Статьи журнала - International Journal of Modern Education and Computer Science
Все статьи: 968
A Survey on Journey of Topic Modeling Techniques from SVD to Deep Learning
Статья научная
Topic modeling techniques have been primarily being used to mine the topics from text corpora. These techniques reveal the hidden thematic structure in a collection of documents and facilitate to build up new ways to browse, search and summarize large archive of texts. A topic is a group of words that frequently occur together. A topic modeling can connect words with similar meanings and make a distinction between uses of words with several meanings. Here we present a survey on journey of topic modeling techniques comprising Latent Dirichlet Allocation (LDA) and non-LDA based techniques and the reason for classify the techniques into LDA and non-LDA is that LDA has ruled the topic modeling techniques since its inception. We have used the three hierarchical classification criteria’s for classifying topic models that include LDA and non-LDA based, bag-of-words or sequence-of-words approach and unsupervised or supervised learning for our survey. Purpose of this survey is to explore the topic modeling techniques since Singular Value Decomposition (SVD) topic model to the latest topic models in deep learning. Also, provide the brief summary of current probabilistic topic models as well as a motivation for future research.
Бесплатно
A Survey on Scheduling Heuristics in Grid Computing Environment
Статья научная
Job scheduling is one of the thrust research area in the discipline of Grid computing. Scheduling in the Grid environment is not only complicated but also known to be NP-Complete problem and that is all due to its unique characteristics. Thus, there are limited opportunities to find an optimal solution. In recent past, many eminent researchers reported a variety of Scheduling Heuristics that can have a substantial impact on the performance of the Grid systems. Unfortunately, it gives rise to difficulty in evaluating and keeping track of those solutions. Therefore, the motivation of this comprehensive study is to present firstly, an in-depth review of the topic under discussion mostly in the perspective of Grid Scheduling environment, and secondly, a proposal for a new state-of-the-art classification of the existing Scheduling Heuristics. All these Heuristics in each category have been further studied based on significant parameters frequently used in Scheduling Heuristics. The final part of this study includes a fair assessment of those mostly used dominating parameters. This report deals with the key concepts behind existing Scheduling Heuristics including Objectives, Types of Job Scheduling, Functionality of Grid, Nature of Grid, and the importance of the proposed classification.
Бесплатно
A Systematic Literature Review on Spell Checkers for Bangla Language
Статья научная
Spell checkers check whether a word is misspelled and provide suggestions to correct it. Detection and correction of spelling errors in Bangla language which is the seventh most spoken native language in the world, is very onerous because of the complex rules of Bangla spelling. There is no systematic literature review on this research topic. In this paper, we present a systematic literature review on checking and correcting spelling errors in Bangla language. We investigate the current methods used for spell checking and find out what challenges are addressed by those methods. We also report the limitations of those methods. Recent relevant studies are selected based on a set of significant criteria. Our results indicate that there are research gaps in this research topic and has a potential for further investigation.
Бесплатно
Статья научная
Vides a proper solution to this limitation. There are broadly three main categories of Vector Space Model: term-document, word-content and pair-pattern matrices. The main aim of this paper is to discuss broadly the three main categories of VSM for semantic analysis of texts and make proper selection for automatic categorizing. The scenario taken up here is categorization of research papers for organizing a national or an international conference based on the proposed methodology. Computers do not understand human language and this makes it difficult when human wants the computer to do some specific task like categorization according to human need. Vector Space Model (VSM) for semantic analysis of texts and make proper selection of one of the three main categories for automatic categorizing of research papers for organizing a national or an international conference based on the proposed methodology.
Бесплатно
A Two Layers Novel Low-Cost and Optimized Embedded Board Based on TMS320C6713 DSP and Spartan-3 FPGA
Статья научная
This paper presents the design and implementation of a new low-cost and minimum embedded board based on TMS320C6713 (PYP 208-PIN (PQFP)) DSP and Spartan-3 (XCS400-4PQG208C) FPGA in two layers with mount elements on two sides of the board. The proposed embedded board was developed satisfactorily for different applications such as data acquisition of sensor’s with serial port, control units, finite state machines, signal processing algorithms, navigation computing, Kalman filtering etc. Goal of the design was to implement as many as possible low-cost and minimum sizes of the board, also to receive input signals in a short time period and as real time. The board features are include: mount elements in two side of the board for minimization of the proposed board and also placed decoupling capacitors (by pass) for the DSP and FPGA in bottom layer of board strictly below these two ICs because should be placed as close as possible to the power supply pins DSP and FPGA, GND polygon layer is used in total top layer and microcomputer ground for DSP & FPGA in bottom layer, use FPGA for two aim ones for implementation of glue logic total of board and interface between serial connectors, use three RS-232 serial port, one RS-422, and SPI serial port on FPGA, use MT48LC16M16A SDRAM-256MB(4*4MB*16), Am29LV400B Flash memory 4 Megabit (512 K x 8-Bit/256 K x 16-Bit) and XCF02S configuration PROM. The size of the proposed embedded board is 11.1cm*17. 7cm so this board is optimized of aspect cost, performance, power, weight, and size.
