popularity of time series data. Although many algorithms to solve this problem have been investigated, there is a challenging demand for supporting similarity search in a fast and accurate way. In this paper, we present a novel approach, TS2BC, to perform time series similarity search efficiently and effectively. TS2BC uses binary code to represent time series and measures the similarity under the Hamming Distance. Our method is able to represent original data compactly and can handle shifted time series and work with time series of different lengths. Moreover, it can be performed with reasonably low complexity due to the efficiency of calculating the Hamming Distance. We extensively compare TS2BC with state-of-the-art algorithms in classification framework using 61 online datasets. Experimental results show that TS2BC achieves better or comparative performance than other the state-of-the-art in accuracy and is much faster than most existing algorithms. Furthermore, we propose an approximate version of TS2BC to speed up the query procedure and test its efficiency by experiment. [ABSTRACT FROM AUTHOR]

This paper presents a novel mechanism to analyze and summarize the statistical correlations among the attributes of a data cube. To perform the analysis and summarization, this paper proposes a new measure of statistical significance. The main reason for proposing the new measure of statistical significance is to have an essential closure property, which is exploited in the summarization stage of the data mining process. In addition to the closure property, the proposed measure of statistical significance has two other important properties. First, the proposed measure of statistical significance is more conservative than the well-known chi-square test in classical statistics and, therefore, inherits its statistical robustness. This paper does not simply employ the chi-square test due to lack of the desired closure property, which may lead to a precision problem in the summarization process. The second additional property is that, though the proposed measure of statistical significance is more conservative than the chi-square test, for most cases, the proposed measure yields a value that is almost equal to the z test, a conventional measurement of statistical significance based on the normal distribution. Based on the closure property addressed above, this paper develops an algorithm to summarize the results from performing statistical analysis in the data cube. Though the proposed measure of statistical significance avoids the precision problem due to having the closure property, its conservative nature may lead to a recall rate problem in the data mining process. On the other hand, if the chi-square test, which does not have the closure property, was employed, then the summarization process may suffer a precision problem. In this paper, we also show a possible application in bioinformatics. We applied the proposed mechanism on a microarray dataset, in order to identify groups of genes with similar expression patterns in subspaces of feature space. [ABSTRACT FROM AUTHOR]

Nowadays more and more information extraction projects need to classify large amounts of text data. The common way to classify text is to build a supervised classifier trained on human-labeled positive and negative examples. In many cases, however, it is easy to label positive examples, but hard to label negative examples. In this paper, we address the problem of building a one-class classifier when only the positive examples are labeled. Previous works on building one-class classifier mostly use positive examples and unlabeled data. In this paper, we show that a configurable one-class classifier such as one-class naive Bayes can be optimized by examining the clustering quality of the classification on target data. We propose to use existing and new quality scores for determining clustering quality of the classification. Experimental analysis with real-world data show that our approach generally achieves high classification accuracy, and in some cases improves the accuracy by more than 10% compared to state-of-art baselines. [ABSTRACT FROM AUTHOR]

Named Entity Recognition (NER) is a challenging learning task of identifying and classifying entity mentions in texts into predefined categories. In recent years, deep learning (DL) methods empowered by distributed representations, such as word- and character-level embeddings, have been employed in NER systems. However, for information extraction in Police narrative reports, the performance of a DL-based NER approach is limited due to the presence of fine-grained ambiguous entities. For example, given the narrative report "Anna stole Ada's car", imagine that we intend to identify the VICTIM and the ROBBER, two sub-labels of PERSON. Traditional NER systems have limited performance in categorizing entity labels arranged in a hierarchical structure. Furthermore, it is unfeasible to obtain information from knowledge bases to give a disambiguated meaning between the entity mentions and the actual labels. This information must be extracted directly from the context dependencies. In this paper, we deal with the Hierarchical Entity-Label Disambiguation problem in Police reports without the use of knowledge bases. To tackle such a problem, we present HELD, an ensemble model that combines two components for NER: a BLSTM-CRF architecture and a NER tool. Experiments conducted on a real Police reports dataset show that HELD significantly outperforms baseline approaches. [ABSTRACT FROM AUTHOR]

