Similar languages with massive parallel corpora are readily implemented by large-scale systems using either Statistical Machine Translation (SMT) or Neural Machine Translation (NMT). Translations involving low-resource language pairs with linguistic divergence have always been a challenge. We consider one such pair, English-Manipuri, which shows linguistic divergence and belongs to the low resource category. For such language pairs, SMT gets better acclamation than NMT. However, SMT’s more prominent phrase- based model uses groupings of surface word forms treated as phrases for translation. Therefore, without any linguistic knowledge, it fails to learn a proper mapping between the source and target language symbols. Our model adopts a factored model of SMT (FSMT3*) with a part-of-speech (POS) tag as a factor to incorporate linguistic information about the languages followed by hand-coded reordering. The reordering of source sentences makes them similar to the target language allowing better mapping between source and target symbols. The reordering also converts long-distance reordering problems to monotone reordering that SMT models can better handle, thereby reducing the load during decoding time. Additionally, we discover that adding a POS feature data enhances the system’s precision. Experimental results using automatic evaluation metrics show that our model improved over phrase-based and other factored models using the lexicalised Moses reordering options. Our FSMT3* model shows an increase in the automatic scores of translation result over the factored model with lexicalised phrase reordering (FSMT2) by an amount of 11.05% (Bilingual Evaluation Understudy), 5.46% (F1), 9.35% (Precision), and 2.56% (Recall), respectively.
{"title":"Reordering of Source Side for a Factored English to Manipuri SMT System","authors":"Indika Maibam, Bipul Syam Purkayastha","doi":"10.32985/ijeces.14.3.6","DOIUrl":"https://doi.org/10.32985/ijeces.14.3.6","url":null,"abstract":"Similar languages with massive parallel corpora are readily implemented by large-scale systems using either Statistical Machine Translation (SMT) or Neural Machine Translation (NMT). Translations involving low-resource language pairs with linguistic divergence have always been a challenge. We consider one such pair, English-Manipuri, which shows linguistic divergence and belongs to the low resource category. For such language pairs, SMT gets better acclamation than NMT. However, SMT’s more prominent phrase- based model uses groupings of surface word forms treated as phrases for translation. Therefore, without any linguistic knowledge, it fails to learn a proper mapping between the source and target language symbols. Our model adopts a factored model of SMT (FSMT3*) with a part-of-speech (POS) tag as a factor to incorporate linguistic information about the languages followed by hand-coded reordering. The reordering of source sentences makes them similar to the target language allowing better mapping between source and target symbols. The reordering also converts long-distance reordering problems to monotone reordering that SMT models can better handle, thereby reducing the load during decoding time. Additionally, we discover that adding a POS feature data enhances the system’s precision. Experimental results using automatic evaluation metrics show that our model improved over phrase-based and other factored models using the lexicalised Moses reordering options. Our FSMT3* model shows an increase in the automatic scores of translation result over the factored model with lexicalised phrase reordering (FSMT2) by an amount of 11.05% (Bilingual Evaluation Understudy), 5.46% (F1), 9.35% (Precision), and 2.56% (Recall), respectively.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":" ","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48980913","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}
Next generation communication systems require better performance to support high - bandwidth, peak data rate, spectral efficiency, mobility, connection density, positioning accuracy, etc. Investigation on efficient modulation technique for next generation has become very important so as to meet its expectations. In this paper performance optimization of universal filtered multicarrier (UFMC) technique for next generation communication systems have been investigated. Dolph-Chebyshev (DC) and Kaiser-Bessel- derived (KBD) filters have been used to optimize power spectral density, channel equalization, bit error rate, and peak to average power ratio (PAPR). It has been observed that KBD filter response is comparatively better than DC filter. Effect of filter length also influences the system performance, filter with bigger length improves performance at the cost of computational complexity. Performance of UFMC has been compared with that of orthogonal frequency division multiplexing (OFDM) technique. The present work of investigations on UFMC that is based on subband filtering is our original research work that has been carried out for its suitability for next generation communication systems. It has simple design structure, lower computational complexities and better performance in terms of BER compared to OFDM and f-ODFM systems. It has comparatively low PAPR than GFDM and FBMC techniques.
