Pub Date : 2018-12-01DOI: 10.21608/mjcis.2018.312007
Abeer M. Hekal, Ahmed I. Saleh, Magdi Zakria
Mobility prediction is one of the main challenges that faced Personal Communication Service (PCS) network. It is probable for many users to move among cells (coverage areas) during their calls. Therefore, the network needs to predict their next location in order to reserve another resource in that next cell to keep their calls going on. In this paper, a new strategy is proposed for mobility prediction named Mixed Mobility Prediction (MMP). It is composed of two predictors. The first one is named Association Rules Predictor (ARP), and the second one is called Weighted Ant Colony Predictor (WACP). In ARP the prediction is based on Association rules in data mining and detecting the time of calls. In WACP the prediction is based on Ant Colony (AC) in swarm intelligence. In addition to that, roads lead to their predicted next locations, and priority of famous places found in those locations. Finally, MMP merges the decisions of both predictors to get the final accurate decision in the absence of sufficient history for a MT. The proposed approach outperformed the compared the state-of-arts methods in terms of; Prediction Accuracy (PA), and Quality of Measure (QM).
{"title":"A new strategy for mobility prediction in the PCS network","authors":"Abeer M. Hekal, Ahmed I. Saleh, Magdi Zakria","doi":"10.21608/mjcis.2018.312007","DOIUrl":"https://doi.org/10.21608/mjcis.2018.312007","url":null,"abstract":"Mobility prediction is one of the main challenges that faced Personal Communication Service (PCS) network. It is probable for many users to move among cells (coverage areas) during their calls. Therefore, the network needs to predict their next location in order to reserve another resource in that next cell to keep their calls going on. In this paper, a new strategy is proposed for mobility prediction named Mixed Mobility Prediction (MMP). It is composed of two predictors. The first one is named Association Rules Predictor (ARP), and the second one is called Weighted Ant Colony Predictor (WACP). In ARP the prediction is based on Association rules in data mining and detecting the time of calls. In WACP the prediction is based on Ant Colony (AC) in swarm intelligence. In addition to that, roads lead to their predicted next locations, and priority of famous places found in those locations. Finally, MMP merges the decisions of both predictors to get the final accurate decision in the absence of sufficient history for a MT. The proposed approach outperformed the compared the state-of-arts methods in terms of; Prediction Accuracy (PA), and Quality of Measure (QM).","PeriodicalId":253950,"journal":{"name":"Mansoura Journal for Computer and Information Sciences","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131191850","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.21608/mjcis.2018.312010
A. M. El-Edkawy, M. El-Dosuky, T. Hamza
Since the advent of DNA computing field; there’s debate about its ability to solve hard computational problems. A way to deal with the debate is to design a computational model for DNA computing and then studying its complexity power. The main contribution of this paper is to propose a DNA computing model that presents a usage of CRISPR in DNA computing field. The model has three basic operations which are merge, CRISPR and detect. The model is argued to be robust, as the three operations are robust, and to be the minimum model for DNA computing in terms of the number of operations. The proposed model has corresponded with a model inspired from the evolution of DNA sequences called Accepting Hybrid Network of Evolutionary Processors (AHNEPs). Based on that correspondence, the proposed model can be used to solve NP problems in polynomial time and PSPACE problems in polynomial space. The model is used to solve, the NP problem, the Hamiltonian Path Problem (HPP) in linear time. The limitations of DNA computing area could be avoided within the model. Also, the study presents “Visual DNA” which is a software that can simulate biochemical operations in DNA computing. Also, the software can make analysis for DNA sequences. So this simulation software will offer a useful tool for the DNA computing implementation because it will help in the analysis of input DNA sequences and in the prediction of output DNA sequences which would be helpful to avoid errors during the experimental process.