Cluster analysis or clustering is one of the most important and widely used techniques for data exploration and knowledge discovery that concerned with partitioning a set of objects in such a way that objects in the same groups, called clusters, are more similar to each other than to those in other clusters. However, obtaining the clusters that exhibit high withincluster similarity or homogeneity and high between-cluster dissimilarity or heterogeneity is critically depended on the similarity notion, which has not been yet clearly defined for clustering purposes. Distance and correlation are the most important and commonly used mathematics and statistics-based similarity measurements in the literature of the clustering, respectively. In this paper, the learning speed of the supervised neural networks is proposed as novel intelligent similarity measurement for unsupervised clustering problems. On the other hand, the main aim of this paper is to answer this question that can convergence speed of the different objects to the given target be used for measuring the similarity. Empirical results of the simulated data sets indicate that the proposed measurement not only can be used as similarity measurement in clustering tasks, but also can produce accurate results. In this way, for first time and in contrast of the literature, it is demonstrated that a supervised model can be used for handling the unsupervised tasks. [ABSTRACT FROM AUTHOR]

Community detection is one of the main challenges in social network analysis. Since the issue of community detection is considered as a NP-hard problem, Evolutionary algorithms have been used as one of the most effective approaches. In this paper, a multi-objective particle swarm optimization algorithm and its extended versions are proposed. The aforementioned algorithm uses an opposition-based method for producing an initial swarm. It optimizes two objective functions at the same time which represents a partition of the network as well as using a mutation operator for handling the problem in high dimensions. The performances of the proposed algorithm and its extended versions have been evaluated on real networks. The result represented the efficiency of proposed methods. Also an optimum value is suggested for aforementioned algorithm that can be said with less complexity and calculations, proposed algorithm achieved the acceptable amount of accuracy. The remarkable thing is the better performance of algorithm as the size of social network grows. [ABSTRACT FROM AUTHOR]

Recommender systems apply machine learning and data mining techniques for filtering unseen information, and they can provide an opportunity to predict whether a user would be interested in a given item. The main types of recommender systems are collaborative filtering (CF) and content-based filtering, which suffer from scalability and data sparsity resulting in poor quality recommendations and reduced coverage. There are two incremental algorithms based on Singular Value Decomposition (SVD) with high scalability for recommender systems which are named the incremental SVD algorithm and incremental Approximating the Singular Value Decomposition (ApproSVD) algorithm. In both mentioned methods, the estimated value of rank for approximating the recommender systems' data matrix is chosen experimentally in the related literature. In this paper, we investigate the role of singular values for estimating a more reliable amount of rank in the mentioned dimensionality reduction techniques to improve the recommender systems' performance. In other words, we offered a strategy for choosing the optimal rank that approximates the data matrix more accurately in incremental algorithms with the help of singular values. The numerical results illustrate that the suggested strategy improves the accuracy of the recommendations and run times of both algorithms when employs for Movielens, Netflix, and Jester dataset. [ABSTRACT FROM AUTHOR]

Feature selection is an important data preprocessing in data mining and machine learning, that can reduce the number of features without deteriorating model's performance. Recently, sparse regression has received considerable attention in feature selection task due to its good performance. However, because the l 2 , 0 -norm regularization term is non-convex, this problem is hard to solve, and most of the existing methods relaxed it by l 2 , 1 -norm. Unlike the existing methods, this paper proposes a novel method to solve the l 2 , 0 -norm regularized least squares problem directly based on iterative hard thresholding, which can produce exact row-sparsity solution for weights matrix, and features can be selected more precisely. Furthermore, two homotopy strategies are derived to reduce the computational time of the optimization method, which are more practical for real-world applications. The proposed method is verified on eight biological datasets, experimental results show that our method can achieve higher classification accuracy with fewer number of selected features than the approximate convex counterparts and other state-of-the-art feature selection methods. [ABSTRACT FROM AUTHOR]

The integration of incomplete and uncertain information has emerged as a crucial issue in many application domains, including data warehousing, data mining, data analysis, and artificial intelligence. This paper proposes a novel approach of mediation-based integration for integrating these types of information from heterogeneous relational databases. We present in detail the different processes in the layered architecture of the proposed flexible mediator system. The integration process of our mediator is based on the use of fuzzy logic and semantic similarity measures for more effective integration of incomplete and uncertain information. We also define fuzzy views over the mediator's global fuzzy schema to express incomplete and uncertain databases and specify the mappings between this global schema and these sources. Moreover, our approach provides intelligent data integration, enabling efficient generation of cooperative answers from similar ones, retrieved by queried flexible wrappers. These answers contain information that is more detailed and complete than the information contained in the initial answers. A thorough experiment verifies our approach improves the performance of data integration under various configurations. [ABSTRACT FROM AUTHOR]