Бесплатно
Статья научная
Assessment and evaluation of Program Educational Objectives (PEOs) and Student Outcomes (SOs) is a challenging task. In this paper, we present a unified framework, which has been developed over a period of more than eight years, for the systematic assessment and evaluation of PEOs and SOs. The proposed framework is based on a balance sampling approach that thoroughly covers PEO/SO assessment and evaluation and also minimizes human effort. This framework is general but to prove its effectiveness, we present a case study where this framework is successfully adopted by our undergraduate computer science program in the department of computer science at King Abdulaziz University, Jeddah. The robustness of the proposed framework is ascertained by an independent evaluation by ABET who awarded us full six years accreditation without any comments or concerns. The most significant value of our proposed framework is that it provides a balanced sampling mechanism for assessment and evaluations of PEOs/SOs that can be adapted by any program seeking ABET accreditation.
Бесплатно
A Wavelet Based Approach for Compression of Color Images
Статья научная
The use of color in image analysis and compression is becoming more and more popular. The high quality color images are in demand, but the bandwidth and power resources are limited, this shows the requirement of effective color image compression algorithm which is suitable to human visual system. However most of the existing algorithms are designed for gray scale visual information. In this work a unique wavelet based approach is proposed for compression of color images. Wavelet families are used to characterize the quality of image by calculating quality estimation parameters, which are, peak signal to noise ratio, energy retained, entropy and redundancy. The entropy calculations are done using color histogram and coding programme is developed for estimation of PSNR, ER and redundancy of the compressed image. The results are analyzed and a set of criteria is determined for the acceptability of coding algorithm. Results show that Biorthogonal wavelet filter outperforms the orthogonal one in quality of compressed image but the orthogonal filter is more energy preserving.
Бесплатно
A case analysis on different registration methods on multi-modal brain images
Статья научная
Many applications of artificial vision need to compare or integrate images of the same object but obtained at different moments of time with different devices (cameras), from different positions, under different conditions, etc. These differences in capture give rise to images with important relative geometric differences that prevent these "Fit" with precision over each other. The registry eliminates these geometric differences so that located pixels in the same coordinates correspond to the same point of the object and, therefore, both images can easily be compared or integrated. The registration of images is essential in disciplines such as remote sensing, radiology, robotic vision, etc. ; Fields, all of them, that overlap images to study environmental phenomena, monitor tumours carcinogenic or to reconstruct the observed scene. This paper also study different measures of similarity used to measure their consistency and a novel procedure is proposed to improve the accuracy of the linear record by pieces. Specifically the elements that influence the estimation are analysed experimentally of probability distributions of the intensity levels of the images. These distributions are the basis for calculating measures of similarity based on entropy as mutual information (MI) or the Entropy correlation coefficient (ECC). Therefore, the effectiveness of these measures depends critically on their correct estimation.
Бесплатно
A clustering algorithm based on local density of points
Статья научная
Data clustering is very active and attractive research area in data mining; there are dozens of clustering algorithms that have been published. Any clustering algorithm aims to classify data points according to some criteria. DBSCAN is the most famous and well-studied algorithm. Clusters are recorded as dense regions separated from each other by spars regions. It is based on enumerating the points in Eps-neighborhood of each point. This paper proposes a clustering method based on k-nearest neighbors and local density of objects in data; that is computed as the total of distances to the most near points that affected on it. Cluster is defined as a continuous region that has points within local densities fall between minimum local density and maximum local density. The proposed method identifies clusters of different shapes, sizes, and densities. It requires only three parameters; these parameters take only integer values. So it is easy to determine. The experimental results demonstrate the superior of the proposed method in identifying varied density clusters.
