International Journal of Computers and Communications

E-ISSN: 2074-1294
Volume 14, 2020

Notice: As of 2014 and for the forthcoming years, the publication frequency/periodicity of NAUN Journals is adapted to the 'continuously updated' model. What this means is that instead of being seperated into issues, new papers will be added on a continuous basis, allowing a more regular flow and shorter publication times. The papers will appear in reverse order, therefore the most recent one will be on top.

Main Page

Submit a paper | Submission terms | Paper format

 


Volume 14, 2020 


Title of the Paper: Regression Analysis of Solar Flares: A Multilayer Perceptron Approach with Feature Selection Techniques

 

Authors: Mohamad Tamer Rabie, Ali Bou Nassif, Maha Alaa Eddin

Pages: 84-90

https://doi.org/10.46300/91013.2020.14.14

Abstract: In this paper, we are going to analyze and test the solar flare dataset from the UCI Machine Learning Repository [10], by improving it using feature selection techniques such as stepwise regression, detecting the most effective attributes, importance ranker using k-fold and leave-one-out cross validation methods. We are going test the model by evaluating the dataset using Multi-linear regression, by looking at the P-values and the VIF to show the effectiveness of the dataset attributes. Multilayer perceptron model will be created using the holdout regression by partitioning the dataset into training and testing to model the testing dataset into an MLP model with 5 hidden layers. The model will show the mean absolute error and variance of the model to test its accuracy.


Title of the Paper: Graduate Admission Prediction Using Machine Learning

 

Authors: Sara Aljasmi, Ali Bou Nassif, Ismail Shahin, Ashraf Elnagar

Pages: 79-83

https://doi.org/10.46300/91013.2020.14.13

Abstract: Student admission problem is very important in educational institutions. This paper addresses machine learning models to predict the chance of a student to be admitted to a master’s program. This will assist students to know in advance if they have a chance to get accepted. The machine learning models are multiple linear regression, k-nearest neighbor, random forest, and Multilayer Perceptron. Experiments show that the Multilayer Perceptron model surpasses other models.


Title of the Paper: Emirati-Accented Emotion Verification based on HMM3s, HMM2s, and HMM1s

 

Authors: Ismail Shahin, Noor Ahmad Al Hindawi

Pages: 73-78

https://doi.org/10.46300/91013.2020.14.12

Abstract: The proposed research is dedicated to verifying the claimed emotion of speaker-independent and text-independent formed on three dissimilar classifiers. The HMM3 short for Third-Order Hidden Markov Model, HMM2 short for Second-Order Hidden Markov Model, and HMM1 short for First-Order Hidden Markov Model are the three classifiers utilized in this study. Our work has been evaluated on our collected Emirati-accented speech corpus which entails 50 speakers of Emirati origin (25 female and 25 male) uttering sentences in six emotions by means of the extracted features by Mel-Frequency Cepstral Coefficients (MFCCs). Our outcomes prove that HMM3 is superior to each of HMM1 and HMM2 to authenticate the claimed emotion. The achieved results formed on HMM3 are very similar to the outcomes attained in the subjective valuation by Arab listeners.


Title of the Paper: A Class of Evolutionary Algorithms for Determining Shortest Path Routing in 5G Ultra Dense Heterogeneous Networks

 

Authors: Debashis Dev Misra, Kandarpa Kumar Sarma, Pradyut Kumar Goswami, Utpal Bhattacharjee

Pages: 63-72

https://doi.org/10.46300/91013.2020.14.11

Abstract: Ultra Dense Network (UDN), an important element of the upcoming 5G networks are characterised by extremely dynamic operations due to the presence and mobility of large number of users spread over small cells of varying sizes. It makes optimal path between the source/destination pairs for communication and data transmission be highly dynamic in nature and hence a challenging issue to deal with. Under such dynamic backdrops, routing procedures have to exhibit robustness, scalability and time efficiency in order to ensure seamless link reliability and Quality of Service (QOS) of the network. In the proposed work, the shortest optimal route of the source and destination pair is found using a combination of evolutionary optimization algorithms namely Genetic Algorithm (GA), Particle Swarm Optimization (PSO) Algorithm and our novel hybrid PSO-GA approach which searches for an optimized solution by determining cost functions of individual fitness state and comparing states generated between individual solutions. Application of all the three above mentioned algorithms to the Shortest Path Routing Problem in UDNs and the results obtained have shown that the hybrid PSO-GA comparatively provided enhanced optimized solution.