{"title":"Performance Optimization of Universal Filtered Multicarrier Technique for Next Generation Communication Systems","authors":"Shatrughna Prasad Yadav","doi":"10.32985/ijeces.14.2.1","DOIUrl":"https://doi.org/10.32985/ijeces.14.2.1","url":null,"abstract":"Next generation communication systems require better performance to support high - bandwidth, peak data rate, spectral efficiency, mobility, connection density, positioning accuracy, etc. Investigation on efficient modulation technique for next generation has become very important so as to meet its expectations. In this paper performance optimization of universal filtered multicarrier (UFMC) technique for next generation communication systems have been investigated. Dolph-Chebyshev (DC) and Kaiser-Bessel- derived (KBD) filters have been used to optimize power spectral density, channel equalization, bit error rate, and peak to average power ratio (PAPR). It has been observed that KBD filter response is comparatively better than DC filter. Effect of filter length also influences the system performance, filter with bigger length improves performance at the cost of computational complexity. Performance of UFMC has been compared with that of orthogonal frequency division multiplexing (OFDM) technique. The present work of investigations on UFMC that is based on subband filtering is our original research work that has been carried out for its suitability for next generation communication systems. It has simple design structure, lower computational complexities and better performance in terms of BER compared to OFDM and f-ODFM systems. It has comparatively low PAPR than GFDM and FBMC techniques.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":" ","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45955895","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}
Atul O. Thakare, Omprakash W. Tembhurne, Abhijeet R. Thakare, Soora Narasimha Reddy
Because of the massive utilization of the world wide web and the drastic use of electronic gadgets to access the online world, there is an exponential growth in the information produced by these hardware gadgets. The data produced by different sources, such as smart transportation, healthcare, and e-commerce, are large, complex, and heterogeneous. Therefore, storing and querying this data, coined "Big Data," is challenging. This paper compares relational databases with a few of the popular NoSQL databases. The performance of various databases in executing join queries, filter queries, and aggregate queries on large datasets are compared on a single node and multinode clusters. The experimental results demonstrate the suitability of NoSQL databases for Big Data Analytics and for supporting large userbase interactive web applications.
{"title":"NoSQL Databases: Modern Data Systems for Big Data Analytics","authors":"Atul O. Thakare, Omprakash W. Tembhurne, Abhijeet R. Thakare, Soora Narasimha Reddy","doi":"10.32985/ijeces.14.2.10","DOIUrl":"https://doi.org/10.32985/ijeces.14.2.10","url":null,"abstract":"Because of the massive utilization of the world wide web and the drastic use of electronic gadgets to access the online world, there is an exponential growth in the information produced by these hardware gadgets. The data produced by different sources, such as smart transportation, healthcare, and e-commerce, are large, complex, and heterogeneous. Therefore, storing and querying this data, coined \"Big Data,\" is challenging. This paper compares relational databases with a few of the popular NoSQL databases. The performance of various databases in executing join queries, filter queries, and aggregate queries on large datasets are compared on a single node and multinode clusters. The experimental results demonstrate the suitability of NoSQL databases for Big Data Analytics and for supporting large userbase interactive web applications.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":" ","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41637079","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}
MANET (Mobile ad hoc networks) are famous in research due to their ad hoc nature and effectiveness during calamities across continents when no framework support is free. Wireless network interfaces have a limited transmission range; nodes may require multiple network hops to trade information across the organization. Each versatile node functions like a switch in such an organization, sending details to the other portable connected nodes. The nodes should not interrupt communication and associate themselves with the correct information transfer. Another significant issue was the development of expandable route discoveries capable of assessing rapid topography variations and numerous network detachments caused by high vehicle quality. This research article describes extensive technological changes, including the components and flaws of current progressive routing algorithms. Routing protocols designed for wired networks, such as the distance vector or connection state conventions, are inadequate for this application because they assume fixed geography and high overheads. This research article includes the MANET-supported routing protocols and their performance analysis across various performance parameters such as packet delivery ratio, average throughput, residual energy, and delay.