Бесплатно
A detailed examination of the enterprise architecture frameworks being implemented in Pakistan
Статья научная
Managing the underlying structure of an enterprise is a daunting task. The business management and IT management alike have to deal with intricate layers of complexity that lies beneath the surface of the day-to-day operations of an enterprise. Without a proper Enterprise Architecture Framework, any organization regardless of size and magnitude of operations is bound to struggle in managing their business strategies. However, choosing a suitable Enterprise Architecture Framework is in itself a pretty hard endeavor that requires a deep dive into the terrifying maze of available Enterprise Architecture Frameworks and their respective characteristics. In this study, we compare the major Enterprise Architectu¬re Frameworks that are currently prevalent in Pakistan. Through a well-crafted questionnaire we conducted a survey and assessed what Enterprise Architecture Frameworks most of the industries in Pakistan are using and the enterprise’s level of satisfaction with the achieved results. By focusing on the trends of Enterprise Architecture Framework implementation in Pakistan we try to offer a unique perspective on the comparative studies of Enterprise Architecture Framework that are usually done on general basis.
Бесплатно
A dynamic feedback-based load balancing methodology
Статья научная
With the recent growth of Internet-based application services, the concurrent accessing requests arriving at the particular servers offering application services are growing significantly. It is one of the critical strategies that employing load balancing to cope with the massive concurrent accessing requests and improve the access performance is. To build up a better online service to users, load balancing solutions achieve to deal with the massive incoming concurrent requests in parallel through assigning and scheduling the work executed by the members within one server cluster. In this paper, we propose a dynamic feedback-based load balancing methodology. The method analyzes the real-time load and response status of each single cluster member through periodically collecting its work condition information to evaluate the current load pressure by comparing the learned load balancing performance with the preset threshold. In this way, since the load arriving at the cluster could be distributed dynamically with the optimized manner, the load balancing performance could thus be maintained so that the service throughput capacity would correspondingly be improved and the response delay to service requests would be reduced. The proposed result is contributed to strengthening the concurrent access capacity of server clusters. According to the experiment report, the overall performance of server system employing the proposed solution is better.
Бесплатно
A feature selection based ensemble classification framework for software defect prediction
Статья научная
Software defect prediction is one of the emerging research areas of software engineering. The prediction of defects at early stage of development process can produce high quality software at lower cost. This research contributes by presenting a feature selection based ensemble classification framework which consists of four stages: 1) Dataset selection, 2) Feature Selection, 3) Classification, and 4) Results. The proposed framework is implemented from two dimensions, one with feature selection and second without feature selection. The performance is evaluated through various measures including: Precision, Recall, F-measure, Accuracy, MCC and ROC. 12 Cleaned publically available NASA datasets are used for experiments. The results of both the dimensions of proposed framework are compared with the other widely used classification techniques such as: “Naïve Bayes (NB), Multi-Layer Perceptron (MLP). Radial Basis Function (RBF), Support Vector Machine (SVM), K Nearest Neighbor (KNN), kStar (K*), One Rule (OneR), PART, Decision Tree (DT), and Random Forest (RF)”. Results reflect that the proposed framework outperformed other classification techniques in some of the used datasets however class imbalance issue could not be fully resolved.
Бесплатно
A framework for ensuring consistency of Web Services Transactions based on WS-BPEL
Статья научная
Transaction processing, as the key technology of web service composition (WSC), has obtained wildly concern. WS-BPEL[1] as a primary web service composition description language, which couldn’t coordinate these web service transactions that distribute in a distributed computing environment reach consistent agreement on the outcome. This paper proposed two kinds of transaction types and coordination mechanisms by analyzing the features of WSC transaction, and a transaction processing coordination model based on BPEL was lastly proposed, by which extending the structure of BPEL firstly and then introduced the coordination mechanism into it. The model was validated by an instance at last.
Бесплатно
A fuzzy based parametric approach for software effort estimation
Статья научная
Accurate Software effort estimation is an ongoing challenge for the modern software engineers in computer science engineering since last 30 years due to the dynamic behavior of the software [1] [2][14]. This is only because of the time and cost estimation during the early stage of the software development is quite difficult and erroneous. So many algorithmic and non algorithmic techniques are used such as SLIM (Software life cycle management), Halstead Model, Bailey-Basil Model, COCOMO model and Function point analysis, etc, but does not estimate all kinds of software accurately. Nowadays these traditional techniques are not acceptable. This research work proposes a new fuzzy model to achieve higher accuracy by multiplying a fuzzy factor with the effort equation predicted empirically. As comparison to both model based and equation based, Model based estimation focused on specific models where as equation based techniques are based on traditional equations. Fuzzy logic is more suitable and flexible to meet the realistic challenges of today’s software estimation process.
