Search

Fragmentation, allocation and replication are techniques widely used in relational databases to improve the performance of operations and reduce their cost in distributed environments. This article shows an analysis of different methods for database fragmentation, allocation and replication and a Web application called FRAGMENT that adopts the work technique that was selected in the analysis stage, because it presents a fragmentation and replication method, it is applied to a cloud environment, it is easy to implement, it focuses on improving the performance of the operations executed on the database, it shows everything necessary for its implementation and is based on a cost model. FRAGMENT analyzes the operations performed in any table of a database, proposes fragmentation schemes based on the most expensive attributes and allocates and replicates a scheme chosen by the user in a distributed environment in the cloud. This work shows a common problem in fragmentation methods, overlapping fragments, and provides an algorithm with an approach to address it. This algorithm results in the predicates that will define each fragment in a distributed environment. To validate the implemented technique, a second web application is presented, dedicated to simulate operations on sites and focused on producing a log file for the main application. Experiments with the TPC-E benchmark demonstrated lower response time of the queries executed against the distributed database generated by FRAGMENT compared with a centralized database.

Vertical fragmentation is a distributed database design technique widely used in relational databases to reduce query execution costs. This technique has been also applied in multimedia databases to take advantage of its benefits in query optimization. Content-based retrieval is essential in multimedia databases to provide relevant data to the user. This paper presents a review of the literature on vertical fragmentation methods to determine if they consider multimedia data and content-based queries, are easy to implement, are complete (i.e., the paper includes all the information necessary to implement the method), and if they provide a cost model for the evaluation of their results. To meet this objective, we analyzed and classified 37 papers referring to vertical fragmentation and/or content-based image retrieval. Finally, we selected the best method from our comparative analysis and presented a web application architecture for applying vertical fragmentation in multimedia databases to optimize content-based queries.

Fragmentation plays a very important role in databases since this achieves an adequate design in distributed environments such as the cloud, which in terms of execution cost of the read and write operations, benefits are obtained. This work presents an improvement to the design of the overlapping fragmentation algorithm shown in previous works, which is part of a method to also carry out allocation and replication. The improvement consists of ordering the predicates by cost and in this way obtaining each predicate contemplating that the fragment overlap will occur, however, the fragments with a higher cost will be kept more intact than those with a lower cost. Experiments with the TPC-E benchmark adapted to a distributed environment demonstrate that our enhanced approach achieves a significant reduction of query response time.

Fragmentation is a design technique widely used in multimedia databases, because it produces substantial benefits in reducing response times, causing lower execution costs in each operation performed. Multimedia databases include data whose main characteristic is their large size, therefore, database administrators face a challenge of great importance, since they must contemplate the different qualities of non-trivial data. These databases over time undergo changes in their access patterns. Different fragmentation techniques presented in related studies show adequate workflows, however, some do not contemplate changes in access patterns. This paper aims to provide an in-depth review of the literature related to dynamic fragmentation of multimedia databases, to identify the main challenges, technologies employed, types of fragmentation used, and characteristics of the cost model. This review provides valuable information for database administrators by showing essential characteristics to perform proper fragmentation and to improve the performance of fragmentation schemes. The reduction of costs in fragmentation methods is one of the most desired main properties. To fulfill this objective, the works include cost models, covering different qualities. In this analysis, a set of characteristics used in the cost models of each work is presented to facilitate the creation of a new cost model including the most used qualities. In addition, different data sets or reference points used in the testing stage of each work analyzed are presented.

Cloud computing is a revolutionary technology for businesses, governments, and citizens. Some examples of Software-as-a-Services (SaaS) of cloud computing are banking apps, e-mail, blog, online news, and social networks. In this chapter, we analyze data sets generated by trending topics on Twitter that emerged from Mexican citizens that interacted during the earthquake of September 19, 2017, using sentiment analysis and supervised learning, based on the Ekman’s six emotional model. We built three classifiers to determine the emotions of tweets that belong to the same topic. The classifiers with the best accuracy for predicting emotions were Naive Bayes and support vector machine. We found that the most frequent predicted emotions were happiness, anger, and sadness; also, that 6.5% of predicted tweets were irrelevant. We provide some recommendations about the use of machine learning techniques in sentiment analysis. Our contribution is the expansion of the emotions range, from three (negative, neutral, positive) to six in order to provide more elements to understand how users interact with social media platforms. Future research will include validation of the method with different data sets and emotions, and the addition of new artificial intelligence techniques to improve accuracy.

