المؤتمر الدولى الثانى لمعامل التأثير العربى
مايو 6-9 - مدينة زويل - القاهرة
....
 
  • الاديب الكبير الدكتور مصطفى رجب - و الدكتور اللواء شوقى صلاح
  • حضور الاستاذ الدكتور عصام شرف رئيس وزراء مصر الاسبق حفل الافتتاح - الاعلامى الاستاذ جمال الشاعر -
  • - قررت اللجنة المنظمة اقامة المؤتمر فى موعدة بمدينة زويل بالقاهرة
 
  ورشة عمل الانظمة الذكية : دكتور ابو العلا عطيفى
 
 
 
 
 

Workshop on Intelligent System and Optimization (ISO’17)

 

Zewail City of Science and Technology, Egypt

May 5, 2017

 

Chair Professor Aboul Ella Hassanien 

 

 

Co-chair Dr. Essam Halim Houssein

 

 

 

8.00 : 3:00

Registration

9 : 9:30

Welcome message by Professor Aboul Ella Hassanien

Keynote speaker

 

The impact of the CloudIoT-Health integration on the future of healthcare systems: A review

By Dr. AShraf Darwish,

Vice chair of the Scientific Research Group in Egypt

9.30-10.15

 

Abstract: Cloud Computing (CC) and Internet of Things (IoT) have emerged as new platforms in the ICT revolution of the 21st century. The adoption of the CloudIoT paradigm in the healthcare field can bring several opportunities to medical IT, and experts believe that it can significantly improve healthcare services and contribute to its continuous and systematic innovation. Moreover, CC and IoT are considered as complementary technologies such that they can enhance and improve each other capacity and capabilities. Their adoption and use are expected to be more pervasive, making them important components of the future of Internet and healthcare applications. This paper presents a review of the current literature on integration of CC and IoT to solving various problems in healthcare applications such as In addition, a brief introduction to cloud computing and internet of things with an application to health care is given. This paper presents a new concept of the integration of CC and IoT for healthcare applications, which is what we; call the CloudIoT-Health paradigm. The term CloudIoT-Health and some key integration issues have been presented in this paper to offer practical vision to integrate current components of CC and the IoT in healthcare applications. This paper aims to present the state of the art and gap analysis of different levels of integration components, analyzing different existing proposals in CloudIoT-Health systems and pointing out some challenges and open research issues. In addition, previous research work has been reviewed CC and IoT integration for healthcare systems. Challenges to be addressed and future directions of research are identified and an extensive bibliography is also included.

 

Session (I): Intelligent System and applications

                                                         Session Chair Dr. Essam Halim

10.30 : 10:45

Moatez Kilany

Emotion Recognition Based on EEG Signals.

 Abstract: The ability to recognize emotional states of people surrounding us is an important portion of natural communication. We present an emotion recognition approach using electroencephalography (EEG) signals based on three emotional scales; valence, arousal and dominance. EEG raw data were pre-processed to remove artifacts using discrete wavelet transform (DWT) support vector regression (SVR) is combined with EHO optimization technique to predict emotional scales. Experimental results proved that SVR has achieved better results in terms of regression accuracy by 99% after applying EHO.

10.45: 11.00

Sara Abdelghafar

Telemetry Data Mining for Space Systems

 Abstract: Space systems are amongst today’s most complex, extremely sensitive and highly sophisticated technical systems, they fulfil their mission in a very special, harsh, and challenging environment. So it is practically impossible to completely eliminate the possibility of anomalies or faults, even if we increase the reliability of the system components to the limit. In addition to, it is extremely difficult to directly inspect or repair a damaged component of these systems once a severe failure occurs. Therefore, early detection and predication of anomalies and faults in the system behaviour is significantly important to avoid disastrous situations such as loss of control.

11.00: 11.15

Mohamed-Abd-Elfattah

Handwritten manuscripts image binarization (problems and solutions)

 Abstract: Since the importance of the optimization algorithms in many real-life applications. Optical character recognition, word spotting, and other application need a cleared binarized image. A cleared binarized image without any distortion or hidden character. In this lecture, a quick review about the handwritten manuscripts problems and solutions will be presented. In addition to, the comparison between different binarization methods will be presented.

