Intelligent Data Analysis: An International Journal

Deviating behavior within an organization can lead to unexpected results. The effects of deviations are often negative, but sometimes also positive. Therefore, it is useful to detect deviations from event logs which record all the behavior of the organization. However, existing model-based and cluster-based approaches are inaccurate or slow when dealing with complex event logs, i.e. logs of less structured processes having many activities and many possible paths. This paper proposes a novel approach that is faster than cluster-based approaches because it creates a so-called profile which is less time-consuming than creating clusters. Furthermore, the approach is also more accurate than model-based approaches because we use an iterative approach to improve the result. Our experiments show that approach outperforms existing techniques in a variety of circumstances.

Benchmarking is among the most widely adopted practices in business today. However, to the best of our knowledge, conducting multidimensional benchmarking in data warehouses has not been explored from a technical efficiency perspective. In this paper, we formulate benchmark queries in the context of data warehousing and business intelligence, and develop algorithms to answer benchmark queries efficiently. Our methods employ a few interesting ideas and the state-of-the-art data cube computation techniques to reduce the number of aggregate cells that need to be computed and indexed. An empirical study using the TPC-H and the Weather data sets demonstrates the efficiency and the scalability of our methods.

A common task in data analysis is to find the appropriate data sample whose properties allow us to infer the parameters and behavior of the data population. In data mining this task makes sense since usually the population is significantly huge, and thus it is required (for practical reasons) to obtain a subset that preserves its properties. In this regard, statistics offers some sampling techniques usually based on asymptotic results from the Central Limit Theorem. The effectiveness of such ways is bounded by several considerations as the sampling strategy (simple with or without replacement, stratified, cluster-based, etc.), the size of the population and the dimensionality of the space of the data. Due to these considerations alternative proposals are necessary. We propose a method based on a measure of information in terms of Shannon’s Entropy. Our idea is to find the optimal sample whose information is as similar as possible to the information of the population, subject to several constraints. Finding such sample represents a hard optimization problem whose feasible space disallows the use of traditional optimization techniques. To solve it, we resort to a breed of Genetic Algorithm called Eclectic Genetic Algorithm. We test our method with synthetic datasets; the results show that our method is suitable. For completeness, we used several datasets from real problems; the results confirm the effectiveness of our proposal and allow us to visualize different applications. Finally, we establish a baseline based on selection instance methods as a point reference to measure the effectiveness of our method.

Conventional clustering algorithms optimize a single criterion, which may not conform to diverse needs of multidimensional data science. This paper proposes a new clustering algorithm that solves multiple clustering issues, called clustering by Marked Point Process (ClusterMPP). It is a new, efficient, scalable and unsupervised density-based clustering algorithm. ClusterMPP simulates a proposed Marked Point Process (MPP) to find clusters of complex shapes present in the raw data space. The outputs of this new algorithm, in the first step, are the observations belonging to each cluster mode called “prototypes”. The classification process is achieved, in the second step, using an improved KNN algorithm. We conduct intensive experiments to compare ClusterMPP with the most well-known algorithms. The results of ClusterMPP proved its efficiency on high complex and scalable datasets.

Social media provides unprecedented opportunities for people to disseminate information and share their opinions and views online. Extracting events from social media platforms such as Twitter could help in understanding what is being discussed. However, event extraction from social text streams poses huge challenges due to the noisy nature of social media posts and dynamic evolution of language. We propose a generic unsupervised framework for exploring events on Twitter which consists of four major steps, filtering, pre-processing, extraction and categorization, and post-processing. Tweets published in a certain time period are aggregated and noisy tweets which do not contain newsworthy events are filtered by the filtering step. The remaining tweets are pre-processed by temporal resolution, part-of-speech tagging and named entity recognition in order to identify the key elements of events. An unsupervised Bayesian model is proposed to automatically extract the structured representations of events in the form of quadruples < entity, keyword, date, location > and further categorize the extracted events into event types. Finally, the categorized events are assigned with the event type labels without human intervention. The proposed framework has been evaluated on over 60 million tweets which were collected for one month in December 2010. A precision of 78.01% is achieved for event extraction using our proposed Bayesian model, outperforming a competitive baseline by nearly 13.6%. Moreover, events are also clustered into coherence groups with the automatically assigned event type labels with an accuracy of 42.57%.

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 within-cluster 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.