Currently, vertical partitioning has been used in multimedia databases in order to take advantage of its potential benefits in query optimization. Nevertheless, most vertical partitioning algorithms are static; this means that they optimize a vertical partitioning scheme (VPS) according to a workload, but if this workload suffers changes, the VPS may be degraded, which would result in long query response time. This paper presents a set of active rules to perform dynamic vertical partitioning in multimedia databases. First of all, these rules collect all the information that a vertical partitioning algorithm needs as input. Then, they evaluate this information in order to know if the database has experienced enough changes to trigger a performance evaluator. In this case, if the performance of the database falls below a threshold previously calculated by the rules, the vertical partitioning algorithm is triggered, which gets a new VPS. Finally, the rules materialize the new VPS. Our active rule base is implemented in the system DYMOND, which is an active rule-based system for dynamic vertical partitioning of multimedia databases. DYMOND's architecture and workflow are presented in this paper. Moreover, a case study is used to clarify and evaluate the functionality of our active rule base. Additionally, authors of this paper performed a qualitative evaluation with the aim of comparing and evaluating DYMOND's functionality. The results showed that DYMOND improved query performance in multimedia databases.

Graphical abstractDisplay Omitted HighlightsThis paper describes the development of an algorithm for training large data sets.The algorithm uses a first stage of SVM with a small data set.The algorithm uses decision trees to find best data points in the entire data set.DT is trained using SV and non-SV found in the first SVM stage.In the second SVM stage the training data represent all data points found by the DT. Support Vector Machine (SVM) has important properties such as a strong mathematical background and a better generalization capability with respect to other classification methods. On the other hand, the major drawback of SVM occurs in its training phase, which is computationally expensive and highly dependent on the size of input data set. In this study, a new algorithm to speed up the training time of SVM is presented; this method selects a small and representative amount of data from data sets to improve training time of SVM. The novel method uses an induction tree to reduce the training data set for SVM, producing a very fast and high-accuracy algorithm. According to the results, the proposed algorithm produces results with similar accuracy and in a faster way than the current SVM implementations.

Training a support vector machine (SVM) with data number n has time complexity between O ( n 2 ) and O ( n 3 ) . Most training algorithms for SVM are not suitable for large datasets. Decision trees can simplify SVM training, however classification accuracy becomes lower when there are inseparable points. This paper introduces a novel method for SVM classification. A decision tree is used to detect the low entropy regions in input space, and Fisher’s linear discriminant is applied to detect the data near to support vectors. The experimental results demonstrate that our approach has good classification accuracy and low standard deviation, the training is significantly faster than the other existing methods.

Univariate decision trees are classifiers currently used in many data mining applications. This classifier discovers partitions in the input space via hyperplanes that are orthogonal to the axes of attributes, producing a model that can be understood by human experts. One disadvantage of univariate decision trees is that they produce complex and inaccurate models when decision boundaries are not orthogonal to axes. In this paper we introduce the Fisher's Tree, it is a classifier that takes advantage of dimensionality reduction of Fisher's linear discriminant and uses the decomposition strategy of decision trees, to come up with an oblique decision tree. Our proposal generates an artificial attribute that is used to split the data in a recursive way. The Fisher's decision tree induces oblique trees whose accuracy, size, number of leaves and training time are competitive with respect to other decision trees reported in the literature. We use more than ten public available data sets to demonstrate the effectiveness of our method.