{"title":"Comparative Study and Performance Analysis of MANET Routing Protocol","authors":"Chetana Hemant Nemade, U. Pujeri","doi":"10.32985/ijeces.14.2.4","DOIUrl":"https://doi.org/10.32985/ijeces.14.2.4","url":null,"abstract":"MANET (Mobile ad hoc networks) are famous in research due to their ad hoc nature and effectiveness during calamities across continents when no framework support is free. Wireless network interfaces have a limited transmission range; nodes may require multiple network hops to trade information across the organization. Each versatile node functions like a switch in such an organization, sending details to the other portable connected nodes. The nodes should not interrupt communication and associate themselves with the correct information transfer. Another significant issue was the development of expandable route discoveries capable of assessing rapid topography variations and numerous network detachments caused by high vehicle quality. This research article describes extensive technological changes, including the components and flaws of current progressive routing algorithms. Routing protocols designed for wired networks, such as the distance vector or connection state conventions, are inadequate for this application because they assume fixed geography and high overheads. This research article includes the MANET-supported routing protocols and their performance analysis across various performance parameters such as packet delivery ratio, average throughput, residual energy, and delay.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":" ","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48206169","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}
Amier Hafizun Ab Rashid, Badrul Hisham Ahmad, Mohamad Zoinol Abidin Abd Aziz, N. Hassan
Nowadays, the compact and multiband antennas are typically required for personal communication devices in the continuously developing wireless communication industry. The fractal antenna with the third iteration of the pentagonal Sierpinski gasket island for WiMAX and WLAN applications is presented in this work. It starts with the fundamental design of the square patch (Antenna A1) and pentagonal patch (Antenna A2). This simulation work is done using CST Microwave Studio simulation studio by applying the concept of the zero, first, second and third fractal iteration. Then, it goes on to use the fractal geometry concept of the Sierpinski gasket island structure with three designs step. The designs consist of the first iteration (Antenna B1), second iteration (Antenna B2) and third iteration (Antenna B3) of fractal geometry. The simulation work of Antenna B3 is compared with the fabrication work of the same design. After that, the measurement of the Antenna B3 is done in laboratory with - 29.55 dB at 3.41 GHz and - 20.40 dB at 5.28 GHz for its operating frequencies with bandwidth of 3.52 GHz and 5.48 GHz, respectively. At targeting 3.5 GHz WiMAX, 5.2 GHz WLAN application and 7.24 GHz of Antenna B3, the antenna shows the – 17.78 dB, - 29.63 dB and – 22.73 dB, respectively, and this value is feasible for WiMAX and WLAN operation.
{"title":"CPW Fractal Antenna with Third Iteration of Pentagonal Sierpinski Gasket Island for 3.5 GHz WiMAX and 5.2 GHz WLAN Applications","authors":"Amier Hafizun Ab Rashid, Badrul Hisham Ahmad, Mohamad Zoinol Abidin Abd Aziz, N. Hassan","doi":"10.32985/ijeces.14.2.2","DOIUrl":"https://doi.org/10.32985/ijeces.14.2.2","url":null,"abstract":"Nowadays, the compact and multiband antennas are typically required for personal communication devices in the continuously developing wireless communication industry. The fractal antenna with the third iteration of the pentagonal Sierpinski gasket island for WiMAX and WLAN applications is presented in this work. It starts with the fundamental design of the square patch (Antenna A1) and pentagonal patch (Antenna A2). This simulation work is done using CST Microwave Studio simulation studio by applying the concept of the zero, first, second and third fractal iteration. Then, it goes on to use the fractal geometry concept of the Sierpinski gasket island structure with three designs step. The designs consist of the first iteration (Antenna B1), second iteration (Antenna B2) and third iteration (Antenna B3) of fractal geometry. The simulation work of Antenna B3 is compared with the fabrication work of the same design. After that, the measurement of the Antenna B3 is done in laboratory with - 29.55 dB at 3.41 GHz and - 20.40 dB at 5.28 GHz for its operating frequencies with bandwidth of 3.52 GHz and 5.48 GHz, respectively. At targeting 3.5 GHz WiMAX, 5.2 GHz WLAN application and 7.24 GHz of Antenna B3, the antenna shows the – 17.78 dB, - 29.63 dB and – 22.73 dB, respectively, and this value is feasible for WiMAX and WLAN operation.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":" ","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43890483","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}
In the online education field, Massive open online courses (MOOCs) have become popular in recent years. Educational institutions and Universities provide a variety of specialized online courses that helps the students to adapt with various needs and learning preferences. Because of this, institutional repositories creates and preserve a lot of data about students' demographics, behavioral trends, and academic achievement every day. Moreover, a significant problem impeding their future advancement is the high dropout rate. For solving this problem, the dropout rate is predicted by proposing an Ensemble Deep Learning Network (EDLN) model depending on the behavior data characteristics of learners. The local features are extracted by using ResNet-50 and then a kernel strategy is used for building feature relations. After feature extraction, the high-dimensional vector features are sent to a Faster RCNN for obtaining the vector representation that incorporates time series data. Then an attention weight is obtained for each dimension by applying a static attention mechanism to the vector. Extensive experiments on a public data set have shown that the proposed model can achieve comparable results with other dropout prediction methods in terms of precision, recall, F1 score, and accuracy.
