Pub Date : 2017-11-01DOI: 10.1109/CSNT.2017.8418532
Shabina Sayed, Shoeb Ahmed, R. Poonia
The breast cancer diagnostic and prognostic problems are mainly in the scope of the widely discussed classification problems. These problems have attracted many researchers in computational intelligence, data mining, and statistics fields. The objective of these predictions is to handle cases for which cancer has not recurred (censored data) as well as case for which cancer has recurred at a specific time. The proposed study uses Breast Cancer Wisconsin (Prognostic) Data Set for training and testing purpose. It has implemented holo entropy enable decision tree(HDT). The proposed strategy utilizes the training data to train the classifier. It categorizes each instance of breast cancer growth as recurrent or non recurrent. It ascertains the precision of the classifier to decide the exact classifier accuracy. In the present situation where there is continuous increment in the breast cancer cases and the expanding number of death cases the proposed strategy can be a guide in the determination of breast cancer.
{"title":"Holo entropy enabled decision tree classifier for breast cancer diagnosis using wisconsin (prognostic) data set","authors":"Shabina Sayed, Shoeb Ahmed, R. Poonia","doi":"10.1109/CSNT.2017.8418532","DOIUrl":"https://doi.org/10.1109/CSNT.2017.8418532","url":null,"abstract":"The breast cancer diagnostic and prognostic problems are mainly in the scope of the widely discussed classification problems. These problems have attracted many researchers in computational intelligence, data mining, and statistics fields. The objective of these predictions is to handle cases for which cancer has not recurred (censored data) as well as case for which cancer has recurred at a specific time. The proposed study uses Breast Cancer Wisconsin (Prognostic) Data Set for training and testing purpose. It has implemented holo entropy enable decision tree(HDT). The proposed strategy utilizes the training data to train the classifier. It categorizes each instance of breast cancer growth as recurrent or non recurrent. It ascertains the precision of the classifier to decide the exact classifier accuracy. In the present situation where there is continuous increment in the breast cancer cases and the expanding number of death cases the proposed strategy can be a guide in the determination of breast cancer.","PeriodicalId":382417,"journal":{"name":"2017 7th International Conference on Communication Systems and Network Technologies (CSNT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126734476","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 : 2017-11-01DOI: 10.1109/CSNT.2017.8418562
Girdhar Gopal Ladha, R. S. Pippal
According to international diabetes federation one in eleven adults have diabetes in 2015 worldwide. Their report also suggest that it may increase by one in ten adults has diabetes by 2040 worldwide. It is dangerous as in maximum chances it is diagnosed in the higher stages. The main aim of this paper is to find the methodologies which can help in early stage diagnosis. This paper provides a detail study and analysis on data mining techniques. A deep analysis has been presented for finding the gaps and also suggested the enhancements in terms of future suggestions. It also provides methodological review along with the impact of diabetes and in the direction of their symptoms identification.
{"title":"A review and analysis on data mining methods to predict diabetes","authors":"Girdhar Gopal Ladha, R. S. Pippal","doi":"10.1109/CSNT.2017.8418562","DOIUrl":"https://doi.org/10.1109/CSNT.2017.8418562","url":null,"abstract":"According to international diabetes federation one in eleven adults have diabetes in 2015 worldwide. Their report also suggest that it may increase by one in ten adults has diabetes by 2040 worldwide. It is dangerous as in maximum chances it is diagnosed in the higher stages. The main aim of this paper is to find the methodologies which can help in early stage diagnosis. This paper provides a detail study and analysis on data mining techniques. A deep analysis has been presented for finding the gaps and also suggested the enhancements in terms of future suggestions. It also provides methodological review along with the impact of diabetes and in the direction of their symptoms identification.","PeriodicalId":382417,"journal":{"name":"2017 7th International Conference on Communication Systems and Network Technologies (CSNT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123079478","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}
This paper presents the DFDM (Dynamic Frequency Division Multiplexing) system, a dynamic digital multi-carrier transceiver concept. The system aims at exploiting the spectrum fragmentation in widely separated spectrum bands, which is due to guard bands provided to channels and random spectrum allocations. The system implements dynamic spectrum exploitation of the white spaces in UHF (Ultra High Frequency) band and also below noise floor communication links.