Plant identification belongs to a specific application domain of data mining. Images of plant leaves are usually used as the main element to distinguish a plant from another. For proper identification, feature extraction is necessary. In the literature, most plant recognition systems use the features along with a classification method, which has been adapted or modified to face this type of application. In this paper, we propose three new geometric features that describe the vertical and horizontal symmetry of leaves. These features are simple to extract from images. According to the results of experiments, when these features are used in conjunction with other well-known geometric characteristics, the performance of classical classification methods is remarkably improved. To show the effectiveness of the proposal, we test seven classifiers with images of leaves publicly available on the Internet.

The performance of classification methods is notably damaged with imbalanced data sets. Although some studies to analyze this behavior have realized before, most of the conclusions obtained from experiments correspond to synthetic data sets. In this paper, we study the relationship between the performance of five classification methods and neighbors of minority class instances. According to the results of experiments, we found strong empiric evidence that the type of neighborhoods of minority class instances affect classification accuracy. Indeed, we observe that the type of neighborhood is more important than the imbalance rate. In order to validate the results, we use ten real-world imbalanced data sets, and measure AUC ROC and True Positive Rates.

Classification methods usually exhibit a poor performance when they are applied on imbalanced data sets. In order to overcome this problem, some algorithms have been proposed in the last decade. Most of them generate synthetic instances in order to balance data sets, regardless the classification algorithm. These methods work reasonably well in most cases; however, they tend to cause over-fitting.

Over the past few years, has been shown that generalization power of Support Vector Machines (SVM) falls dramatically on imbalanced data-sets. In this paper, we propose a new method to improve accuracy of SVM on imbalanced data-sets. To get this outcome, firstly, we used undersampling and SVM to obtain the initial SVs and a sketch of the hyperplane. These support vectors help to generate new artificial instances, which will take part as the initial population of a genetic algorithm. The genetic algorithm improves the population in artificial instances from one generation to another and eliminates instances that produce noise in the hyperplane. Finally, the generated and evolved data were included in the original data-set for minimizing the imbalance and improving the generalization ability of the SVM on skewed data-sets.

Support Vector Machine (SVM) is an important classification method used in a many areas. The training of SVM is almost O(n^{2}) in time and space. Some methods to reduce the training complexity have been proposed in last years. Data selection methods for SVM select most important examples from training data sets to improve its training time. This paper introduces a novel data reduction method that works detecting clusters and then selects some examples from them. Different from other state of the art algorithms, the novel method uses a decision tree to form partitions that are treated as clusters, and then executes a guided random selection of examples. The clusters discovered by a decision tree can be linearly separable, taking advantage of the Eidelheit separation theorem, it is possible to reduce the size of training sets by carefully selecting examples from training sets. The novel method was compared with LibSVM using public available data sets, experiments demonstrate an important reduction of the size of training sets whereas showing only a slight decreasing in the accuracy of classifier.

In pattern recognition and data mining a data set is named skewed or imbalanced if it contains a large number of objects of certain type and a very small number of objects of the opposite type. The imbalance in data sets represents a challenging problem for most classification methods, this is because the generalization power achieved for classic classifiers is not good for skewed data sets. Many real data sets are imbalanced, so the development of new methods to face this problem is necessary. The SVM classifier has an exceptional performance for data sets that are not skewed, however for imbalanced sets the optimal separating hyper plane is not enough to achieve acceptable results. In this paper a novel method that improves the performance of SVM for skewed data sets is presented. The proposed method works by exciting the support vectors and displacing the separating hyper plane towards majority class. According to the results obtained in experiments with different skewed data sets, the method enhances not only the accuracy but also the sensitivity of SVM classifier on this kind of data sets.

Border points are those instances located at the outer margin of dense clusters of samples. The detection is important in many areas such as data mining, image processing, robotics, geographic information systems and pattern recognition. In this paper we propose a novel method to detect border samples. The proposed method makes use of a discretization and works on partitions of the set of points. Then the border samples are detected by applying an algorithm similar to the presented in reference [8] on the sides of convex hulls. We apply the novel algorithm on classification task of data mining; experimental results show the effectiveness of our method.