11.15: 11.30

Ragia Ibrahim

Group Impact: Local Influence Maximization in
Social Networks

Abstract: Influence maximization defined as the problem of selecting influential small set of nodes that maximize influence spread over the social network. Influence maximization considered in number of domains, emergence situations, viral marketing, education, collaborative activities and political elections. In this paper, we propose Local Information Maximization LIM. considering group impact in terms of local propagation where the influencer(s) of each community has a direct effect on the nodes in the same community. We conduct experiments on synthetic data set and compare the performance of the LIM to various heuristics.

11.30: 11.45

Usama Mokhtar

Binary moth flam features selection approach 

 Abstract: The feature selection process is one of the most important tasks for pattern recognition and classification. The main goal of feature selection is to find a minimal feature subset from a problem domain with high accuracy in representing the original features. In this paper, a new binary version of the moth flame optimization (MFO) is presented. Then this approach is used to select optimal feature subset for classification purposes. In the proposed (BMFO) approach, sigmoidal function is used to squash the continuously updated position, then randomly threshold these values to find the updated binary moth flame position.

11.45: 12.00

Abdelazim Galal

Binary Whale Optimization Algorithm for feature selection

 Abstract: binary version of the whale optimization algorithm (WOA) is proposed and used to select optimal feature subset for classification purpose. WOA is one of the latest bioinspired optimization techniques which Simulate hunting purpose of whale. This algorithm are hired in feature selection domain for finding feature subset maximizing the classification accuracy while minimizing the number of selected feature. The results were compared to two common optimizer Particle Swarm Optimization and Genetic Algorithm.

12.15: 12:30

Mahamad Eissa

A Fast Fragmented Pairwise Local Alignment Technique using Particle Swarm Optimization

 Abstract: Pairwise local sequence alignment is a main task in a bioinformatics that finds the common subsequences between two DNA, protein or RNA biological sequences. Smith-Waterman algorithm was developed to align two biological sequences locally and it is the most accurate local alignment technique. The main disadvantage of it is the huge increasing of execution time with increasing the length of sequences. This paper proposed a faster technique using the fragmentation of the sequences into short fragments. Smith-Waterman alignment is performed on these fragments to find common subsequences of the whole sequences. Particle swarm optimization is one of the stochastic optimization techniques was chosen for guiding quickly the search into the fragments maximum alignment’ score. The proposed technique was tested on real protein sequences have average length 300 bases to verify its alignment function which gave identical result of that using Smith-Waterman. For more long sequences the proposed algorithm successes to find the common subsequences with accuracy 97% but the speed up over Smith-Waterman algorithm has quadratic increasing with increasing the lengths of sequences.

12.30: 12:45

Sameh Basha

 

 

 

 

Neutrosophic Logic in Data Mining

Abstract: This talk presents a non-trivial process of dealing with incomplete, imprecise, and inconsistent knowledge in classification and clustering of large-scale data using neutrosophic logic (NL). Neutrosophic logic is adapted for representing different forms of knowledge. The presented system generalizes using fuzzy logic in data mining by describing every logical variable with its degree of truth, degree of indeterminacy, and degree of falsity. In classification stage these degrees are obtained from truth, indeterminacy, and falsity membership functions extracted from the fuzzy trapezoidal membership function. Then, it is followed by an extraction of the "IF-THEN" rules which used in the classification phase. Also in clustering stage we generalize fuzzy c-mean technique by neutrosophic c mean. The performance of the proposed systems is tested on three real-world databases Iris, Wine, and Wisconsin Diagnostic Breast Cancer (WDBC). In a series of experiments, we compare the performance of the proposed neutrosophic classification system with that of the fuzzy classification system and we compare the performance of the proposed neutrosophic c mean with that of the fuzzy c mean.