Sentiment analysis is a field that has experienced considerable growth over the last decade. This area of research attempts to determine the opinions of people on something or someone. This article introduces a novel technique for association rule extraction in text called Extended Association Rules in Semantic Vector Spaces (AR-SVS). The objective of this analysis is to explore the feasibility of applying AR-SVS in the field of opinion mining and sentiment analysis. This new method is based on the construction of association rules, which are extended through a similarity criteria for terms represented in a semantic vector space. The method was evaluated on a sentiment analysis data set consisting of scientific paper reviews. A quantitative and qualitative analysis is done with respect to the classification performance and the generated rules. The results show that the method is competitive compared to the baseline provided by Naïve Bayes and Support Vector Machines. Furthermore, previous work on the evaluation of scientific paper reviews (the Scoring Algorithm) has been used in conjunction with association rules to obtain a method that shows a superior behaviour compared to the baseline. Finally, additional experiments are performed on various multidomain data sets in order to evaluate the results of AR-SVS in different settings. [ABSTRACT FROM AUTHOR]

Time series classification and class imbalance problem are two common issues in a multitude of real-life scenarios. This paper simultaneously explores both issues with deep convolution neural networks (CNNs). Because standard networks treat the majority and minority classes with same class weights, most CNN-based networks fail to classify imbalanced time series. Until recently, there is very little work applying deep learning to imbalanced time series classification (ITSC). Thus, we propose an adaptive cost-sensitive learning strategy to address the ITSC problem. The standard CNN is modified to a cost-sensitive network (CS-CNN), which is able to punish the misclassified samples using a class-dependent cost matrix. Moreover, this cost matrix is automatically updated based on overall class distribution and the CS-CNN's training performance. The proposed method is extended to FCN, LSTM-FCN and ResNet. It is experimentally tested on five public benchmark UCR datasets and a real-life large volume dataset. Four cost-sensitive CNN-based networks are compared with several data samplers and two traditional ITSC methods. The modified networks are superior in all metrics. Results show that cost-sensitive networks successfully complete the ITSC tasks. [ABSTRACT FROM AUTHOR]

Dynamic environment data generators are very often in real-world that produce data streams. A data source of a dynamic environment generates data streams in which the underlying data distribution changes very frequently with respect to time and hence results in concept drifts. As compared to the stationary environment, learning in the dynamic environment is very difficult due to the presence of concept drifts. Learning in dynamic environment requires evolutionary and adaptive approaches to be accommodated with the learning algorithms. Ensemble methods are commonly used to build classifiers for learning in a dynamic environment. The ensemble methods of learning are generally described at three very crucial aspects, namely, the learning and testing method employed, result integration method and forgetting mechanism for old concepts. In this paper, we propose a novel approach called Age Decay Accuracy Weighted (ADAW) ensemble architecture for learning in concept drifting data streams. The ADAW method assigned weights to the component classifiers based on its accuracy and its remaining life-time in the ensemble is such a way that ensures maximum accuracy. We empirically evaluated ADAW on benchmark artificial drifting data stream generators and real datasets and compared its performance with ten well-known state-of-the-art existing methods. The experimental results show that ADAW outperforms over the existing methods. [ABSTRACT FROM AUTHOR]

Biological network classification is an eminently challenging task in the domain of data mining since the networks contain complex structural information. Conventional biochemical experimental methods and the existing intelligent algorithms still suffer from some limitations such as immense experimental cost and inferior accuracy rate. To solve these problems, in this paper, we propose a novel framework for Biological graph classification named Biogc, which is specifically developed to predict the label of both small-scale and large-scale biological network data flexibly and efficiently. Our framework firstly presents a simplified graph kernel method to capture the structural information of each graph. Then, the obtained informative features are adopted to train different scale biological network data-oriented classifiers to construct the prediction model. Extensive experiments on five benchmark biological network datasets on graph classification task show that the proposed model Biogc outperforms the state-of-the-art methods with an accuracy rate of 98.90% on a larger dataset and 99.32% on a smaller dataset. [ABSTRACT FROM AUTHOR]

Time series clustering has been attracted great interest in the last decade. Most time series clustering works focus on clustering algorithms and similarity measures. Recently a u-shapelet-based time series clustering method has been proposed which can not only hold a high performance of clustering but also offer an acceptable interpretation of clustering result. Time complexity is the major issue in the process of discovering u-shapelets, particularly for huge datasets. In this paper, we propose a Random Local Search algorithm that reduces the time to discover u-shapelets, meanwhile keeps or even improves the quality of clustering. Our algorithm first randomly samples subsequences from time series to reduce the time cost of the u-shapelet search problem. It then uses a local search strategy to make clustering result more excellent and stable. We test our approach extensively on 27 UCR time series datasets, and obtain improved clustering accuracies over existing approaches. The experiment shows that our method outperforms the primitive algorithm by up to 2 orders of magnitude on some datasets in runtime. Because the method allows a fast u-shapelet discovery, it is feasible to apply u-shapelets clustering on large datasets. [ABSTRACT FROM AUTHOR]