本文提出了动态数字多载波收发器DFDM (Dynamic Frequency Division Multiplexing)系统。该系统旨在利用由于向信道提供保护带和随机分配频谱而产生的宽频带内的频谱碎片。该系统实现了对超高频(UHF)频段和噪声底以下通信链路空白频段的动态频谱利用。
{"title":"DFDM — Dynamic frequency division multiplexing","authors":"Muralidhar Reddy Challa, Bharath Simha Reddy Eedula, Gnana Pavan Bombothu, Ram Mohan Rao Kanugu","doi":"10.1109/CSNT.2017.8418506","DOIUrl":"https://doi.org/10.1109/CSNT.2017.8418506","url":null,"abstract":"This paper presents the DFDM (Dynamic Frequency Division Multiplexing) system, a dynamic digital multi-carrier transceiver concept. The system aims at exploiting the spectrum fragmentation in widely separated spectrum bands, which is due to guard bands provided to channels and random spectrum allocations. The system implements dynamic spectrum exploitation of the white spaces in UHF (Ultra High Frequency) band and also below noise floor communication links.","PeriodicalId":382417,"journal":{"name":"2017 7th International Conference on Communication Systems and Network Technologies (CSNT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132511594","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}
Most of the electronic circuit components do not receive the clock at same time due to various factors involved in circuitry. Phase locked loop is a precision and familiar circuit for high frequency and high accuracy application with very short interlocking time. This paper presents All Digital Phase Locked Loop (ADPLL) and has been analysed for the required applications on the basis of its cost, power consumption and speed of operation for phase locked loop. In the given ADPLL system phase detection system has been realized by generating analytic signal using Hilbert transform and then computing the instantaneous phase using CORDIC algorithm. The loop filter of the ADPLL has been designed using a low pass filter and is used to discard the higher order harmonics. The proposed architecture is implemented using VHDL code and is synthesized using Xilinx ISE 9.2 software. To validate its functionality, verification and simulation is done by using the Modelsim SE 6.2C. The ADPLL is planned for 100 MHz central frequency. The work in this paper mainly deals with the power efficiency of ADPLL.
由于电路中涉及的各种因素,大多数电子电路元件不能同时接收时钟。锁相环是一种高频、高精度、联锁时间极短的精密电路。本文介绍了全数字锁相环(ADPLL),并在其成本、功耗和锁相环运行速度的基础上,对其应用进行了分析。在给定的ADPLL系统中,相位检测系统是通过希尔伯特变换产生解析信号,然后用CORDIC算法计算瞬时相位来实现的。ADPLL的环路滤波器采用低通滤波器设计,用于去除高次谐波。所提出的架构使用VHDL代码实现,并使用Xilinx ISE 9.2软件进行综合。为了验证其功能,可以使用Modelsim SE 6.2C进行验证和仿真。ADPLL的中心频率为100mhz。本文主要研究了ADPLL的功率效率问题。
{"title":"FPGA implementation and power efficient CORDIC based ADPLL for signal processing and application","authors":"Akarshika Singhal, Anjana Goen, Tanutrushna Mohapatra","doi":"10.1109/CSNT.2017.8418560","DOIUrl":"https://doi.org/10.1109/CSNT.2017.8418560","url":null,"abstract":"Most of the electronic circuit components do not receive the clock at same time due to various factors involved in circuitry. Phase locked loop is a precision and familiar circuit for high frequency and high accuracy application with very short interlocking time. This paper presents All Digital Phase Locked Loop (ADPLL) and has been analysed for the required applications on the basis of its cost, power consumption and speed of operation for phase locked loop. In the given ADPLL system phase detection system has been realized by generating analytic signal using Hilbert transform and then computing the instantaneous phase using CORDIC algorithm. The loop filter of the ADPLL has been designed using a low pass filter and is used to discard the higher order harmonics. The proposed architecture is implemented using VHDL code and is synthesized using Xilinx ISE 9.2 software. To validate its functionality, verification and simulation is done by using the Modelsim SE 6.2C. The ADPLL is planned for 100 MHz central frequency. The work in this paper mainly deals with the power efficiency of ADPLL.","PeriodicalId":382417,"journal":{"name":"2017 7th International Conference on Communication Systems and Network Technologies (CSNT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134212893","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 : 2017-11-01DOI: 10.1109/CSNT.2017.8418500