{"title":"Ensemble Deep Learning Network Model for Dropout Prediction in MOOCs","authors":"G. Kumar, Amarkant Singh, Ashok Sharma","doi":"10.32985/ijeces.14.2.8","DOIUrl":"https://doi.org/10.32985/ijeces.14.2.8","url":null,"abstract":"In the online education field, Massive open online courses (MOOCs) have become popular in recent years. Educational institutions and Universities provide a variety of specialized online courses that helps the students to adapt with various needs and learning preferences. Because of this, institutional repositories creates and preserve a lot of data about students' demographics, behavioral trends, and academic achievement every day. Moreover, a significant problem impeding their future advancement is the high dropout rate. For solving this problem, the dropout rate is predicted by proposing an Ensemble Deep Learning Network (EDLN) model depending on the behavior data characteristics of learners. The local features are extracted by using ResNet-50 and then a kernel strategy is used for building feature relations. After feature extraction, the high-dimensional vector features are sent to a Faster RCNN for obtaining the vector representation that incorporates time series data. Then an attention weight is obtained for each dimension by applying a static attention mechanism to the vector. Extensive experiments on a public data set have shown that the proposed model can achieve comparable results with other dropout prediction methods in terms of precision, recall, F1 score, and accuracy.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":" ","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44051765","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}
B. V. Ravisankar Devarakonda, Venkateswararao Nandanavam
Spectrum sensing is one of the key tasks of cognitive radio to monitor the activity of the primary user. The sensing accuracy of the secondary user is dependent on the signal-to-noise ratio of the primary user signal. A novel Multi-head Attention-based spectrum sensing for Cognitive Radio is proposed through this work to increase the detection probability of the primary user at a low signal- to-noise ratio condition. A radio machine learning dataset with a variety of digital modulation schemes and varying signal-to-noise ratios served as a training source for the proposed model. Further, the performance metrics were evaluated to assess the performance of the proposed model. The experimental results indicate that the proposed model is optimized in terms of the amount of training time required which also has an increase of 27.6% in the probability of detection of the primary user under a low signal-to-noise ratio when compared to other related works that use deep learning.