Break

Session (II): Optimization in Real Life Problems

Session Chair Dr. Tarek Gaber,  Faculty of Computers and Informatics, Canal Suze University

1.00: 1.15

Ahmed Metwalli Anter

Bio-inspiring computational segmentation approach for improving greeness
identification in agricultural images

Abstract: In Precision Agriculture (PA) automatic image segmentation for plant identification is an important issue to be addressed. In this paper we propose a new successful segmentation approach to distinguish between both soil and green parts in maize plant based on bio-inspired Particle Swarm Optimization algorithm (PSO). The proposed approach consists of the following steps: (1) Preprocessing adaptive color histogram equalization uses to adjust image intensities and to enhance contrast, (2) Green index is extracted based on vegetation color to guide PSO for good segmentation. PSO inspired by birds flocking in search of food. PSO consists of a number of particles that move on search space in search of the global optimum, 3) Multilevel thresholds are extracted from PSO based on the image intensity and the problems of Otsu threshold is solved, then (4) maize plant identified based on four level threshold. This approach gives useful results in different atmospheric and daylight conditions especially when the quality of imaging has low greenness.

1.15: 1.30

Khaled Ahmed

Intelligent approaches for social network analysis 

 Abstract: Social network analysis (SNA) is the process of investigating social structures through the use of network and graph theories. It characterizes networked structures in terms of nodes (individual actors, people, or things within the network) and the ties, edges, or links (relationships or interactions) that connect them. Examples of social structures commonly visualized through social network analysis include social media networks, memes spread, friendship and acquaintance networks, and collaboration graphs.

1.30: 1.45

Takwa Mohamed 

Prediction approach of possible mutations in influenza A virus based on spiking neural network

 Abstract: Continual and accumulated mutations in hemagglutinin (HA) protein of influenza A virus generate novel antigenic strains that cause annual epidemics. In this paper, a prediction approach based on Spiking Neural Network (SNN) is proposed to predict antigenic variants using a collected sequences of influenza A viruses. Furthermore, the prediction of mutation is treated as a classification problem, which can be solved using regression techniques. Thus, the spiking neural network is utilized to predicting the mutation positions in H1 neuraminidase from influenza A virus. Then, the amino-acid mutating probability to predicting the would-be-mutated amino acids at predicted positions is applied. Finally, the results confirm the possibility of prediction of mutation using this approach and pave the way for future development.

1.45: 2.00

Asmaa Hamad

EEG Signals Classification for Epileptic Detection based on Whale Optimization enhanced Support Vector Machine

Abstract: Epilepsy is one of the most widespread a chronic neurological disorders of the brain that affect millions of the world’s populations. It is characterized by recurrent seizures, which are physical reactions to sudden, usually brief, excrescent electrical discharges in a group of brain cells. Hence, seizure identification has great importance in clinical therapy of epileptic patients. Electroencephalogram (EEG) is most commonly used in epilepsy detection since it includes precious physiological information of the brain. However, it could be a challenge to detect the subtle but critical changes included in EEG signals due to the non-stationary and chaotic nature of it. In this paper, the discrete wavelet transform (DWT) is applied to EEG signals to pre-process, decompose it till the 4th level of decomposition tree. Various features like entropy, min, max, mean, median, standard deviation, variance, skewness, energy and relative wave energy (RWE) were computed in terms of detailed coefficients and the approximation coefficients of the last decomposition level. Furthermore, the whale optimization algorithm (WOA) was utilized to find the effective feature subset of EEG from a larger feature pool and the selected subset can provide better classification performance compared with using the whole set, also WOA used to optimize the SVM parameters, finally SVM with a Radial basis kernel function (SVM-RBF) was used for the classification process. The experimental results demonstrate that the highest classification accuracy (100%) for normal subject data versus epileptic data. The experimental results confirmed that the proposed WOA-SVM approach, able to detect epileptic and could thus further enhance the diagnosis of epilepsy.

2.00: 2.15

Shahd Tarek

Microscopic Images in Blood Diseases 

Abstract: Blood consist of Plasma, Red Blood Cells, White Blood Cells and Platelets, changing in blood condition show the development of diseases and help in the diagnosis of blood cells disorder . There are many disorders include benign disorders, as well as cancers that occur in blood. This presentation will give short brief to summarize blood diseases, including the definition, types, symptoms that helps in early detection through visual inspection of microscopic images.