Title of the Paper: Distribution Problems, Their Modifications and Applications

 

Authors: Miloš Šeda, Jan Roupec, JindÅ™iška Šedová

Pages: 57-62

https://doi.org/10.46300/91013.2020.14.10

Abstract: In this paper, we deal with well-known distribution problems and discuss their restrictions, extensions and modifications including a possible application in agriculture. We show that the transportation problem can be transformed to an assignment problem using special constraints, but because of NPhardness it needs quite different methods of its solving. Another modification of the transportation problem, the crop problem, has an application in agriculture, but we must deal with uncertain data. We propose a genetic algorithm and fuzzy logic approach for solving these problems.


Title of the Paper: One Approach to Solve Some Problems of Management Under Uncertainty

 

Authors: Teimuraz Tsabadze, Archil Prangishvili

Pages: 48-56

https://doi.org/10.46300/91013.2020.14.9

Abstract: We consider the approaches to making decisions for control problems in nonstandard situations with a lack of the previous experience and incomplete knowledge of the considered problem. In such cases we usually cannot do without expert evaluations which lead to the process of group decision-making, and it becomes necessary to solve a problem of alternatives aggregation. It has been proposed to solve such problems by means of fuzzy sets. The approach is based on the coordination index and the similarity of finite collections of fuzzy sets and takes into account the specific character of the fuzzy aggregation operator. The approach is discussed and its algorithm is presented. An example of the application of the proposed method is given.


Title of the Paper: Ant Decision Systems for Combinatorial Optimization with Binary Constraints

 

Authors: Nicolas Zufferey

Pages: 44-47

https://doi.org/10.46300/91013.2020.14.8

Abstract: In this paper is considered a problem (P) which consists in minimizing an objective function f while satisfying a set of binary constraints. Function f consists in minimizing the number of constraints violations. Problem (P) is NP-hard and has many applications in various fields (e.g., graph coloring, frequency assignment, satellite range scheduling). On the contrary to exact methods, metaheuristics are appropriate algorithms to tackle medium and large sized instances of (P). A specific type of ant metaheuristics is designed to tackle (P), where in contrast with state-of-the-art ant algorithms, an ant is a decision helper and not a constructive procedure.


Title of the Paper: Intelligent Offer Model for Game Online Item with Principle Hybrid Filtering Together with Categorical Data Technique

 

Authors: Khammapun Khantanapoka

Pages: 39-43

https://doi.org/10.46300/91013.2020.14.7

Abstract: This paper focus to Intelligent offer pattern, The commonly used raw material, wearing: costumes, earrings, gloves, gloves, boots, shoes, costumes, hair, necklace, two hands, costume masks, hats, costumes, weapons, clothes, watches, rings, plaques and feature character. Robot and various tools are built to response of players. One of the expectations of most players is items selection which is selected from the User Interface. It is obtrusive noise of view screen while players select a few of the items. Object Items of games offer to each player overlap on the screen. It helps reduce event to switch the screen to maximize revenue from the purchase, sale, and exchange object items in the online game. we use Double Two dimension Principle Component Analysis for the feature extraction data behavior which is behavior a repeated in each player such as character of clothes, characteristic of choose activity , characteristic of use game items. We built Collaborative group pattern (CGP) and analysis player separate to each corporative group. Adding create offer game items in each corporative group. Each player will receive the best offers. The game operator has the opportunity to sell game items online as possible by without interrupting game play. The results of experiment showed that this method can offer a game item which is player want to sequence in the top of the list, and average 11.41 percentages is garbage items which player does not want, by game items which each player need to be in the order of 1-10 of menu.


Title of the Paper: Lanczos Model Reduction for Switched Linear Systems

 

Authors: Mohamed Kouki, Mehdi Abbes, Abdelkader Mami

Pages: 33-38

https://doi.org/10.46300/91013.2020.14.6

Abstract: In today the methods reduction of large-scale linear time invariant and complexe systems are very many, the best choices today is the used of the krylov subspace methods based on moment matching. As hybrid dynamical systems are of rising spread and complexity, for these reasons, we present in this paper two model reduction methods applied to linear switched system. Which is an important class of hybrid and non linear system. Tow methods for reduction systems are present. In first part we present the modified non symmetric Lanczos algorithm, which is numerically efficient and applicable of any order. In second part we present the modified global lanczos algorithm, it is also numerically efficient, applicable of any order and having a best numerical stability. The effectivity and suitability of these new methods is illustrated by one simulation example.