Clinical research and drug development trials generate large amounts of data. Due to the dispersed nature of clinical trial data across multiple sites and heterogeneous databases, it remains a challenge to harness these trial data for analytics to gain more understanding about the implementation of studies as well as disease processes. Moreover, the veracity of the results from analytics is difficult to establish in such datasets. We make a two-fold contribution in this paper: First, we provide a mechanism to extract task-relevant data using Master Data Management (MDM) from a clinical trial database with data spread over several domain datasets. Second, we provide a method for validating findings by collaborative utilization of multiple data mining techniques, namely: classification, clustering, and association rule mining. Overall, our approach aims at extracting useful knowledge from data collected during clinical trials to enable the development of faster and cheaper clinical trials that more accurate and impactful. For a demonstration of the efficacy of our proposed methods, we utilized the following datasets: (1) the National Institute on Drug Abuse (NIDA) data share repository and (2) the data from the Osteoarthritis initiative (OAI), where we found real-world implications in validating the findings using multiple data mining methods in a collaborative manner. The comparative results with existing state of the art techniques show the usefulness and high accuracy of our methods. [ABSTRACT FROM AUTHOR]

This paper use multinomial nave Bayes to improve multi-label classification methods in a number of ways. First, we use the value weighting method, a new fine-grained weighting method, to calculate the weights of the feature values. Second, we employ a co-training method to incorporate the dependencies among the class values. The results of our experiments show that the proposed approach outperforms other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Regression trees are powerful tools in data mining for analyzing data sets. Observations are usually divided into homogeneous groups, and then statistical models for responses are derived in the terminal nodes. This paper proposes a new approach for regression trees that considers the dependency structures among covariates for splitting the observations. The mathematical properties of the proposed method are discussed in detail. To assess the accuracy of the proposed model, various criteria are defined. The performance of the new approach is assessed by conducting a Monte-Carlo simulation study. Two real data sets on classification and regression problems are analyzed by using the obtained results. [ABSTRACT FROM AUTHOR]

The article discusses various papers published within this issue, including one by V. Nguyen and T. Washio on modeling dynamics substate chains among massive states and another by T. Calders, et al. on mining frequent items in stream using flexible windows.

A probabilistic radial basis function (PRBF) network is an effective non-linear classifier. However, similar to most other neural network models it is non-transparent, which makes its predictions difficult to interpret. In this paper we show how a one-variable-at-a-time and an all-subsets explanation method can be modified for an equivalent and more efficient use with PRBF network classifiers. We use several artificial and real-life data sets to demonstrate the usefulness of the visualizations and explanations of the PRBF network classifier. [ABSTRACT FROM AUTHOR]

Class imbalance is often a problem in various real-world datasets, where one class contains a small number of data and the other contains a large number of data. It is notably difficult to develop an effective model using traditional data mining and machine learning algorithms without using data preprocessing techniques to balance the dataset. Oversampling is often used as a pretreatment method for imbalanced datasets. Specifically, synthetic oversampling techniques focus on balancing the number of training instances between the majority class and the minority class by generating extra artificial minority class instances. However, the current oversampling techniques simply consider the imbalance of quantity and pay no attention to whether the distribution is balanced or not. Therefore, this paper proposes an entropy difference and kernel-based SMOTE (EDKS) which considers the imbalance degree of dataset from distribution by entropy difference and overcomes the limitation of SMOTE for nonlinear problems by oversampling in the feature space of support vector machine classifier. First, the EDKS method maps the input data into a feature space to increase the separability of the data. Then EDKS calculates the entropy difference in kernel space, determines the majority class and minority class, and finds the sparse regions in the minority class. Moreover, the proposed method balances the data distribution by synthesizing new instances and evaluating its retention capability. Our algorithm can effectively distinguish those datasets with the same imbalance ratio but different distribution. The experimental study evaluates and compares the performance of our method against state-of-the-art algorithms, and then demonstrates that the proposed approach is competitive with the state-of-art algorithms on multiple benchmark imbalanced datasets. [ABSTRACT FROM AUTHOR]