Multi-label learning has attracted significant attention from machine learning and data mining over the last decade. Although many multi-label classification algorithms have been devised, few research studies focus on multi-assignment clustering (MAC), in which a data instance can be assigned to multiple clusters. The MAC problem is practical in many application domains, such as document clustering, customer segmentation and image clustering. Additionally, specifying the number of clusters is always a difficult but critical problem for a certain class of clustering algorithms. Hence, this work proposes a nonparametric multi-assignment clustering algorithm called multi-assignment Chinese restaurant process (MACRP), which allows the model complexity to grow as more data instances are observed. The proposed algorithm determines the number of clusters from data, so it provides a practical model to process massive data sets. In the proposed algorithm, we devise a novel prior distribution based on the similarity graph to achieve the goal of multi-assignment, and propose a Gibbs sampling algorithm to carry out posterior inference. The implementation in this work uses collapsed Gibbs sampling and compares with several methods. Additionally, previous evaluation metrics used by multi-label classification are inappropriate for MAC, since label information is unavailable. This work further devises an evaluation metric for MAC based on the characteristics of clustering and multi-assignment problems. We conduct experiments on two real data sets, and the experimental results indicate that the proposed method is competitive and outperforms the alternatives on most data sets.

Instance-based learning (IBL) methods predict the class label of a new instance based directly on the distance between the new unlabeled instance and each labeled instance in the training set, without constructing a classification model in the training phase. In this paper, we introduce a novel class-based feature weighting technique, in the context of instance-based distance methods, using the Ant Colony Optimization meta-heuristic. We address three different approaches of instance-based classification: k -Nearest Neighbours, distance-based Nearest Neighbours, and Gaussian Kernel Estimator. We present a multi-archive adaptation of the ACO ℝ algorithm and apply it to the optimization of the key parameter in each IBL algorithm and of the class-based feature weights. We also propose an ensemble of classifiers approach that makes use of the archived populations of the ACO ℝ algorithm. We empirically evaluate the performance of our proposed algorithms on 36 benchmark datasets, and compare them with conventional instance-based classification algorithms, using various parameter settings, as well as with a state-of-the-art coevolutionary algorithm for instance selection and feature weighting for Nearest Neighbours classifiers.

Recruiting prospective students efficiently and effectively is a very important challenge for universities, mainly because of the increasing competition and the relevance of enrollment-generated revenues. This work provides an intelligent system for modeling the student enrollment decisions problem. A nested logit classifier was constructed to predict which prospective students will eventually enroll in different Bachelor degree programs of a small-sized, private Chilean university. Feature selection is performed to identify the key features that influence the student decisions, such as socio-demographic variables (gender, age, school type, among others), admission efforts, and admission test results. Our results suggest that on-campus activities are far more productive than career fairs and other efforts performed off campus, demonstrating the importance of bringing prospective students to the university. Furthermore, variables such as gender, school type, and declared university and Bachelor degree program preferences are shown to be relevant in successfully modeling the student’s choice of university.

A Deep Belief Network (DBN) is a generative probabilistic graphical model that contains many layers of hidden variables and has excelled among deep learning approaches. DBN can extract suitable features, but improving these networks for obtaining features with more discrimination ability is an important issue. One of the important improvements is sparsity in hidden units. In sparse representation, we have the property that learned features can be interpreted, i.e., correspond to meaningful aspects of input, and are more efficient. One of the main problems in sparsity techniques is to find the best hyper-parameters values which need dozens of experiments to obtain them. In this paper, a dynamic hyper-parameter value setting is proposed for resolving this problem. This proposed method does not need to set parameters manually. According to the results, our new dynamic method achieves acceptable recognition accuracy on test sets in different applications, including image, speech and text. According to these experiments, the proposed method can find hyper-parameters dynamically without losing much accuracy.

Two approaches to data mining with association rules are compared – the apriori algorithm and the ASSOC procedure. The first one was developed for market basket analysis at the beginning of 1990s. An association rule is understood as an implication between conjunctions of attribute-value pairs. The ASSOC procedure is an implementation of the GUHA method of mechanizing hypothesis formation developed since the 1960s. ASSOC deals with association rules – general relations of two general Boolean attributes. Arules – a computational environment for mining association rules based on apriori and the 4ft-Miner procedure – an implementation of the ASSOC procedure are discussed and compared. It is shown that the arules approach to missing information does not correspond to Kleene’s approach and this can lead to a large number of misleading rules. It is also shown that a secured completion developed for the ASSOC procedure avoids this problem.

The problem of selecting learning algorithms has been studied by the meta-learning community for more than two decades. One of the most important task for the success of a meta-learning system is gathering data about the learning process. This data is used to induce a (meta) model able to map characteristics extracted from different data sets to the performance of learning algorithms on these data sets. These systems are built under the assumption that the data are generated by a stationary distribution, i.e., a learning algorithm will perform similarly for new data from the same problem. However, many applications generate data whose characteristics can change over time. Therefore, a suitable bias at a given time may become inappropriate at another time. Although meta-learning has been used to continuously select a learning algorithm in data streams, data characterization has received less attention in this context. In this study, we provide a set of guidelines to support the proposal of characteristics able to describe non-stationary data over time. This guidance considers both the order of arrival of the examples and the type of variables involved in the base-level learning. In addition, we analyze the influence of characteristics regarding their dependence on data morphology. Experimental results using real data streams showed the effectiveness of the proposed data characterization general scheme to support algorithm selection by meta-learning systems. Moreover, the dependent meta-features provided crucial information for the success of some meta-models.