Статьи журнала - International Journal of Information Technology and Computer Science
Все статьи: 1195
Статья научная
Authors here tried to use the WEKA tool to evaluate the performance of various classifiers on a dataset to come out with the optimum classifier, for a particular application. A Classifier is an important part of any machine learning application. It is required to classify various classes and get to know whether the predicted class lies in the true class. There are various performance analysis measures to judge the efficiency of a classifier and there are many tools which provide oodles of classifiers. In the present investigation, Bayes Net, Naive Bayes and their combination have been implemented using WEKA. It has been concluded that the combination of Bayes Net and Naive Bayes provides the maximum classification efficiency out of these three classifiers. Such a hybridization approach will always motivate for combining different classifiers to get the best results.
Бесплатно
Comparative analysis of multiple sequence alignment tools
Статья научная
The perfect alignment between three or more sequences of Protein, RNA or DNA is a very difficult task in bioinformatics. There are many techniques for alignment multiple sequences. Many techniques maximize speed and do not concern with the accuracy of the resulting alignment. Likewise, many techniques maximize accuracy and do not concern with the speed. Reducing memory and execution time requirements and increasing the accuracy of multiple sequence alignment on large-scale datasets are the vital goal of any technique. The paper introduces the comparative analysis of the most well-known programs (CLUSTAL-OMEGA, MAFFT, BROBCONS, KALIGN, RETALIGN, and MUSCLE). For programs’ testing and evaluating, benchmark protein datasets are used. Both the execution time and alignment quality are two important metrics. The obtained results show that no single MSA tool can always achieve the best alignment for all datasets.
Бесплатно
Comparative descriptive analysis of breast cancer tissues using k-means and self-organizing map
Статья научная
Data mining is a descriptive and predictive data analytical technique that discovers meaningful and useful knowledge from dataset. Clustering is one of the descriptive analytic techniques of data mining that uses latent statistical information that exists among dataset to group them into meaningful and or useful groups. In clinical decision making, information from medical tests coupled with patients’ medical history is used to make recommendations, and predictions. However, these voluminous medical datasets analysis is always dependent of individual analyzer that might have in one way or the other introduced human error. In other to solve this problem, many automated analyses have been proposed by researchers using various machine learning techniques and various forms of dataset. In this paper, dataset from electrical impedance imaging of breast tissues are clustered using two unsupervised algorithms (k-means and self-organizing map). Result of the performances of these machine learning algorithms as implemented with R i368 version 3.4.2 shows a slight outperformance of K-means in terms of classification accuracy over self-organizing map for the considered dataset.
Бесплатно
Статья научная
Since various pseudo-random algorithms and sequences are used for cryptography of data or as initial values for starting a secure communication, how these algorithms are analyzed and selected is very important. In fact, given the growingly extensive types of pseudo-random sequences and block and stream cipher algorithms, selection of an appropriate algorithm needs an accurate and thorough investigation. Also, in order to generate a pseudo-random sequence and generalize it to a cryptographer algorithm, a comprehensive and regular framework is needed, so that we are enabled to evaluate the presented algorithm as quick as possible. The purpose of this study is to use a number of pseudo-random number generators as well as popular cryptography algorithms, analyze them in a standard framework and observe the results obtained in each stage. The investigations are like a match between different algorithms, such that in each stage, weak algorithms are eliminated using a standard method and successful algorithms enter the next stage so that the best algorithms are chosen in the final stage. The main purpose of this paper is to certify the approved algorithm.
Бесплатно
Comparing the Performances of Ensemble-classifiers to Detect Eye State
Статья научная
Brain signals required for the brain-computer interface are obtained through the electroencephalography (EEG) method. EEG data is used in the analysis of many problems such as epileptic seizure detection, bipolar mood disorder, attention deficit, and detection of the sleep state of the vehicle driver. It is very important to determine whether the eye is open or closed, which is a substantial organ for the determination of the cognitive state of the person. The aim of this paper is to present a stable and successful model for detecting the eye states that are opened or closed. In this context, the performances of several ensemble classifiers were examined on the Emotiv EEG Neuroheadset dataset, which has 14 features excluding the target variable, 14980 records that have 8225 eye states opened and 6755 eye states closed. In the experiments, firstly the min-max normalization process was applied to the dataset, and then the classification performances of these classifiers were evaluated via a 5-fold cross-validation technique. The performance of each model was measured using accuracy, sensitivity, and specificity metrics. The obtained results show that the Random Forest algorithm is an acceptable level with 92.61% value of accuracy, 94.31% value of sensitivity and 91.36% value of specificity for detecting the eye state.