In data mining, finding frequent itemsets is a critical step to discovering association rules. The number of frequent itemsets may, however, be huge if the threshold of minimum support is set at a low value or the number of items in the transaction database to be mined is large. In the past, some approaches were thus proposed to keep frequent itemsets with compact representation. For example, the approach of maximal itemsets keeps a borderline composed of the maximal itemsets, which separate frequent itemsets from non-frequent ones. It can recover all the frequent itemsets, but cannot get their actual frequencies back. On the contrary, the approach of closed itemsets can correctly recover each frequent itemset and its frequency. Besides, another approach called reference itemsets can recover each frequent itemset and approximately estimate its frequency. In this paper, we propose an efficient algorithm to recover each frequent itemset and its approximate frequency based on the kept maximal itemsets, frequent 1-itemsets, their supports, and some key information. The maximal frequent itemsets are used to recover all frequent itemsets, which are then organized into a simple flow network with levels. Next, the kept key information is used to derive approximate supports of the frequent itemsets in the flow network through the flow process. Finally, a series of experiments are conducted to show the compression effects of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Recently, music recommender systems have been proposed to help users obtain the interested music. Traditional recommender systems making attempts to discover users' musical preferences by ratings always suffer from problems of rating diversity, rating sparsity and lack of ratings. These problems result in unsatisfactory recommendation results. To deal with traditional problems, in this paper, we propose a novel music recommender system, namely Multi-modal Music Recommender system (MMR), which integrates social and collaborative information to predict users' preferences. In this work, the playcounts are transformed into collaborative information to cope with problem of lack of rating information, while item tags and artist tags are employed as social information to cope with problems of rating diversity and rating sparsity. Through optimizing the integrated social-and-collaborative information, the users' preferences can be inferred more accurately and efficiently. The experimental results reveal that, three problems can be alleviated significantly and our proposed method outperforms other state-of-the-art recommender systems in terms of RMSE (Root Mean Square Error) and NDCG (Normalized Discount Cumulative Gain). [ABSTRACT FROM AUTHOR]

Association Rules (ARs) and Approximate Dependencies (ADs) are significant fields in data mining and the focus of many research efforts. This knowledge, extracted by traditional mining algorithms becomes inexact when new data operations are executed, a common problem in real-world applications. Incremental mining methods arise to avoid re-runs of those algorithms from scratch by re-using information that is systematically maintained. These methods are useful to extract knowledge in dynamic environments. However, the implementation of algorithms only to maintain previously discovered information creates inefficiencies. In this paper, two active algorithms are proposed for incremental maintenance of previous discovered ARs and ADs, inspired by efficient computation of changes. These algorithms operate over a generic form of measures to efficiently maintain a wide range of rule metrics simultaneously. We also propose to compute data operations at real-time, in order to create a reduced relevant instance set. The algorithms presented do not discover new knowledge; they are just created to efficiently maintain previously extracted valuable information. Experimental results in real education data and repository datasets show that our methods achieve a good performance. In fact, they can significantly improve traditional mining, incremental mining, and a naïve approach. [ABSTRACT FROM AUTHOR]

To increase their competitiveness, many industrial companies monitor their production process, collecting large amount of measurements. This paper describes a technique using this data to improve the performance of a monitored process. In particular we wish to find a set of rules, i.e. intervals on a reduced number of parameters, for which an output value is maximized. The model-free optimization problem to solve is to find a box, restricted on a limited amount of dimensions, with the maximum mean value of the included points. This article compares a machine learning-based heuristic to the solution computed by a mixed-integer linear program on real-life databases from steel and glass manufacturing. Computational results show that the heuristic obtains comparable solutions to the mixed integer linear approach. However, the exact approach is computationally too expensive to tackle real life databases. Results show that the restriction of five process parameters, on these databases, may improve the quality of the process by 50%. [ABSTRACT FROM AUTHOR]

Outlier detection is an interesting issue in data mining and machine learning. In this paper, to detect outliers, an information-entropy-based k-nearest neighborhood relevant outlier factor algorithm is proposed that is combined with Shannon information theory and the triangle pruning strategy. The algorithm accounts for the data points whose k-nearest neighbors are distributed on the edge of the range within the designated radius. In particular, the neighborhood influence on each point is considered to address the problem of information concealment and submergence. Information entropy is used to calculate the weights to distinguish the importance of each attribute. Then, based on the attribute weights, the improved pruning strategy reduces the computational complexity of the subsequent procedures by removing some inliers and obtaining the outlier candidate dataset. Finally, according to the weighted distance between the objects in the candidate dataset and those in the original dataset, the algorithm calculates the dissimilarity between each object and its k-nearest neighbors. The data points with the top r dissimilarity are regarded as the outliers. Experimental results show that, compared to existing methods, the proposed approach improves pruning and detection rates while maintaining the coverage rate. [ABSTRACT FROM AUTHOR]