Komal Verma, R. Pandey, Nidhi Srivastava
Experts from different specialized fields trust that another era of Computer Science is emerging. Sometime ago, applications use to run on the individual PC framework's hard drive. With taking a break, the application started to be put away on a nearby server of an organization and from that point the applications were keep running on the individual framework. The following stage to this advancement of software engineering time is moving both the applications and information to the ‘Cloud’, where all information is put away at a virtual server and can be gotten to from anyplace through Internet. Microsoft Azure is a Microsoft's open Cloud Computing Platform. It gives an extensive variety of cloud administrations from Data Storage, Analysis to Computing and Networking of Data. Analysis through Machine learning is one of the administrations that Azure Offers. For dissecting one can use programming languages like R, Python and Visual Studio. R gives a Programming domain broadly utilized for graphically analysis and measurable computing.
{"title":"Inventory supply & shelfing through data analytics","authors":"Komal Verma, R. Pandey, Nidhi Srivastava","doi":"10.1109/CSNT.2017.8418500","DOIUrl":"https://doi.org/10.1109/CSNT.2017.8418500","url":null,"abstract":"Experts from different specialized fields trust that another era of Computer Science is emerging. Sometime ago, applications use to run on the individual PC framework's hard drive. With taking a break, the application started to be put away on a nearby server of an organization and from that point the applications were keep running on the individual framework. The following stage to this advancement of software engineering time is moving both the applications and information to the ‘Cloud’, where all information is put away at a virtual server and can be gotten to from anyplace through Internet. Microsoft Azure is a Microsoft's open Cloud Computing Platform. It gives an extensive variety of cloud administrations from Data Storage, Analysis to Computing and Networking of Data. Analysis through Machine learning is one of the administrations that Azure Offers. For dissecting one can use programming languages like R, Python and Visual Studio. R gives a Programming domain broadly utilized for graphically analysis and measurable computing.","PeriodicalId":382417,"journal":{"name":"2017 7th International Conference on Communication Systems and Network Technologies (CSNT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133490292","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 : 2017-11-01DOI: 10.1109/CSNT.2017.8418526
Prateeksha Dhantre, R. Prasad, P. Saurabh, B. Verma
Human skin detection strives to spot skin from the pictures. Automatic skin detection is considered as a significantly difficult and complex as skin image differ on the aspects of contents due to variation in size, style, orientation, alignment coupled with different contrast and background. This paper proposes a skin detection approach using localization, tracking, extraction, enhancement, and recognition. This approach remains sensitive to the color palette and uses edge detection technique. Also, color classification box incorporates a deep impact on the performance of the rule. The proposed approach detects single as well as multiple persons in a picture. Promising results are obtained on variety of pictures, except in few pictures wherever color distinction is hard to even when edge detection rule.