Бесплатно
Comparison of Time Concept Modeling for Querying Temporal Information in OWL and RDF
Статья научная
Ontology is an important factor in the integration of heterogeneous semantic information. Description logic, as a formal language for expressing ontologies, does not include the necessary features to create a temporal dimension in the relationships among concepts. It is critical to introduce time concepts to model temporal data and relate them to other non-temporal data recorded in ontology. Current query languages in the semantic web are not able to respond to temporal questions; thus, another important issue is to have the appropriate methods for answering temporal questions. In this paper, temporal modeling methods in OWL and RDF are assessed and the temporal query languages for expressing queries in the semantic web are categorized and compared.
Бесплатно
Comparison of on Demand Routing Protocols
Статья научная
A routing protocol is used to facilitate communication in ad hoc network. The primary goal of such a routing protocol is to provide an efficient and reliable path between a pair of nodes. The routing protocols for ad hoc network can be categorized into three categories: table driven, on demand and hybrid routing. The table driven and hybrid routing strategies require periodic exchange of hello messages between nodes of the ad hoc network and thus have high processing and bandwidth requirements. On the other hand on demand routing strategy creates routes when required and hence is very much suitable for ad hoc network. This paper therefore examines the performance of three on demand routing protocols at application layer using QualNet-5.01 simulator.
Бесплатно
Статья научная
Informal education will be successful as an alternative for the community because not all people are able to receive formal education. This study uses a qualitative method with a systematic literature review (SLR) technique to look for learning community components in informal education to support learning in the culinary community in the new normal era of Covid-19. The author collects, studies, and analyzes reference sources according to the specified keywords. Found 53 papers from 2002 to 2021 with background authors from academia, industry, and the public sector with reference sources from journals, conferences, white papers, and research reports. Systematic literature review results obtained 6 components of learning community in informal education, namely content, forum, method, technology, figure/layout, and human/social resources. The six components as a reference and the author's first step in the next research through searching for the characteristics of the learning community in the culinary field, then making a learning model of the culinary community. Because of the importance of the learning community component in informal education to help community members share knowledge, solve problems, share common goals and interests among community members.
Бесплатно
Comprehensive Evaluation of Energy Transportation Corridor in China
Статья научная
In order to offer reference for the energy transportation channels in future, we evaluated the existing system with the information entropy theory. By calculating the base value for the scale of construction and the equilibrium degree with the data of annual investment in the fixed assets from 1986 to 2010, we measured the scale and its rationality, and analyzed the relation between transportation channels and the information entropy combined with the maximum entropy methods. Empirical results show that the partial information entropy Si still have certain rising space. We can improve the rationality of the energy transportation channels by improving the partial information entropy Si. And we should not let the rail transportation channels have excessive investment growth, and then secondly transfer some of the highway investment to the transportation channels of waterway or pipeline while finally advance the construction step of the pipeline transportation channels. This conclusion provides a quantitative basis for measuring the rationality of the construction scale of the energy transportation channels. The quantitative evaluation of the construction scale of energy transportation channels is a important supplement to the traditional research and provide evidence for further development. And we can obtain more suggestions on the development strategy emphasis in future.
Бесплатно
Статья научная
A Mobile Adhoc Network is a multi-hop self-configuring network without any fixed infrastructure. Due to mobility of nodes, dynamic topology and highly dynamic environment, designing and implementing stable routing in Mobile Ad-hoc Networking is a major challenge and a critical issue. This paper analyses the performance analysis of on demand routing protocol, Dynamic Source Routing (DSR), Adhoc on Demand Distance Vector Routing (AODV) and table driven protocol, Destination-Sequenced Distance Vectoring (DSDV) using a network simulator NS2. Different types of test scenario have designed with fixed number of nodes but varying mobility. Different performance metric values like, throughput, delay, normalized network load, end to end delay, dropped packets, packets delivery ratio have been observed. The experimental results have been analysed and recommendation based on the obtained results has been proposed about the significance of each protocol in different scenarios and situations. The simulation results show that both protocols are good in performance in their own categories. We believe that findings of this paper will help the researcher to find the best protocol under predefined condition with varied mobility. We believe that this research will help the researcher to identify and further investigate any particular metrics value of AODV, DSR and DSDV.