Despite the benefits of data mining in a wide range of applications, this technique has raised some issues related to the privacy and security of individuals. Due to these issues, data owners may prevent to share their sensitive information with data miners. On the other hand, in distributed environments, other issues related to the distribution of data will raise, which will make the preservation of privacy more challengeable. To solve these problems, different privacy preserving data mining (PPDM) techniques have been introduced. In this paper, a new privacy preserving clustering (PPC) technique for horizontally and vertically distributed datasets is proposed. The proposed technique uses Haar wavelet transforms (HWT) and scaling data perturbation (SDP) to achieve both data hiding and data reduction for protecting private numerical attribute values in distributed datasets. The results of our evaluations demonstrated that the proposed technique provides a proper degree of privacy and quality of clustering for distributed datasets and also runs fast. Our experiments have also shown that the proposed technique provides better privacy and clustering results comparing to the other existing privacy preserving clustering techniques applicable to distributed datasets. The proposed algorithms and the results of their experimental evaluations using different datasets are presented in this paper. [ABSTRACT FROM AUTHOR]

Traditional research on preserving privacy in data mining focuses on time-invariant privacy issues. With the emergence of time series data mining, traditional snapshot-based privacy issues need to be extended to be multi-dimensional with the addition of time dimension. We find current techniques to preserve privacy in data mining are not effective in preserving time-domain privacy. We present the data flow separation attack on privacy in time series data mining, which is based on blind source separation techniques from statistical signal processing. Our experiments with real data show that this attack is effective. By combining the data flow separation method and the frequency matching method, an attacker can identify data sources and compromise time-domain privacy. We propose possible countermeasures to the data flow separation attack in the paper. [ABSTRACT FROM AUTHOR]

Sentiment classification in text documents is an active data mining research topic in opinion retrieval and analysis. Different from previous studies concentrating on the development of effective classifiers, in this paper, we focus on the extraction and validation of unexpected sentences issued in sentiment classification. In this paper, we propose a general framework for determining unexpected sentences. The relevance of the extracted unexpected sentences is assessed in the context of text classification. In the experiments, we present the extraction of unexpected sentences for sentiment classification within the proposed framework, and then evaluate the influence of unexpected sentences on the quality of classification tasks. The experimental results show the effectiveness and usefulness of our proposed approach. [ABSTRACT FROM AUTHOR]

In some cases, currency crises are followed by strong recessions (e.g., recent Asian and Argentinean crises), but in other cases they are not. This paper uses Self-Organizing Maps (SOM) to search for meaningful associations between speculative attacks' real effects and 28 variables that characterize the economic, financial, legal, and socio-political structure of the country at the onset of the attack. SOM is a neural network-based generalization of Principal Component Analysis (PCA) that provides an efficient non-linear projection of the multidimensional data space on a curved surface. This paper finds a strong association of speculative attacks` real effects with fundamentals and the banking sector structure. [ABSTRACT FROM AUTHOR]

Mining frequent simultaneous attribute co-variations in numerical databases is also called frequent gradual pattern problem. Few efficient algorithms for automatically extracting such patterns have been reported in the literature. Their main difference resides in the variation semantics used. However in applications with temporal order relations, those algorithms fail to generate correct frequent gradual patterns as they do not take this temporal constraint into account in the mining process. In this paper, we propose an approach for extracting frequent gradual patterns for which the ordering of supporting objects matches the temporal order. This approach considerably reduces the number of gradual patterns within an ordered data set. The experimental results show the benefits of our approach. [ABSTRACT FROM AUTHOR]

By providing a complete record of time use for a given population, time use studies enable investigators to test various hypotheses concerning that behavior. However, the large number and variety of activity combinations that are relevant in time allocation choices and, therefore, time use analysis, makes measuring or even fully identifying all of them impossible without the proper data mining tools. In this paper, we propose a framework for mining sequences of activities to capture more complex patterns than those currently available on how individuals organize their days. The proposed framework was applied to the American Time Use Surveys (ATUS) dataset to explore individual time allocation behavior, identifying sequences of activities that are frequent. For example, patterns such as the preferred activities that are performed before and after specific activities (such as paid work or leisure) are discussed in terms of their frequency. Such patterns are not easy to reveal using traditional descriptive analysis. [ABSTRACT FROM AUTHOR]