{"title":"A hybrid approach for human skin detection","authors":"Prateeksha Dhantre, R. Prasad, P. Saurabh, B. Verma","doi":"10.1109/CSNT.2017.8418526","DOIUrl":"https://doi.org/10.1109/CSNT.2017.8418526","url":null,"abstract":"Human skin detection strives to spot skin from the pictures. Automatic skin detection is considered as a significantly difficult and complex as skin image differ on the aspects of contents due to variation in size, style, orientation, alignment coupled with different contrast and background. This paper proposes a skin detection approach using localization, tracking, extraction, enhancement, and recognition. This approach remains sensitive to the color palette and uses edge detection technique. Also, color classification box incorporates a deep impact on the performance of the rule. The proposed approach detects single as well as multiple persons in a picture. Promising results are obtained on variety of pictures, except in few pictures wherever color distinction is hard to even when edge detection rule.","PeriodicalId":382417,"journal":{"name":"2017 7th International Conference on Communication Systems and Network Technologies (CSNT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114223959","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 : 2017-11-01DOI: 10.1109/CSNT.2017.8418533
K. Vamsi, Raman Chadha, B. Ramkumar, S. Prasad
Nowadays generation moves upon, digital forgeries also increasing with new trending tools for general concerns illegally. Moreover, applications are used for morphing/tampering an image to judge the world's computation. Spliced Location of any images, we pinpointed a probable approach to grab the forgery section easily and clearly. The approaches used are Super-pixels identification, Discrete Cosine Transform, Scale-invariant feature transform along with Kurtosis mapping, passive/blind forgery assumes a worthy part to search for spliced images without certain information which increases the execution of retrieval of duplicity image and consumption of time. In this proposed methodology, the controlled mechanism for "n" iteration is calculated with the help of estimation local noise variance algorithm. Approach narrates the splicing methodology in consign way to Speculate the loop-hole detection mechanism i.e., Gives information about a traced image spliced area for verification.
{"title":"Image splicing detection using HMRF superpixel segmentation","authors":"K. Vamsi, Raman Chadha, B. Ramkumar, S. Prasad","doi":"10.1109/CSNT.2017.8418533","DOIUrl":"https://doi.org/10.1109/CSNT.2017.8418533","url":null,"abstract":"Nowadays generation moves upon, digital forgeries also increasing with new trending tools for general concerns illegally. Moreover, applications are used for morphing/tampering an image to judge the world's computation. Spliced Location of any images, we pinpointed a probable approach to grab the forgery section easily and clearly. The approaches used are Super-pixels identification, Discrete Cosine Transform, Scale-invariant feature transform along with Kurtosis mapping, passive/blind forgery assumes a worthy part to search for spliced images without certain information which increases the execution of retrieval of duplicity image and consumption of time. In this proposed methodology, the controlled mechanism for \"n\" iteration is calculated with the help of estimation local noise variance algorithm. Approach narrates the splicing methodology in consign way to Speculate the loop-hole detection mechanism i.e., Gives information about a traced image spliced area for verification.","PeriodicalId":382417,"journal":{"name":"2017 7th International Conference on Communication Systems and Network Technologies (CSNT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121315945","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 : 2017-11-01DOI: 10.1109/CSNT.2017.8418513
Kasula Chaithanya Pramodh, Iluri Nikhil, J. Ranjith Singh
The present paper focuses on developing an SNMP-JSON translator which automatically translates the data in an MIB file written in ASN.1 to JSON. Further, we integrate the SNMP Agents, Managers and implement the standard SNMP operations by using sockets of TCP/IP suite in a client-server architecture. The result is an integrated Network Management System which performs better than the existing XML and HTTP based approach.
{"title":"Implementation of SNMP-JSON translator and integrating SNMP agents with JSON based network management system","authors":"Kasula Chaithanya Pramodh, Iluri Nikhil, J. Ranjith Singh","doi":"10.1109/CSNT.2017.8418513","DOIUrl":"https://doi.org/10.1109/CSNT.2017.8418513","url":null,"abstract":"The present paper focuses on developing an SNMP-JSON translator which automatically translates the data in an MIB file written in ASN.1 to JSON. Further, we integrate the SNMP Agents, Managers and implement the standard SNMP operations by using sockets of TCP/IP suite in a client-server architecture. The result is an integrated Network Management System which performs better than the existing XML and HTTP based approach.","PeriodicalId":382417,"journal":{"name":"2017 7th International Conference on Communication Systems and Network Technologies (CSNT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128977184","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 : 2017-11-01DOI: 10.1109/CSNT.2017.8418565
P. Anitha, C. Rao, S. Babu
Email spam is a kind of electronic spam, which tends to be a more difficult problem nowadays among all internet challenges. Spam mails are mostly sent in commercial purpose, some of them may contain malware links that lead to phishing websites. The aim of this study is to classify into ham and spam emails with an optimized and well efficient classification technique. Ham holds emails that are legitimate or legally valid message can get accepted by users. Spam emails are unwanted emails that a user doesn't want and to get rid of it. This study emphasizes on the improvement in classifying all mails into these two groups with minimal requirement of training and with an accuracy of hundred percent. Here in this study, Modified Naïve Bayes (MNB) classifier ensured the requirements with very low percentage of training and produces accurate results than existing Naïve Bayes (NB) or Supporting Vector Machine (SVM) classifier.