2.15: 2.30

Ayat Taha

Arabian Horse Identification System based on Muzzle Print

Abstract: Classical animal identification and tracking methods such as ear tags, branding, tattooing, and electrical methods have long been in use; however, their performance is limited due to their vulnerability to losses, duplications, fraud, and security challenges. In this paper, we propose Arabian Horse identification system based on muzzle print images. In the proposed system, the Scale Invariant Feature Transform (SIFT) is used to extract feature of muzzle print images and SIFT algorithm is applied to match input feature with the feature that stored in the database. In order to enhance the robustness of the proposed system, the Random Sample Consensus (RANSAC) algorithm has been coupled with the SIFT match output to remove the mismatches point and achieve more robustness. The experimental results showed that the superiority of the proposed system is 100% identification accuracy in reasonable processing time.

2.30: 2.45

Samar Seif 

Multimodal biometrics recognition system using feature fusion 

 Abstract: Biometrics basically is a measure and analysis of human physiological and behavioral characteristics to identify and verify the person. Multimodal biometrics has drawn lot of attention in recent days as it provides more reliable scheme for person verification multimodal biometrics includes the fusion of information from different modalities. Multi biometrics combines more than one biometric trait; hence fusion of these traits plays a central role in designing multimodal biometric system. Fusion can be performed at data level, feature level, match score level or at decision level. Multimodal biometric systems elegantly address several of the problems present in unimodal systems. By combining multiple sources of information, these systems improve matching performance, increase population coverage, deter spoofing, and facilitate indexing. Incorporating user- specific parameters can further improve performance of these systems. With the widespread deployment of biometric systems in several civilian and government applications.

2.45: 3.00

Ramadan Babers

Blockchain Technology

Abstract: Blockchain is a decentralized transaction and data management technology. Blockchain technology has been increasing as it provides security, anonymity and data integrity without any third party organization in control of the transactions. The reason for the interest in Blockchain is as it offers a way for market participants to access dematerialized assets directly without always going through other participants needlessly. Blockchain can automate trading, clearing, and settlement functions on financial market. Bitcoin is a digital currency, payments system and decentralized ledger used in blockchain technology.

Break

Session (II): New trends

Session Chair Professor Ashraf Darwish

3.15: 3.30

      Abdalla Essam

 

Quantum Image Processing

Abstract: Quantum Image Processing (QIMP) is a new field that utilize the power of quantum computing technologies to capture, manipulate, and recover quantum image in different formats and for different purposes. QIMP algorithms and applications can investigate the problems and challenges in any topic in image processing like image filtering, image segmentation and feature extraction.

 

3.30: 3.45

Rizk masoud

 

A Hybrid Sine Cosine Optimization Algorithm for Solving Global Optimization Problems

 Abstract: In this paper, a hybrid sine cosine optimization (denoted as HSCO) algorithm is proposed, in which a local search strategy is hybridized with the sine cosine optimization. The local search strategy aims to enhance the solutions and to prevent premature convergence of the population. By this way, the algorithm can avoid the running without any improvements in the obtained results. The simulations were conducted on a set of the benchmark problems and compared to other optimization techniques that reported in the literature. The obtained results demonstrate the superiority of the proposed HSCO compared to other optimization techniques that are reported in the literature.

 

3.45: 4.00

Yasmine S. Moemen

 

Using Molecular Mechanic Study in Evaluation the Role of natural resources in Preventing Type II Diabetes

 Abstract: Type II Diabetes (T2D) is affected more than 285 million people worldwide. The International Diabetes Federation has announced that the Middle East and North Africa (MENA) region is considered the highest prevalence of diabetes in the world (37 million) and this expected to rise dramatically. India has the second highest incidence of diabetes (31.7 million), China (20.8 million) and finally the USA (17.7 million) while. Many synthetic compounds like Gliptin, Metformin and Pioglitazone are used as antidiabetic. Although the mechanism of action of these compounds still fully unknown and not decisive in treating T2D. Dipeptidyl peptidase-4 (DPP4) enzyme is considered target enzyme to control T2D, where a strong relation exist between DPP4 and hemoglobin A1c (HbA1c) which observed in T2D patients. The phenolic compounds present in in natural resources, are associated to hypoglycemic besides many other therapeutic effects like antioxidant, antihypertensive, hypocholesterolemic and cardioprotective activity. The current study aims to screen the mechanism of action of such compounds as DPP4 inhibitors, using molecular docking procedures and molecular mechanic simulation for better accuracy.