Uncertain data are data accompanied with probability, which makes frequent itemset mining more challenging. This paper focuses on the problem of mining probabilistic maximal frequent itemsets. We redefine the concept of probabilistic maximal frequent itemset to be consistent with the traditional definition and provide a better view on how to devise pruning strategies. A tree-based index called the probabilistic maximal frequent itemset tree is constructed to maintain the probabilistic frequent itemsets. We proposed a depth-first probabilistic maximal frequent itemset mining algorithm to bottom-up generate the exact results, in which support and expected support are used to estimate the range of probabilistic support, enabling the frequency of an itemset to be inferred with less runtime and memory usage. Also, superset pruning is employed to further reduce mining cost. Nevertheless, certain probabilistic supports have to be computed when the minimum support is low, which may result in highly increased mining speed. This problem is addressed in our approximate probabilistic maximal frequent itemset mining method, which uses the expected support to directly compute the probabilistic support. Theoretical analysis and experimental studies demonstrate that our proposed algorithms have high accuracy, expend less computational time and use less memory, and significantly outperform the TODIS-MAX [20] state-of-the-art algorithm. [ABSTRACT FROM AUTHOR]

This paper describes the use of a concept hierarchy for improving the results of association rule mining. Given a large set of tuples with demographic information and personal interest information, association rules can be derived, that associate ages and gender with interests. However, it is a problem to come up with rules with high support whenever the mined data set is sparse. On the other hand, if rules with high support can be generated, they tend to involve interests that are too abstract to be of practical use. To overcome the first problem, we have developed a method of raising data instances to higher levels in the ontology. In this paper we give a formal definition of the raising operation. We also show that in some cases data mining with raised data leads to rules that better represent the reality. In order to avoid the second problem, namely rules that are too abstract, we formulate a notion of an optimal target level for the raising operation. We then derive two estimates for this optimal raising level. Knowing to which level to raise reduces the computational effort of raising to several levels and reduces the user effort of selecting those mined rules that best fit her/his needs from a large candidate set. [ABSTRACT FROM AUTHOR]

Focuses on mining official data, which are data collected in censuses and statistical surveys by the National Statistics Institute, as well as administrative and registration records collected by government departments and local authorities. Purpose of official data; Information on geo-referenciation; Overview on the articles published in a 2003 issue of "Intelligent Data Analysis."

The validation of any database mining methodology goes through an evaluation process where benchmarks availability is essential. In this paper, we aim to randomly generate relational database benchmarks that allow to check probabilistic dependencies among the attributes. We are particularly interested in Probabilistic relational models (PRMs). These latter extend Bayesian networks (BNs) to a relational data mining context that enable effective and robust reasoning about relational data structures. Even though a panoply of works have focused, separately, on Bayesian networks and relational databases random generation, no work has been identified for PRMs on that track. This paper provides an algorithmic approach allowing to generate random PRMs from scratch to cover the absence of generation process. The proposed method allows to generate PRMs as well as synthetic relational data from a randomly generated relational schema and a random set of probabilistic dependencies. This can be of interest for machine learning researchers to evaluate their proposals in a common framework, as for databases designers to evaluate the effectiveness of the components of a database management system. [ABSTRACT FROM AUTHOR]

Large amounts of data generated everyday by different organizations for various purposes have catalyzed research opportunities related to data science. Publishing raw data may raise security concerns among the users or actors who have provided some sensitive information in the raw data. Over the years, it has been observed that attackers can very easily exploit the sensitive information as well as the identity of the users from the raw data. Thus, to protect the identity of the users in the anonymized data, the notion of k-anonymity and its improved version l-diversity have been proposed. Several algorithms have been developed to achieve k-anonymity as well as l-diversity for both relational micro-data and social network data. In this paper, we propose an approach based on the eigenvector centrality value of the individual nodes to achieve k-anonymity as well as l-diversity by adding noise nodes in the raw data. In the process of adding noise nodes, we focused on adding noise nodes in such an intelligent manner so that they have very little influence on the anonymized data. Our proposed algorithm also ensures minimal changes in the social influence of the nodes in the anonymized data which are already present in the raw data. Through various measures and experiments, we establish the effectiveness of our proposed algorithm over the several existing social network anonymization techniques. [ABSTRACT FROM AUTHOR]