Бесплатно
Статья научная
The paper is presenting theoretical analytical model and computer simulation of electromagnetic transient process in a transformer. Transformer parameters in a selected transitional process have been analyzed. Theoretical model refers to an energetic transformer with concentrated parameters with consideration of parameters of mutual inductance M. Simulation was performed on a personal computer using the software program MATLAB SIMULINK. The computer simulation confirmed the possibility of transitional process analysis in transformer’s windings with concentrated parameters.
Бесплатно
Confidence Analysis of a Solo Sign-On Device for Distributed Computer Networks
Статья научная
Solo sign-on (SSO) is a new authentication mechanism that enables a legal user with a single credential to be authenticated by multiple service providers in a distributed computer network. Recently, a SSO scheme proposed and claimed its security by providing well organized security arguments. But their scheme is actually insecure as it fails to meet credential privacy and soundness of authentication. Specifically, we present two impersonation attacks i.e., credential recovering attack and impersonation attack without credentials. So we propose a more authentication scheme that overcomes these attacks and flaws by make use of efficient verifiable encryption of RSA signatures. We promote the formal study of the soundness of authentication as one open problem.
Бесплатно
Congestion Aware Multipath Routing: Aggregation Network Applicability and IPv6 Implementation
Статья научная
Currently, the service provider network capacity utilization is low due to the shortest multipath based routing protocols Opens Shortest Path First (OSPF) and Intermediate System-to-Intermediate System (ISIS). Due to inefficient routing approach, certain paths can be overloaded and link capacity is required while alternative paths are unused. The overall network has to be dimensioned with higher link bandwidth requirements introducing additional line, linecard, routing engine and overall solution cos. This paper provides improved Congestion Aware Multipath Routing (CAMRv2) algorithm overview. The new network routing algorithm allows higher throughput, network load-balancing and stability to ensure lower congestion and data drop on critical links. The algorithm discovers unused network resources and dynamically adapts to the actual traffic load and displacement. The focus in this paper is on new parameters for path computation performance improvement. Additionally, detailed IPv6 source routing CAMRv2 implementation for parallel coexistence with present networks is presented. Finally, the new routing algorithm is simulated in several scenarios over aggregation network. The result of simulations have proved better performance and resource utilization of the proposed algorithm in sparse aggregation network in terms of load-balancing between uplinks to the core network.
Бесплатно
Conservativity Principle Violations for Ontology Alignment: Survey and Trends
Статья научная
Ontology matching techniques are a solution to overcome the problem of interoperability between ontologies. However, the generated mappings suffer from logical defects that influence their usefulness. In this paper we present a detailed analysis of the problem so-called conservativity principle; alignment between ontologies should never generate new knowledge compared to those generated by reasoning solely on ontologies. We also study the sub-problems; Ontology change and Satisfiability preservation problems and compare the related works and their way to detect and repair conservativity principle. At the end we present a set of open research issues.
Бесплатно
Статья научная
The paper presents a method and tools for consistency checking in UML design of an object-oriented software system. The proposed method uses graph representation of UML diagrams and first-order predicate logic to specify consistency rules mostly on the cross-diagram level. Classification of consistency rules is presented. Two approaches to implementation of con-sistency checking are discussed and compared.
Бесплатно
Construction of High-accuracy Ensemble of Classifiers
Статья научная
There have been several methods developed to construct ensembles. Some of these methods, such as Bagging and Boosting are meta-learners, i.e. they can be applied to any base classifier. The combination of methods should be selected in order that classifiers cover each other weaknesses. In ensemble, the output of several classifiers is used only when they disagree on some inputs. The degree of disagreement is called diversity of the ensemble. Another factor that plays a significant role in performing an ensemble is accuracy of the basic classifiers. It can be said that all the procedures of constructing ensembles seek to achieve a balance between these two parameters, and successful methods can reach a better balance. The diversity of the members of an ensemble is known as an important factor in determining its generalization error. In this paper, we present a new approach for generating ensembles. The proposed approach uses Bagging and Boosting as the generators of base classifiers. Subsequently, the classifiers are partitioned by means of a clustering algorithm. We introduce a selection phase for construction the final ensemble and three different selection methods are proposed for applying in this phase. In the first proposed selection method, a classifier is selected randomly from each cluster. The second method selects the most accurate classifier from each cluster and the third one selects the nearest classifier to the center of each cluster to construct the final ensemble. The results of the experiments on well-known datasets demonstrate the strength of our proposed approach, especially applying the selection of the most accurate classifiers from clusters and employing Bagging generator.