{"title":"Improving DNA Computing through CRISPR based Model and Visual DNA Tool","authors":"A. M. El-Edkawy, M. El-Dosuky, T. Hamza","doi":"10.21608/mjcis.2018.312010","DOIUrl":"https://doi.org/10.21608/mjcis.2018.312010","url":null,"abstract":"Since the advent of DNA computing field; there’s debate about its ability to solve hard computational problems. A way to deal with the debate is to design a computational model for DNA computing and then studying its complexity power. The main contribution of this paper is to propose a DNA computing model that presents a usage of CRISPR in DNA computing field. The model has three basic operations which are merge, CRISPR and detect. The model is argued to be robust, as the three operations are robust, and to be the minimum model for DNA computing in terms of the number of operations. The proposed model has corresponded with a model inspired from the evolution of DNA sequences called Accepting Hybrid Network of Evolutionary Processors (AHNEPs). Based on that correspondence, the proposed model can be used to solve NP problems in polynomial time and PSPACE problems in polynomial space. The model is used to solve, the NP problem, the Hamiltonian Path Problem (HPP) in linear time. The limitations of DNA computing area could be avoided within the model. Also, the study presents “Visual DNA” which is a software that can simulate biochemical operations in DNA computing. Also, the software can make analysis for DNA sequences. So this simulation software will offer a useful tool for the DNA computing implementation because it will help in the analysis of input DNA sequences and in the prediction of output DNA sequences which would be helpful to avoid errors during the experimental process.","PeriodicalId":253950,"journal":{"name":"Mansoura Journal for Computer and Information Sciences","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130046421","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.21608/mjcis.2018.312008
Hatem M Noaman, Shahenda S. Sarhan, M. Rashwan
Arabic Modern texts are commonly written without diacritization, which is a critical task for other Arabic processing tasks as word sense disambiguation, automatic speech recognition, and text to speech, where word meaning or pronunciation is decided based on the diacritic signs assigned to each letter. This paper presents a novel approach for automatic Arabic text diacritization using deep encode-decode recurrent neural networks that is followed by several text correction techniques, to improve the overall system output accuracy. Experimental results of the proposed system on Wikinews test set show superior performance and are competitive with those of the-state-of-the-art diacritization methods. Namely, our method achieves morphological diacritization Word Error Rate (WER) 3.85% and Diacritic Error Rate (DER) 1.12%.
{"title":"A Hybrid Approach for Automatic Morphological Diacritization of Arabic Text","authors":"Hatem M Noaman, Shahenda S. Sarhan, M. Rashwan","doi":"10.21608/mjcis.2018.312008","DOIUrl":"https://doi.org/10.21608/mjcis.2018.312008","url":null,"abstract":"Arabic Modern texts are commonly written without diacritization, which is a critical task for other Arabic processing tasks as word sense disambiguation, automatic speech recognition, and text to speech, where word meaning or pronunciation is decided based on the diacritic signs assigned to each letter. This paper presents a novel approach for automatic Arabic text diacritization using deep encode-decode recurrent neural networks that is followed by several text correction techniques, to improve the overall system output accuracy. Experimental results of the proposed system on Wikinews test set show superior performance and are competitive with those of the-state-of-the-art diacritization methods. Namely, our method achieves morphological diacritization Word Error Rate (WER) 3.85% and Diacritic Error Rate (DER) 1.12%.","PeriodicalId":253950,"journal":{"name":"Mansoura Journal for Computer and Information Sciences","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127704032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.21608/mjcis.2018.312004
I. I. Saada, R. H. Sakr, M. Rashad
A mobile ad hoc network (MANET) has many undesirable challenges, black hole (BH) is one of the most significant challenges in MANET. Many researchers have proposed researches which depend on different strategies to face this challenge. The problem is how much the ability of these strategies to detect and block BH in order to empty the network from the threats, and how much these solutions could be implemented practically to treat with threats. However, the required strategy should also maintain network performance at the same time. The idea of this research has been applied on AODV based MANET, the strategy deals with cooperative black hole by using a deceptive message with virtual address that is not in the addresses range [1]. The main interest in this paper beside proposing the new idea, is to convert this idea into strategy by providing the necessary details, in addition to implement the strategy by hardware and software models. These models have been built in order to illustrate how black hole works and how the strategy can detect and block the black hole practically in AODV based MANET. This paper presents the method in an integrated manner. In this method, the detection process starts from the previous node, not from the source node. However, this method can choose the shortest path and can deal with more than one black hole. These additions will preserve the performance of network, the hardware model has been provided with liquid crystal displays LCD put on each node. LCD is used to appear each step, and to make the system full-acknowledgment.