Бесплатно
Статья научная
Many approaches to sentiment analysis benefit from polarity lexicons. Existing methods proposed for building such lexicons can be grouped into two categories: (1) Lexicon based approaches which use lexicons such as dictionaries and WordNet, and (2) Corpus based approaches which use a large corpus to extract semantic relations among words. Adjectives play an important role in polarity lexicons because they are better polarity estimators compared to other parts of speech. Among natural languages, Turkish, similar to other non-English languages suffers from the shortage of polarity resources. In this work, a hybrid approach is proposed for building adjective polarity lexicon, which is experimented on Turkish combines both lexicon based and corpus based methods. The obtained classification accuracies in classifying adjectives as positive, negative, or neutral, range from 71% to 91%.
Бесплатно
Статья научная
This paper focuses on the personality traits of students and stress scale they had to face in undergraduate level. With the advancement of computer science and machine learning based applications, we have tried to inter-correlate the terms. In the area of computational psychology, it is important to understand participants’ psychological behavior using personality traits and predict how he/she is going to react on a certain level of the stressed situation. For the experiment, we have collected data of around 150 participants. The personality traits data are collected using the standard survey named The Big Five Personality Test created by IPIP organization and stress scale measurements are collected using scale devised by Sheldon Cohen named as Perceived Stress Scale hosted by Mind garden. The data are taken from Bangladeshi computer science undergraduate students and kept anonymous. In this paper, we have applied nine different machine learning based classification models are built for mapping the traits with stress scales. For performance evaluation, we have utilized precision, recall, f1-score, and accuracy. From the experimental findings, we found that Sequential Minimal Optimization (SMO) and k-NN classifier gives the highest prediction accuracy which is approximately 70%.
Бесплатно
A neoteric optimization methodology for cloud networks
Статья научная
Cloud computing is distinctively marked by its capability of providing on demand virtualized IT resources in a pay as you go fashion. Due to its popularity, the cloud computing users are increasing day by day which has become an important challenge for cloud providers. They need to serve their users in a best possible manner. The providers should not only provide their users a secure access to resources but also need to maintain a proper balance of QOS parameters like throughput, end-to-end delay, packet delivery ratio, jitter, response time, etc. The paper proposes an approach of using a meta-heuristic algorithm called Genetic Algorithm (GA) to optimize QOS parameters like packet delivery ratio and end to end delay in cloud networks. The intelligent optimization algorithms address several shortcomings of existing protocols by improving QOS parameters in an optimum manner. The results are simulated through MATLAB based simulator and the simulated results of proposed approach exhibit optimized parameters when compared to conventional method of shortest path cloud routing approach.
Бесплатно
A novel approach to predict high blood pressure using ABF function
Статья научная
High Blood Pressure (HBP) is a state in the biological system of human beings developed due to physical and psychological changes. Nowadays, it is a most prevalent problem in human beings irrespective of age, place, and profession. The HBP victims are increasing rapidly across the globe. HBP is undiagnosed in the majority of the patients because most of the affected people are not aware of it. To overcome this problem, this paper proposes a new approach that uses ABF (Arterial Blood Flow)-function to predict a person is prone to HBP. In this approach, the impact factor for each attribute is calculated based on the attribute value. Both attribute value and corresponding impact factor are used by ABF function to predict a person is prone to HBP. We experimented proposed approach on real-time data set, which consists of 1100 patient records in the age group between 18 and 65. Our approach outperforms regarding predictive accuracy over j48, Naive Bayes and Rule-based classifiers.
Бесплатно
A novel optimization based algorithm to hide sensitive item-sets through sanitization approach
Статья научная
Association rule hiding an important issue in recent years due to the development of privacy preserving data mining techniques for hiding the association rules. One of the mostly used techniques to hide association rules is the sanitization of the database. In this paper, a novel algorithm MPSO2DT has been proposed based on the Particle Swarm Optimization (PSO) in order to reduce the side effects. The aim is to reduce the side effects such as Sensitive item-set hiding failure, Non-sensitive misses, extra item-set generations and Database dissimilarities along with the reduction of running time and complexities through transaction deletion.
Бесплатно
A novel system for generating simple sentences from complex and compound sentences
Статья научная
In the field of natural language processing, simple sentence has a great importance; especially for multiple choice question generation, automatic text summarization, opinion mining, machine translation and information retrieval etc. Most of these tasks use simple sentences and include a sentence simplification module as pre-processing or post-processing task. But dedicated tasks for sentence simplification are hardly found. Here we have proposed a novel system for generating simple sentences from complex and compound sentences. Our proposed system is an initiative for simplifying sentence by converting complex and compound sentences into simple ones. Along with this the system classifies the simple sentences of an input corpus from other types of sentences. To generate simple sentences from complex and compound sentences we have proposed a novel algorithm which takes the dependency parsing of the input text and produce simple sentences as output. The experimental result demonstrates that the proposed technique is a promising one.
Бесплатно
