Future Computing and Informatics Journal最新文献

Data mining is one of the vital techniques that could be applied in different fields such as medical, educational and industrial fields. Extracting patterns from spatial data is very useful to be used for discovering the trends in the data. However, analyzing spatial data is exhaustive due to its details as it is related to locations with a special representation such as longitude and latitude. This paper aims at proposing an approach for applying data mining techniques over spatial data to find trends in the data for decision support. Basic information considering spatial data is presented with presenting the proposed approach aiming to be applied in the Egyptian organizations to prove its applicability.

Spam emails are not necessary, though they are harmful as they include viruses and spyware, so there is an emerging need for detecting spam emails. Several methods for detecting spam emails were suggested based on the methods of machine learning, which were submitted to reduce non-relevant emails and get results of high precision for spam email classification. In this work, a new predictive method is submitted based on antlion optimization (ALO) and boosting termed as ALO-Boosting for solving spam emails problem. ALO is a computational model imitates the preying technicality of antlions to ants in the life cycle. Where ALO was utilized to modify the actual place of the population in the separate seeking area, thus obtaining the optimum feature subset for the better classification submit based on boosting classifier. Boosting classifier is a classification algorithm that points to a group of algorithms which modifies soft learners into powerful learners. The proposed procedure is compared against support vector machine (SVM), k-nearest neighbours algorithm (KNN), and bootstrap aggregating (Bagging) on spam email datasets in a set of implementation measures. The experimental outcomes show the ability of the proposed method to successfully detect optimum features with the smallest value of selected features and a high precision of measures for spam email classification based on boosting classifier.

This paper studies the advances in time series forecasting models using artificial neural network methodologies in a systematic literature review. The systematic review has been done using a manual search of the published papers in the last 11 years (2006–2016) for the time series forecasting using new neural network models and the used methods are displayed. In the covered period in the study, the results obtained found 17 studies that meet all the requirements of the search criteria. Only three of the obtained proposals considered a process different to the autoregressive of a neural networks model. These results conclude that, although there are many studies that presented the application of neural network models, but few of them proposed new neural networks models for forecasting that considered theoretical support and a systematic procedure in the construction of model. This leads to the importance of formulating new models of neural networks.

Now, the web pages contain opinions on almost anything, at review sites, forums, discussion groups, and blogs which called user generated content. They contain valuable information for different users such as persons or organizations, the processes of collecting, analyzing and classifying them to positive or negative opinions in addition to summarizing the opinions are considered a very important research issue. Summarizing opinions helps users to explore the opinion of others about the key aspects of a topic or an entity. The proposed opinion summarization system receives a document that contains sentences expressing opinions about an entity and generates a summary considering the important aspects, their relations, their sentiments and the textual evidences, as expressed in the reviews. In this paper we present a linguistic approach to summarize the opinionated documents across different domains, our evaluation based on a dataset of hotels, cars and various products reviews. The reviews collected from Tripadvisor, Amazon and Edmunds, each review document consist of a set of unordered, redundant reviews sentence, there are approximately 100 sentences per review document. The summary depends on the type of the opinion which is direct, comparative, or superlative. Each type is assigned to a specialist who is responsible for the summary.

The Publisher regrets that this article is an accidental duplication of an article that has already been published in <FCIJ, 2 (2017) 39 - 47>, http://dx.doi.org/10.1016/j.fcij.2017.05.001. The duplicate article has therefore been withdrawn.

The full Elsevier Policy on Article Withdrawal can be found at https://www.elsevier.com/about/our-business/policies/article-withdrawal

Lately, there is a growing interest in conducting research on optimization algorithms due to their wide range of engineering applications. One of the optimization algorithms' categories is evolutionary algorithms which are inspired from the natural behavior of animals and humans. Further, each of the evolutionary algorithms has its own advantages and disadvantages in convergence accuracy and computational time. In the present paper, a novel solution search algorithm taken from the team games is introduced. This evolutionary algorithm named Team Game Algorithm (TGA) involves passing a ball, making mistakes and substitution operators. Comparing the TGA's results to the outcomes of other well-known algorithms for unimodal and multimodal test functions elucidates the successful design of the proposed heuristic algorithm.

Image forensics is an important area of research used to indicate if a particular image is original or subjected to any kind of tampering. Images are essential part of judgment in tribunals. For forensic analysis, image forgery-detection techniques used to identify the forged images. In this paper, an effective algorithm to indicate Copy Move Forgery in digital image presented. The Scale Invariant Feature Transform (SIFT) and Fuzzy C-means (FCM) for clustering are utilized in the proposed algorithm. A number of numerical experiments performed using the MICC-220 dataset. The authors created an additional dataset, which consisted of 353 color images. The proposed algorithm tested by using both datasets where the average detection time on the MICC-220 data set is reduced by 14.67% over the existing traditional SIFT-based algorithm. For the created dataset, the average detection time reduced by 15.91% over the existing traditional SIFT-based algorithm.

High code coverage is measured by the process of software testing typically using automatic test case generation tools. This standard approach is usually used for unit testing to improve software reliability. Most automated test case generation tools focused just on code coverage without considering its cost and redundancy between generated test cases. To obtain optimized high code coverage and to ensure minimum cost and redundancy a Multi-Objectives Evolutionary Algorithm approach (MOEA) is set in motion. An efficient approach is proposed and applied to different algorithms from MOEA Frame from the separate library with three fitness functions for Coverage, Cost, and Redundancy. Four MEOA algorithms have been proven reliable to reach above the 90 percent code coverage: NSGAII, Random, SMSEMOA,v and ε-MOEA. These four algorithms are the key factors behind the MOEA approach.

Data clustering is one of the fundamental techniques in scientific analysis and data mining, which describes a dataset according to similarities among its objects. Partition based clustering algorithms are the most popular and widely used clustering technique. In this information era, due to the digitization of every field, the huge volume of data is available to data analysts. The quick growth of such datasets makes decade old computing platforms, programming paradigms, and clustering algorithms become inadequate to obtain knowledge from these datasets. To cluster such large datasets, Hadoop distributed platform, MapReduce programming paradigm and modified clustering algorithms are being used to shrink the computational time by distributing clustering job across multiple computing nodes. This paper provides a comprehensive review of Hadoop and MapReduce and their components. This paper aims to survey recent research works on partition based clustering algorithms which use MapReduce as their programming paradigm. In many recent works, the traditional partition based clustering algorithms like K-means, K-prototypes, K-medoids, K-modes and Fuzzy C-means are modified for MapReduce paradigm in order to obtain different clustering objectives on different datasets for reducing the computational time. The contribution of this paper is (1) to provide an overview of clustering challenges in real world large dataset clustering and the role of MapReduce programming paradigm and its supporting platforms in dealing the challenges for several tasks in different datasets and (2) to review recent works in partition based clustering using MapReduce paradigm for different clustering objectives for different datasets employing different strategies.