{"title":"Software and Hardware Implementation of Local Intrusion Detection Strategy","authors":"I. I. Saada, R. H. Sakr, M. Rashad","doi":"10.21608/mjcis.2018.312004","DOIUrl":"https://doi.org/10.21608/mjcis.2018.312004","url":null,"abstract":"A mobile ad hoc network (MANET) has many undesirable challenges, black hole (BH) is one of the most significant challenges in MANET. Many researchers have proposed researches which depend on different strategies to face this challenge. The problem is how much the ability of these strategies to detect and block BH in order to empty the network from the threats, and how much these solutions could be implemented practically to treat with threats. However, the required strategy should also maintain network performance at the same time. The idea of this research has been applied on AODV based MANET, the strategy deals with cooperative black hole by using a deceptive message with virtual address that is not in the addresses range [1]. The main interest in this paper beside proposing the new idea, is to convert this idea into strategy by providing the necessary details, in addition to implement the strategy by hardware and software models. These models have been built in order to illustrate how black hole works and how the strategy can detect and block the black hole practically in AODV based MANET. This paper presents the method in an integrated manner. In this method, the detection process starts from the previous node, not from the source node. However, this method can choose the shortest path and can deal with more than one black hole. These additions will preserve the performance of network, the hardware model has been provided with liquid crystal displays LCD put on each node. LCD is used to appear each step, and to make the system full-acknowledgment.","PeriodicalId":253950,"journal":{"name":"Mansoura Journal for Computer and Information Sciences","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132293138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.21608/mjcis.2018.312002
I. S. Farahat, A. Tolba, Waleed Eladrosy
Healthcare is the major problem that faces the individuals around the world especially oldsters and disables. The development of the object to object technology (Internet of things (IoT)) is improved until it can be used to overcome the healthcare problem. The progress of the IoT extends to help the science in remote healthcare and predict the disease of the patients before it happened. But with these improvements the patient data become in danger because the patient data are sent from patient side over the internet to the physician so it becomes available for any attackers to attack the data and modified or stolen it in its way to the doctor or to the patient. So the healthcare field is having a security and privacy issue. This paper introduces system that can solve this problem by changing the data before it leaves the patient. So the transfer data is secured enough to protect the patients' data. The proposed system using a three technique one for encoding data to change the shape of data and compressed it and one technique for encrypt encoded data with AES but with rotated key and last one for make an authentication mechanism to protect data from output access and permit any person that have username and password credential to access patient's data. The authentication mechanism is built at the website with a private IP the proposed system is developed with low cost hardware to minimize the cost of the product with high efficiency.
{"title":"Enhanced Security of the Internet of Medical Things (IOMT)","authors":"I. S. Farahat, A. Tolba, Waleed Eladrosy","doi":"10.21608/mjcis.2018.312002","DOIUrl":"https://doi.org/10.21608/mjcis.2018.312002","url":null,"abstract":"Healthcare is the major problem that faces the individuals around the world especially oldsters and disables. The development of the object to object technology (Internet of things (IoT)) is improved until it can be used to overcome the healthcare problem. The progress of the IoT extends to help the science in remote healthcare and predict the disease of the patients before it happened. But with these improvements the patient data become in danger because the patient data are sent from patient side over the internet to the physician so it becomes available for any attackers to attack the data and modified or stolen it in its way to the doctor or to the patient. So the healthcare field is having a security and privacy issue. This paper introduces system that can solve this problem by changing the data before it leaves the patient. So the transfer data is secured enough to protect the patients' data. The proposed system using a three technique one for encoding data to change the shape of data and compressed it and one technique for encrypt encoded data with AES but with rotated key and last one for make an authentication mechanism to protect data from output access and permit any person that have username and password credential to access patient's data. The authentication mechanism is built at the website with a private IP the proposed system is developed with low cost hardware to minimize the cost of the product with high efficiency.","PeriodicalId":253950,"journal":{"name":"Mansoura Journal for Computer and Information Sciences","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125450045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-12-01DOI: 10.21608/mjcis.2018.312006
Naira Elazab, N. Mekky, A. Atwan, Hassan H. Soliman
In this paper, the point is to give steady shadows amongst virtual and genuine objects for outdoor scene without requiring any earlier data. The technique gives a real-time framework for recreating reliable virtual shadows in an outdoor scene. The proposed procedure firstly handles the shadow areas because of genuine things; secondly ensures the pixels in genuine shadow are prevented from more rendering and finally rendering shadows because of virtual things. The main research contribution, in addition to shadow refinement, is handling the overlap between real and virtual shadows. Consequently those shadows for virtual things are depend on the shadows of real things. The experimental results create the impression that the recommended system has basically upgraded that authenticity about steady outside augmented reality rendering, in this manner taking care of the issue of practical augmented reality frameworks. Results show a comparison of shadow rendering between proposed technique and other techniques in different resolutions.
