Intelligent Data Analysis: An International Journal

Pattern mining over data streams is critical to a variety of applications such as understanding and predicting weather phenomena or outdoor surveillance. Most of the current techniques attempt to discover relationships between time-point events but are not practical for discovering dependencies over interval-based events. In this work, we present a new approach to mine dependencies between streams of interval-based events that links two events if they occur in a similar manner, one being often followed by the other one after a certain time interval in the data. The proposed method is robust to temporal variability of events and determines the most appropriate time intervals whose validity is assessed by a χ^2 test. As several intervals may redundantly describe the same dependency, the approach retrieves only the most specific intervals with respect to a dominance relationship over temporal dependencies, and thus avoids the classical problem of pattern flooding in data mining. The TEDDY algorithm, TEmporal Dependency DiscoverY, prunes the search space while guaranteeing the discovery of all valid and significant temporal dependencies. We present empirical results on simulated and real-life data to show the scalability and the robustness of our approach. The dependency relationships define a graph that supports intelligent analysis as illustrated by two case studies: Outdoor surveillance of a building via video camera and motion sensors, and assistance for road deicing operations based on the humidity and temperature measurements at the urban scale. These applications demonstrate the efficiency and the effectiveness of our approach.

In recent years, a plethora of approaches have been proposed to deal with the increasingly challenging task of mining concept-drifting data streams. However, most of these approaches can only be applied to uni-dimensional classification problems where each input instance has to be assigned to a single output class variable. The problem of mining multi-dimensional data streams, which includes multiple output class variables, is largely unexplored and only few streaming multi-dimensional approaches have been recently introduced. In this paper, we propose a novel adaptive method, named Locally Adaptive-MB-MBC (LA-MB-MBC), for mining streaming multi-dimensional data. To this end, we make use of multi-dimensional Bayesian network classifiers (MBCs) as models. Basically, LA-MB-MBC monitors the concept drift over time using the average log-likelihood score and the Page-Hinkley test. Then, if a concept drift is detected, LA-MB-MBC adapts the current MBC network locally around each changed node. An experimental study carried out using synthetic multi-dimensional data streams shows the merits of the proposed method in terms of concept drift detection as well as classification performance.

Various ensemble methods are proposed to aggregate hierarchical clusterings. These methods combine a set of hierarchical clusterings into a single representative clustering with an improved quality. The quality of this representative hierarchical clustering intensively depends on the aggregation operator (aggregator) used in the combination. However, choosing from the large pool of aggregators is a challenging task. To facilitate this task, in this paper, we first introduce aggregator types, triangular norms and averaging operators, and then compile a list of main properties and parametric characteristics of these aggregators. An extra property which is needed in hierarchical clustering combination is also defined. Thereafter, a set of experiments is designed to elect the optimized hierarchy aggregator from the variety of these aggregators. The out coming results from the experiments are proved to be compatible with the previous applications in the field of hierarchical clustering ensembles.

Data mining methods extract knowledge from huge amounts of data. Recently with the explosion of mobile technologies, a new type of data appeared. The resulting databases can be described as spatiotemporal data in which spatial information (e.g., the location of an event) and temporal information (e.g., the date of the event) are included. In this article, we focus on spatiotemporal patterns extraction from this kind of databases. These patterns can be considered as sequences representing changes of events localized in areas and its near surrounding over time. Two algorithms are proposed to tackle this problem: the first one uses \emph{a priori} strategy and the second one is based on pattern-growth approach. We have applied our generic method on two different real datasets related to: 1) pollution of rivers in France; and 2) monitoring of dengue epidemics in New Caledonia. Additionally, experiments on synthetic data have been conducted to measure the performance of the proposed algorithms.

Attribute reduction is a very important task in Rough Set Theory. In this context, in recent years several attribute reduction algorithms based on the discernibility matrix have been proposed. In this paper, we go back to the binary discernibility matrix and reformulate some basic concepts that allow us to build on them an algorithm for computing reducts. The proposed algorithm takes advantage of the binary representation used for the discernibility matrix and depending on the fulfillment of certain pruning properties several candidates can be jumped (pruned). In this way, the number of candidates to be tested is reduced. Additionally, from the binary discernibility matrix, we define a more compact and ordered matrix, which allows reducing even more the amount of candidates to be evaluated, and the cost of verifying each one of them.

The proposed algorithm is able to compute all the reducts of an information table, but it can also be used to find a reduct, or a certain number of them. Experiments over standard datasets show that our algorithm has a good performance.

In the coming years, the share of hybrid electric vehicles is expected to grow significantly in personal transportation. Vehicles that can be charged from the electrical grid, such as plug-in hybrids, could introduce problems for the distribution network, especially if the vehicle adoption is spatially concentrated and the charging happens unmanaged. Thus, where and when the hybrid vehicle adoption occurs is an important question for policy makers and planners.

Currently, promoted by the European Union directives such as INSPIRE and PSI, there is a trend of public sector data being harmonised and opened for free usage across the Europe. The vast amount of information in the various registers of the society has a huge and largely untapped potential for modelling societal and environmental phenomena. To discover correlations and relationships from such large databases, data mining methods can be useful.

In this paper, we utilise a data mining approach to identify relationships from public sector data between hybrid vehicle adoption and small area level socio-demography. Based on the discovered model, we assess how favourable each area type is for the adoption. The approach combines Self-Organizing Map based data compression, a feature selection using Genetic Algorithm and a linear regression model.