The Particle Swarm Optimization (PSO) is a heuristic optimization technique-based swarm intelligence that can be applied to solving many real-world optimization problems. However, the standard PSO algorithm can easily get trapped in the local optima and has slow convergence speed, and these drawbacks have hindered its further development in all fields. In this paper, a new optimization method based on neighbor heuristic and Gaussian cloud learning is introduced in order to improve the performance of traditional PSO (NHPSO). The NHPSO consists of two main steps. First, by analyzing the relationship among particles in the evolutionary process, a neighbor heuristic mechanism is performed to improve the search efficiency and convergence speed. In addition, a Gaussian cloud learning strategy is introduced to enhance population diversity and balance the global and local search abilities. The performance of the NHPSO is tested using 12 benchmark functions and 6 shifted functions. Results show that NHPSO is superior to the recent variants of PSO in terms of convergence speed, solution accuracy, algorithm efficiency and robustness. [ABSTRACT FROM AUTHOR]

Outlier detection is an important task in data mining. In this paper, a novel outlier detection algorithm is proposed, which integrates the local density with the global distance seamlessly. In the proposed method, an integrated outlier factor is used to measure the detecting accuracy. A comprehensive experimental study on both synthetic and real-life datasets shows that the proposed method is more effective than some typical outlier detection methods, including Relative Density-based Outlier Score (RDOS), INFLuenced Outlierness (INFLO), Local Outlier Factor (LOF) and Local Distance-based Outlier detection Factor (LDOF). [ABSTRACT FROM AUTHOR]

Data gathered from real world often contains label noise, which is harmful to the quality of data. Moreover, any data mining process suffers a deterioration when it is applied on noisy data. In this paper, a new approach is proposed to improve data quality by correcting mislabeled data. The proposed method employs a procedure to estimate the level of the noise in the data and combines this noise estimation with a correction process. A clustering method and k nearest neighbors approach are applied in the correction process. Extensive experimental results using real-world data sets from UCI machine learning repository are provided. The empirical study shows that our approach successfully improves data quality in many cases and outperforms several correction methods. [ABSTRACT FROM AUTHOR]

The identification of frequent patterns plays a key role in mining association rules. FP-growth is a fundamental algorithm for frequent pattern mining. It employs a prefix tree structure (FP-Tree) and a recursive mining process to discover frequent patterns. However, the performance of FP-growth is closely related to the total number of recursive calls, which leads to poor performance when multiple conditional FP-trees are required to be constructed. This paper proposes highly compressed FP-tree (HCFP-tree). This increases prefix sharing and reduces the number of nodes in the prefix tree. Based on HCFP-tree, we design a new algorithm called HCFP-growth. This algorithm greatly reduces the number of recursive calls required to mine full frequent patterns. Experiments conducted on various types of datasets demonstrate that HCFP-growth is always among the fastest algorithms. It also consumes the least memory in many cases, and its memory consumption is comparable to that of existing algorithms in other cases. [ABSTRACT FROM AUTHOR]

The use of multivariate time series generation in industrial settings such as the automotive industry continues to increase. The complexity of data analysis requirements in such industries has led to an urgent need to develop effective methods for extracting structural information from data based on the clustering of system behavior time series. Because there are complex interactions between vehicle data variables, the time series clustering of single variables can lead to insufficient results. To the best of our knowledge, only univariate dynamic time warping (DTW) approaches have thus far been applied in an automotive context. To close this research gap, this paper presents a review of generic approaches in multivariate dynamic time warping (MDTW) to determine the most promising approaches for use in the automotive domain. Four approaches are found to be particularly useful for tasks such as the objective assessment of subjective driving perceptions. [ABSTRACT FROM AUTHOR]

Transfer learning empowers machine learning algorithms the ability to train a model on a given task, capture the existing relationship in the data and reuse it for another task in the same or similar domain. In this paper, we present a Multi-view deep unsupervised transfer leaning via joint auto-encoder coupled with dictionary learning (MVT-DAE). In the proposed approach knowledge transfer is done in two stages. First, we perform multi-view dictionary learning based on low-rank tensor regularization in the source domain to learn the common intrinsic relationship among views. Then we transfer the learned dictionaries to the target domain, and we construct a new representation of both domain via sparse coding. In the second stage, we use two deep auto-encoders (DAE) from both sources to perform parameters transfer. Each DAE consist of an embedding layer and a label layer. The embedding layer reconstructs the input while the label layer is used to encode the label information. To relax the distribution distance of the embedding and label layer between the source and the target domain a joint maximum mean discrepancy (JMMD) is employed. Learning can be done via Stochastic gradient descent by sharing the embedding and label layer weights with the target domain. We conduct extensive experiments in two real-world datasets which demonstrated the effectiveness of our approach compared with the different state of the art baseline. [ABSTRACT FROM AUTHOR]