{"title":"An Enhanced Real-Time Shadow Rendering Technique in Outdoor Augmented Reality","authors":"Naira Elazab, N. Mekky, A. Atwan, Hassan H. Soliman","doi":"10.21608/mjcis.2018.312006","DOIUrl":"https://doi.org/10.21608/mjcis.2018.312006","url":null,"abstract":"In this paper, the point is to give steady shadows amongst virtual and genuine objects for outdoor scene without requiring any earlier data. The technique gives a real-time framework for recreating reliable virtual shadows in an outdoor scene. The proposed procedure firstly handles the shadow areas because of genuine things; secondly ensures the pixels in genuine shadow are prevented from more rendering and finally rendering shadows because of virtual things. The main research contribution, in addition to shadow refinement, is handling the overlap between real and virtual shadows. Consequently those shadows for virtual things are depend on the shadows of real things. The experimental results create the impression that the recommended system has basically upgraded that authenticity about steady outside augmented reality rendering, in this manner taking care of the issue of practical augmented reality frameworks. Results show a comparison of shadow rendering between proposed technique and other techniques in different resolutions.","PeriodicalId":253950,"journal":{"name":"Mansoura Journal for Computer and Information Sciences","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114070260","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.21608/mjcis.2018.311997
Fatma Faek, O. Nomir, Ehab Essa, Ebrahim El-henawy
A combination of wavelet transforms and contrast limited adaptive histogram equalization (CLAHE) techniques are used to efficiently enhance videos. The proposed technique handles the noises within video frames and enhances the resolution of the video. Lifting wavelet transform (LWT) and stationary wavelet transform (SWT) are applied to separate original frame into the low-frequency sub-bands, and the high-frequency sub-bands. Thereafter, we applied the interpolation to correct the coefficients of the high-frequency and the original frame separately. Next, Inverse Lifting wavelet transform (ILWT) is utilized for the integration of each all these enhanced sub-band. Finally, the CLAHE algorithm is applied to make the details of the frame more visible, and meaningful to generally improve the resolution of the video. The output video shows that the proposed technique enhances the quality of resolution videos under various environmental conditions, alleviates noises and avoids the over-enhancement problems.
{"title":"Wavelet-based Video Enhancement Using Contrast Limited Adaptive Histogram Equalization","authors":"Fatma Faek, O. Nomir, Ehab Essa, Ebrahim El-henawy","doi":"10.21608/mjcis.2018.311997","DOIUrl":"https://doi.org/10.21608/mjcis.2018.311997","url":null,"abstract":"A combination of wavelet transforms and contrast limited adaptive histogram equalization (CLAHE) techniques are used to efficiently enhance videos. The proposed technique handles the noises within video frames and enhances the resolution of the video. Lifting wavelet transform (LWT) and stationary wavelet transform (SWT) are applied to separate original frame into the low-frequency sub-bands, and the high-frequency sub-bands. Thereafter, we applied the interpolation to correct the coefficients of the high-frequency and the original frame separately. Next, Inverse Lifting wavelet transform (ILWT) is utilized for the integration of each all these enhanced sub-band. Finally, the CLAHE algorithm is applied to make the details of the frame more visible, and meaningful to generally improve the resolution of the video. The output video shows that the proposed technique enhances the quality of resolution videos under various environmental conditions, alleviates noises and avoids the over-enhancement problems.","PeriodicalId":253950,"journal":{"name":"Mansoura Journal for Computer and Information Sciences","volume":"290 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122822706","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.21608/mjcis.2018.311999
Ahmed T. Sharafeldeen, M. F. Alrahmawy, S. Elmougy
Rectangles for bipartite graphs are like triangles for unipartite graphs as both represent the smallest cycles in such graphs. Rectangle Counting is considered an important task in many bipartite network analysis metrics and is considered the core of computing such metrics, especially in cluster coefficient, bitruss, etc. However, there are few efficient algorithms to deal with this problem, especially in a large bipartite graph. In this work, we use MapReduce to enhance an algorithm to count rectangles in a large bipartite graph. The results show that our proposed MapReduce-based algorithm gives a better execution time than the existing algorithms, especially when it is applied in very large bipartite graphs.