4.00: 4.15

Ahmed Hassan

 

Chemoinformatics and Drug Discovery

 Abstract: Discoveries of new drugs is very costly process. To reduce costs for drugs design, new techniques are needed. Chemoinformatics implements the informational techniques and computer science like machine learning and graph theory to discover the chemical compounds properties, such as toxicity or biological activity. This is done through analyzing their molecular structure (molecular graph). Therefore, there is an increasing need for algorithms to analyze and classify graph data to predict the activity of molecules.

4.15: 4.30

Sally Elghamrawy

 Spectrum Sensing Optimization in Cognitive Radio Network Using Genetic Artificial Bee Colony (GABC) Algorithm

 Abstract: Due to the exponential increase in the number of wireless devices and in data rates demands, the shortage of the free frequency spectrum resources becomes a critical problem. As a result, there is an urgent need for proposing dynamic techniques that have the ability to avail the available spectrum. Recently, Cognitive radio (CR) is considered as an emerging technology that eases off the problem of spectrum shortage by optimizing the usage of the available spectrum through giving the unlicensed users the opportunity to exploit the spectrum without causing any interference with the primary users’ usage. Therefore, the unlicensed users must sense the spectrum used by the primary users, through the use of spectrum holes, to access the unoccupied channels. Many efforts had been made for solving the spectrum sensing challenges. In this paper, a hybrid Genetic Artificial Bee Colony (GABC) algorithm is proposed to accommodate the unlicensed users in optimal possible space in the spectrum. GABC combines the advantages of the Genetic algorithm GA along with the Artificial Bee Colony ABC algorithm to select the more informative genes in order to optimize spectrum allocation without reaching the local optima caused by applying GA. The simulation results show promising performance of GABC in optimizing spectrum sensing, when compared with other intelligence algorithms.

 

4.30: 4.45

Tarek Gaber

Human thermal face recognition approach  

 Abstract: Accurate human identification is very crucial in many applications including the passports utilization, cellular telephones, driving licenses and other computer vision applications. Face recognition, as a natural mode of identification among humans, is the most appealing modality for human identification. There are many successful face recognition systems based on visual images. However, they are still prone to errors (variable illumination and disguises including facial wear and hair). To address these errors, investigating face image outside the visible part of the spectrum (wavelength ranging from 390 to 750nm) has been explored. The main idea of thermal imaging is that each object emits infrared energy different than other objects, i.e., each object has different thermal signatures. In the case of the human face, this signature is primarily derived from the pattern of the superficial blood vessels existed under the face skin. Because of the vein and tissue structure of each face are unique for each person thus its thermal image is also unique.

This paper proposes a human thermal face recognition approach with two variants based on Random linearOracle (RLO) ensembles. For the two approaches, the Segmentation-based Fractal Texture Analysis (SFTA) algorithmwas used for extracting features and the RLO ensembleclassifier was used for recognizing the face from its thermalimage. For the dimensionality reduction, one variant (SFTALDA-RLO) was used the technique of Linear DiscriminantAnalysis (LDA) while the other variant (SFTA-PCA-RLO) wasused the Principal Component Analysis (PCA). The classifiersmodel was built using the RLO classifier during the trainingphase and in the testing phase then this model was usedto identify the unknown sample images. The two variantswere evaluated using the Terravic Facial IR Database and theexperimental results showed that the two variants achieved agood recognition rate at 94.12% which is better than related work.

Closing and Certificates

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 
 
 
Sponsors
 
 
 
 
   
 
  All rights reserved 2015 - NSP, www.naturalspublishing.com