{"title":"Email spam classification using neighbor probability based Naïve Bayes algorithm","authors":"P. Anitha, C. Rao, S. Babu","doi":"10.1109/CSNT.2017.8418565","DOIUrl":"https://doi.org/10.1109/CSNT.2017.8418565","url":null,"abstract":"Email spam is a kind of electronic spam, which tends to be a more difficult problem nowadays among all internet challenges. Spam mails are mostly sent in commercial purpose, some of them may contain malware links that lead to phishing websites. The aim of this study is to classify into ham and spam emails with an optimized and well efficient classification technique. Ham holds emails that are legitimate or legally valid message can get accepted by users. Spam emails are unwanted emails that a user doesn't want and to get rid of it. This study emphasizes on the improvement in classifying all mails into these two groups with minimal requirement of training and with an accuracy of hundred percent. Here in this study, Modified Naïve Bayes (MNB) classifier ensured the requirements with very low percentage of training and produces accurate results than existing Naïve Bayes (NB) or Supporting Vector Machine (SVM) classifier.","PeriodicalId":382417,"journal":{"name":"2017 7th International Conference on Communication Systems and Network Technologies (CSNT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128405253","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 : 2017-11-01DOI: 10.1109/CSNT.2017.8418525
Amisha Kumari, Urjita Thakar
Classification is a popular technique used to predict group membership for data samples in datasets. A multi-class or multinomial classification is the problem of classifying instances into more than two classes. With the emerging technology, the complexity of multi-class data has also increased thereby leading to class imbalance problem. With an imbalanced dataset, a machine learning algorithm can not make an accurate prediction. Therefore, in this paper Hellinger distance based oversampling method has been proposed. It is useful in balancing the datasets so that minority class can be identified with high accuracy without affecting accuracy of majority class. New synthetic data is generated using this method to achieve balance ratio. Testing has been done on five benchmark datasets using two standard classifiers KNN and C4.5. The evaluation matrix on precision, recall and fmeasure are drawn for two standard classification algorithms. It is observed that Hellinger distance reduces risk of overlapping and skewness of data. Obtained results show increase of 20% in classification accuracy compared to classification of imbalance multi-class dataset.
{"title":"Hellinger distance based oversampling method to solve multi-class imbalance problem","authors":"Amisha Kumari, Urjita Thakar","doi":"10.1109/CSNT.2017.8418525","DOIUrl":"https://doi.org/10.1109/CSNT.2017.8418525","url":null,"abstract":"Classification is a popular technique used to predict group membership for data samples in datasets. A multi-class or multinomial classification is the problem of classifying instances into more than two classes. With the emerging technology, the complexity of multi-class data has also increased thereby leading to class imbalance problem. With an imbalanced dataset, a machine learning algorithm can not make an accurate prediction. Therefore, in this paper Hellinger distance based oversampling method has been proposed. It is useful in balancing the datasets so that minority class can be identified with high accuracy without affecting accuracy of majority class. New synthetic data is generated using this method to achieve balance ratio. Testing has been done on five benchmark datasets using two standard classifiers KNN and C4.5. The evaluation matrix on precision, recall and fmeasure are drawn for two standard classification algorithms. It is observed that Hellinger distance reduces risk of overlapping and skewness of data. Obtained results show increase of 20% in classification accuracy compared to classification of imbalance multi-class dataset.","PeriodicalId":382417,"journal":{"name":"2017 7th International Conference on Communication Systems and Network Technologies (CSNT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126450880","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}