{"title":"Enhanced Algorithms for Counting Rectangles in Large Bipartite Graphs using MapReduce","authors":"Ahmed T. Sharafeldeen, M. F. Alrahmawy, S. Elmougy","doi":"10.21608/mjcis.2018.311999","DOIUrl":"https://doi.org/10.21608/mjcis.2018.311999","url":null,"abstract":"Rectangles for bipartite graphs are like triangles for unipartite graphs as both represent the smallest cycles in such graphs. Rectangle Counting is considered an important task in many bipartite network analysis metrics and is considered the core of computing such metrics, especially in cluster coefficient, bitruss, etc. However, there are few efficient algorithms to deal with this problem, especially in a large bipartite graph. In this work, we use MapReduce to enhance an algorithm to count rectangles in a large bipartite graph. The results show that our proposed MapReduce-based algorithm gives a better execution time than the existing algorithms, especially when it is applied in very large bipartite graphs.","PeriodicalId":253950,"journal":{"name":"Mansoura Journal for Computer and Information Sciences","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127590441","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.21608/mjcis.2018.311994
Abeer Saber, Aya M. Al-Zoghby, S. Elmougy
Semantic web introduces new benefits for many research topics on big-data. It semantically maintains a large amount of data and provides meaningful meaning of unstructured data contents. Big data refers to large scale. It is used to describe a massive collection of datasets in different formats. The semantic and structural heterogeneity are the biggest problems that still face the aggregating, integrating, and storing big data. In this paper, we solved both of the problems of columns redundancy that are produced from the semantic heterogeneity and the problem of structural heterogeneity through developing and implementing a new ETL model based on semantic and ontology technologies. Geospatial data is used as a case study because its integration is complex and usually suffers from the variety of resources and the representation of the produced big data. The results of using this model showed that it solves the problem of heterogeneity in several data sources and it improves the data integration and representation.
{"title":"ETL Semantic Model for Big Data Aggregation, Integration, and Representation","authors":"Abeer Saber, Aya M. Al-Zoghby, S. Elmougy","doi":"10.21608/mjcis.2018.311994","DOIUrl":"https://doi.org/10.21608/mjcis.2018.311994","url":null,"abstract":"Semantic web introduces new benefits for many research topics on big-data. It semantically maintains a large amount of data and provides meaningful meaning of unstructured data contents. Big data refers to large scale. It is used to describe a massive collection of datasets in different formats. The semantic and structural heterogeneity are the biggest problems that still face the aggregating, integrating, and storing big data. In this paper, we solved both of the problems of columns redundancy that are produced from the semantic heterogeneity and the problem of structural heterogeneity through developing and implementing a new ETL model based on semantic and ontology technologies. Geospatial data is used as a case study because its integration is complex and usually suffers from the variety of resources and the representation of the produced big data. The results of using this model showed that it solves the problem of heterogeneity in several data sources and it improves the data integration and representation.","PeriodicalId":253950,"journal":{"name":"Mansoura Journal for Computer and Information Sciences","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132638878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.21608/mjcis.2018.311995
Muhammad H. Zayyan, S. Elmougy, M. F. Alrahmawy
In this paper, we recognize objects within images by collecting information from a large number of random-size patches of the image. The different backgrounds accompany the foreground object demand to have a learning system to identify each patch as belonging to the object category or to the background category. We strengthen a recent method called Evolution-COnstructed (ECO), which is based on the ensemble learning approach which combines several weak classifier. The improvement is relying on decreasing the overfitting problem. Two different improving ideas are proposed: 1) Pooling operation, which is applied to the weak classifiers data, 2) Random Forest algorithm, which combines the weak classifiers outcomes. Experimental results are reported for classification of 9 categories of Caltech-101 data sets and proved that our modifications boost the performance over the base method and other existing methods.
{"title":"Representation Learning Framework of Object Recognition via Feature Construction","authors":"Muhammad H. Zayyan, S. Elmougy, M. F. Alrahmawy","doi":"10.21608/mjcis.2018.311995","DOIUrl":"https://doi.org/10.21608/mjcis.2018.311995","url":null,"abstract":"In this paper, we recognize objects within images by collecting information from a large number of random-size patches of the image. The different backgrounds accompany the foreground object demand to have a learning system to identify each patch as belonging to the object category or to the background category. We strengthen a recent method called Evolution-COnstructed (ECO), which is based on the ensemble learning approach which combines several weak classifier. The improvement is relying on decreasing the overfitting problem. Two different improving ideas are proposed: 1) Pooling operation, which is applied to the weak classifiers data, 2) Random Forest algorithm, which combines the weak classifiers outcomes. Experimental results are reported for classification of 9 categories of Caltech-101 data sets and proved that our modifications boost the performance over the base method and other existing methods.","PeriodicalId":253950,"journal":{"name":"Mansoura Journal for Computer and Information Sciences","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125141720","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：