Бесплатно
Construction of Strength Two Mixed Covering Arrays Using Greedy Mutation in Genetic Algorithm
Статья научная
Metaheuristic methods are capable of solving a wide range of combinatorial problems competently. Genetic algorithm (GA) is a metaheuristic search based optimization algorithm that can be used to generate optimal Covering Arrays (CAs) and Mixed Covering Arrays (MCAs) for pair-wise testing. Our focus in the work presented in this paper is on the strategies of performing mutation in GA to enhance the overall performance of GA in terms of solution quality and computational time (number of generations). This is achieved by applying a greedy approach to perform mutation at a position that minimizes the loss of existing distinct pairs in the parent CA/MCA and ensures that the generated offspring is of good quality. Experiments are conducted on several benchmark problems to evaluate the performance of the proposed greedy based GA with respect to the existing state-of-the-art algorithms. Our evaluation shows that the proposed algorithm outperforms its GA counterpart by generating better quality MCA in lesser number of generations. Also the proposed approach yields better/comparable results compared to the existing state-of-the-art algorithms for generating CAs and MCAs.
Бесплатно
Content Based Image Recognition by Information Fusion with Multiview Features
Статья научная
Substantial research interest has been observed in the field of object recognition as a vital component for modern intelligent systems. Content based image classification and retrieval have been considered as two popular techniques for identifying the object of interest. Feature extraction has played the pivotal role towards successful implementation of the aforesaid techniques. The paper has presented two novel techniques of feature extraction from diverse image categories both in spatial domain and in frequency domain. The multi view features from the image categories were evaluated for classification and retrieval performances by means of a fusion based recognition architecture. The experimentation was carried out with four different popular public datasets. The proposed fusion framework has exhibited an average increase of 24.71% and 20.78% in precision rates for classification and retrieval respectively, when compared to state-of-the art techniques. The experimental findings were validated with a paired t test for statistical significance.
Бесплатно
Статья научная
Content-based image retrieval (CBIR) is the process of retrieving similar images of a query image from a source of images based on the image contents. In this paper, color and texture features are used to represent image contents. Color layout descriptor (CLD) and gray-level co-occurrence matrix (GLCM) are used as color and texture features respectively. CLD and GLCM are efficient for representing images with local dominant regions. For retrieving similar images of a query image, the features of the query image is matched with that of the images of the source. We use cityblock distance for this feature matching purpose. K-nearest images using cityblock distance are the similar images of a query image. Our CBIR approach is scale invariant as CLD is scale invariant. Another set of features, GLCM defines color patterns. It makes the system efficient for retrieving similar images based on spatial relationships between colors. We also measure the efficiency of our approach using k-nearest neighbors algorithm. Performance of our proposed method, in terms of precision and recall, is promising and better, compared to some recent related works.
Бесплатно
Content-based Fish Classification Using Combination of Machine Learning Methods
Статья научная
Fish species recognition is an increasing demand to the field of fish ecology, fishing industry sector, fisheries survey applications, and other related concerns. Traditionally, concept-based fish specifies identification procedure is used. But it has some limitations. Content-based classification overcomes these problems. In this paper, a content-based fish recognition system based on the fusion of local features and global feature is proposed. For local features extraction from fish image, Local Binary Pattern (LBP), Speeded-Up Robust Feature (SURF), and Scale Invariant Feature Transform (SIFT) are used. To extract global feature from fish image, Color Coherence Vector (CCV) is used. Five popular machine learning models such as: Decision Tree, k-Nearest Neighbor (k-NN), Support Vector Machines (SVM), Naïve Bayes, and Artificial Neural Network (ANN) are used for fish species prediction. Finally, prediction decisions of the above machine learning models are combined to select the final fish class based on majority vote. The experiment is performed on a subset of ‘QUT_fish_data’ dataset containing 256 fish images of 21 classes and the result (accuracy 98.46%) shows that though the proposed method does not outperform all existing fish classification methods but it outperforms many existing methods and so, the method is a competitive alternative in this field.
Бесплатно