Title of the Paper: Validity of Use of Various Concepts of Risk Management and Risk Engineering in Practice

 

Authors: Dana Prochazkova

Pages: 25-32

https://doi.org/10.46300/91013.2020.14.5

Abstract: The paper passes judgement of concepts of management disciplines and the engineering disciplines directed to trade-off (negotiation) with risks that have been used with aim to ensure the safety of buildings, territories and human communities considered as systems. Individual concepts of these disciplines fulfil different goals, are based on various assumptions, have different demands on knowledge, data, forces, sources and means, and therefore, they involve different measures and activities for implementation in practice. The investigation, the results of which are furthermore presented, categorizes tasks in practice in which it is necessary to use very advanced procedures and in which simple ones are sufficient. The special attention is paid to engineering domain and to cases in which advanced procedures may be used for ensuring the safety of both, the system of systems and the system of systems´ vicinity.


Title of the Paper: A New Method for Fuzzy Ranking Based on Possibility and Necessity Measures

 

Authors: Mohammad R. Sadeghi Moghadam, Tooraj Karimi, Mohammad B. Menhaj, Somaye Rahimi

Pages: 19-24

https://doi.org/10.46300/91013.2020.14.4

Abstract: In this paper, a new method to rank fuzzy numbers is presented. The proposed method based on Possibility and Necessity Measures is called PNM. According to possibility and necessity measures, eight indexes are calculated to extract four rules to rank fuzzy numbers. Also a method to evaluate each rule validation especially when rules’ outcomes yield conflict conclusions is presented. To test PNM performance, some controversial triangular fuzzy numbers are considered. Additionally, four extracted rules are compared with each other and fully analyzed. Furthermore, PNM is compared with other recently proposed methods. Results confirm that PNM is capable to rank a variety of fuzzy numbers and their images with any selected bandwidths, interval and any degree of closeness.


Title of the Paper: Application of Model Transformation for Optimized Building Energy Management

 

Authors: Quoc Dung Ngo, Yanis Hadj-Said, Stephane Ploix, Bernard Parisse, Ujjwal Maulik

Pages: 11-18

https://doi.org/10.46300/91013.2020.14.3

Abstract: This paper describes an approach aiming to automatically transform a model describing a high level physical behavior model into two different optimized building energy management application models. The first step consists in building a hinge model composed of element models. Then based on MDE approach, this model is projected, according to transformation processes, to application models. This paper presents core specifications of manipulation and transformation of hinge model. To illustrate this approach, an example of transformation into both an acausal anticipative model based on mixed integer linear programming problem and a non-linear causal model for fast simulated annealing optimization are shown. These models are used for energy management of a smart building platform named PREDIS/MHI.


Title of the Paper: Multi-output Hybrid GA-NN with Adaptive Mechanism

 

Authors: Faridah Sh Ismail, Nordin Abu Bakar

Pages: 5-10

https://doi.org/10.46300/91013.2020.14.2

Abstract: This research presents a hybrid Genetic Algorithm Neural Network (GA-NN) model to simulate the physical tests procedures of Medium Density Fiberboard (MDF). Data included in the model are related to MDF properties and its fiber characteristics. Multilayer Perceptron NN is a reliable supervised machine learning model. The model learns from seven inputs fed to the network to produce prediction of three targets. In order to avoid result from local optimum scenario, GA optimizes synaptic weights of the network towards reducing prediction error. The research used a fixed probability rates for crossover and mutation for hybrid GA-NN model. GA-NN model is further improved using adaptive mechanism to help identify the best probability rates for crossover and mutation. Fitness value refers to Sum of Squared Error, which is the accumulation of network error in the Output Layer. The population fitness distribution will guide best rates for each epoch. Performance comparisons are among three models; namely NN with Back Propagation (BP), hybrid GA-NN and hybrid GA-NN with adaptive mechanism. Results show the hybrid GA-NN model perform much better than NN model with back propagation optimizer. Adaptive mechanism in GA helps increase capability to converge at zero sooner than the ordinary GA.


Title of the Paper: An Improved User Authentication Scheme for Hierarchical Wireless Sensor Networks without Tamper-Proof Smart Card

 

Authors: Min-Shiang Hwang, Eko Fajar Cahyadi, Yuen-Cheng Chou, Cheng-Ying Yang

Pages: 1-4

https://doi.org/10.46300/91013.2020.14.1

Abstract: Recently, Maitra et al. proposed an efficient and robust user authentication scheme for hierarchical wireless sensor networks. They claimed that their scheme does not need tamper-proof smart card and resisted different possible attacks include smart card stolen attack, impersonation attack, privileged insider attack, replay attack, off-line password guessing attack, theft attack, session key recovery attack, denial of service attack, and cluster head capture attack. However, we find some weaknesses of his scheme in this article. We show that their scheme is vulnerable to off-line password guessing with smart card stolen attack.