Stock price prediction with high accuracy may offer significant opportunities for the investors who make decisions on making profit or having high gains over the stocks in stock markets. In this study, four predictive models have been developed for classification task in predicting the direction of movement in the sessional, daily, weekly, and monthly Istanbul Stock Exchange National (ISEN) 100 Index using five years of data. Multilayer perceptron (MLP), which comprises artificial neural networks(ANN), Logistic Regression (LR), and Bagging of Logistic Regression (BLR) classification techniques are used in the models. During the prediction, four datasets are used and the following factors are taken into account: data of macroeconomic indicators, gold prices, oil prices, foreign exchange prices, stock price indexes in various countries, and the data of the ISEN 100 index for past sessions and prior days, which are used as input variables in the datasets. In connection with that, the most effective factors of these input variables were determined by using some feature selection methods. As a result, prediction performances showed that using reduced datasets consisting of only selected the most important features induced a predictive model of each dataset for classification modelling with a better predictive accuracy than using original datasets. Experimental results showed that prediction performances of the models, which are 64.13%, 63.09%, 81.54%, and 100% for the sessional, daily, weekly, and monthly datasets respectively, were found by MLP significantly better than the other classifiers used in this study.

Eye movements mainly consist of fixations and saccades. The identification of eye fixations plays an important role in the process of eye-movement data research. At present, there is no standard method for identifying eye fixations. In this paper, eye movements are regarded as spatial-temporal trajectories. Hence, we present a spatial-temporal trajectory clustering algorithm for eye fixations identification. The main idea of the algorithm is based on Density-Based Spatial Clustering Algorithm with Noise (DBSCAN), which is commonly used in spatial clustering data. In order to apply DBSCAN to our spatial-temporal clustering data, we modified its original concept and algorithm. In addition, the optimum dispersion threshold (Eps) is derived automatically from the data sets with the aid of the `gap statistic' theory. Using the confusion matrix measurement method, we compared the classification results obtained by our algorithm with four other expert algorithms for eye fixations identification show the proposed algorithm demonstrated an equal or better performance. Also, the robustness of our algorithm to additional noise in Points of Gaze (PoGs) data and changes in sampling rate has been verified.

In frequent pattern mining, items are considered as having the same importance in a database and their occurrence are represented as binary values in transactions. In real-world databases, however, items not only have relative importance but also are represented as non-binary values in transactions. High utility pattern mining is one of the most essential issues in the pattern mining field, which recently emerged to address the limitation of frequent pattern mining. Meanwhile, tree construction with a single database scan is significant since a database scan is a time-consuming task. In utility mining, an additional database scan is necessary to identify actual high utility patterns from candidates. In this paper, we propose a novel tree structure, namely SIQ-Tree (Sum of Item Quantities), which captures database information through a single-pass. Moreover, a restructuring method is suggested with strategies for reducing overestimated utilities. The proposed algorithm can construct the SIQ-Tree with only a single scan and decrease the number of candidate patterns effectively with the reduced overestimation utilities, through which mining performance is improved. Experimental results show that our algorithm outperforms a state-of-the-art one in terms of runtime and the number of generated candidates with a similar memory usage.

In the United States, Obama's presidency administration has moved forward with Open Government Initiative to open public data. Similarly, Tim Berners-Lee, the inventor of the World Wide Web, has led data.gov.uk in England. With the active movement of these global governments, open public data have let citizens know government policies clearly. However, since it is not trivial for people to clearly understand raw public data, active studies have carried out for good visualization. As an example, in 2009, Andrew Odewahn showed how to visualize U.S. Senate social graphs from 1991 until 2009. Given a polling data set in public, a graph consisting of a set of nodes with edges could be created, where each node is a senator and the edge between two nodes represents the similar voting behavior of the senators. However, since such a graph is considerably dense and complex, it is hard to visualize and also figure out the meaning of the graph. Toward this problem, in this paper, we propose novel graph visualization techniques based on friend matching and hierarchical matching. In our empirical study, we applied our visualization algorithms to the voting data of the 18th National Assembly in Korea, Senate bills in U.S., and U.N. voting data. Then, we showed the effectiveness of our proposed methods through user test.

In application fields such as linguistic and computer vision there is an increasing need of reference data for the empirical analysis of new methods and the assessment of different algorithms. Current evaluations are based on few real-life collections or on artificial data generators built on models that are too simplistic to cover real scenarios and to allow researchers to identify crucial limitations of their algorithms. We propose a flexible approach to generate high-dimensional vectors, with directional properties controlled by the distribution of their pair-wise cosine distances. The generation method is formulated as a non-linear continuous optimization problem, which is solved with a computationally efficient local search algorithm. We show with an empirical study that our approach can create large high-dimensional data collections with desired properties in reasonable time.

Entity Resolution (ER) is the process of identifying references referring to the same entity from one or more data sources. In the ER process, most existing approaches exploit the content information of references, categorized as content-based ER, or additionally consider linkage information among references, categorized as context-based ER. However, in new applications of ER, such as in the genealogical domain, the very limited linkage information among references results in a disjoint graph in which the existing content-/context-based ER techniques have very limited applicability. Therefore, in this paper we propose first, to use the homophily principle for augmentation of the original input graph by connecting the potential similar references, and second, to use a Random Walk based approach to consider contextual information available for each reference in the augmented graph. We evaluate the proposed method by applying it to a large genealogical dataset and we succeed to predict 420,000 reference matches with precision 92% and discover six novel and informative patterns among them which can not be detected in the original disjoint graph.