{"title":"Multi-Head Attention-Based Spectrum Sensing for Cognitive Radio","authors":"B. V. Ravisankar Devarakonda, Venkateswararao Nandanavam","doi":"10.32985/ijeces.14.2.3","DOIUrl":"https://doi.org/10.32985/ijeces.14.2.3","url":null,"abstract":"Spectrum sensing is one of the key tasks of cognitive radio to monitor the activity of the primary user. The sensing accuracy of the secondary user is dependent on the signal-to-noise ratio of the primary user signal. A novel Multi-head Attention-based spectrum sensing for Cognitive Radio is proposed through this work to increase the detection probability of the primary user at a low signal- to-noise ratio condition. A radio machine learning dataset with a variety of digital modulation schemes and varying signal-to-noise ratios served as a training source for the proposed model. Further, the performance metrics were evaluated to assess the performance of the proposed model. The experimental results indicate that the proposed model is optimized in terms of the amount of training time required which also has an increase of 27.6% in the probability of detection of the primary user under a low signal-to-noise ratio when compared to other related works that use deep learning.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":" ","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43848063","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}
Significant attention is paid to static analysis methods for Worst Case Execution Time Analysis of programs. However, major effort has been focused on WCET analysis of sequential programs and only a little work is performed on that of multithreaded programs. Shared computer architectural units such as shared instruction cache pose a special challenge in WCET analysis of multithreaded programs. The principle used to improve the precision of shared instruction cache analysis is to shrink the set of interferences, from competing threads to an instruction in a thread that may be accessed from shared instruction cache, using static analysis extended to barriers. An Algorithm that address barrier synchronization and used by the simulator is designed and benchmark programs consisting of both barrier synchronization and computation task synchronization are presented. Improvements in precision upto 20 % are observed while performing the proposed WCET analysis on benchmark programs.
{"title":"Design and Implementation of a Simulator for Precise WCET Estimation of Multithreaded Program","authors":"P. Padma, P. Dharishini, P. V. R. Murthy","doi":"10.32985/ijeces.14.2.11","DOIUrl":"https://doi.org/10.32985/ijeces.14.2.11","url":null,"abstract":"Significant attention is paid to static analysis methods for Worst Case Execution Time Analysis of programs. However, major effort has been focused on WCET analysis of sequential programs and only a little work is performed on that of multithreaded programs. Shared computer architectural units such as shared instruction cache pose a special challenge in WCET analysis of multithreaded programs. The principle used to improve the precision of shared instruction cache analysis is to shrink the set of interferences, from competing threads to an instruction in a thread that may be accessed from shared instruction cache, using static analysis extended to barriers. An Algorithm that address barrier synchronization and used by the simulator is designed and benchmark programs consisting of both barrier synchronization and computation task synchronization are presented. Improvements in precision upto 20 % are observed while performing the proposed WCET analysis on benchmark programs.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"1 1","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"69473533","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}
Intrusion detection models using machine-learning algorithms are used for Intrusion prediction and prevention purposes. Wireless sensor network has a possibility of being attacked by various kinds of threats that will de-promote the performance of any network. These WSN are also affected by the sensor networks that send wrong information because of some environmental causes in- built disturbances misaligned management of the sensors in creating intrusion to the wireless sensor networks. Even though signified routing protocols cannot assure the required security in wireless sensor networks. The idea system provides a key solution for this kind of problem that arises in the network and predicts the abnormal behavior of the sensor nodes as well. But built model by the proposed system various approaches in detecting these kinds of intrusions in any wireless sensor networks in the past few years. The proposed system methodology gives a phenomenon control over the wireless sensor network in detecting the inclusions in its early stages itself. The Data set pre-processing is done by a method of applying the minimum number of features for intrusion detection systems using a machine learning algorithm. The main scope of this article is to improve the prediction of intrusion in a wireless sensor network using AI- based algorithms. This also includes the finest feature selection methodologies to increase the performance of the built model using the selected classifier, which is the Bayes category algorithm. Performance accuracy in the prediction of different attacks in wireless sensor networks is attained at nearly 95.8% for six selected attributes, a Precision level of 0.958, and the receiver operating characteristics or the area under the curve is equal to 0.989.
{"title":"Feature Selection Model using Naive Bayes ML Algorithm for WSN Intrusion Detection System","authors":"Deepa Jeevaraj, B. Karthik, T. Vijayan, M. Sriram","doi":"10.32985/ijeces.14.2.7","DOIUrl":"https://doi.org/10.32985/ijeces.14.2.7","url":null,"abstract":"Intrusion detection models using machine-learning algorithms are used for Intrusion prediction and prevention purposes. Wireless sensor network has a possibility of being attacked by various kinds of threats that will de-promote the performance of any network. These WSN are also affected by the sensor networks that send wrong information because of some environmental causes in- built disturbances misaligned management of the sensors in creating intrusion to the wireless sensor networks. Even though signified routing protocols cannot assure the required security in wireless sensor networks. The idea system provides a key solution for this kind of problem that arises in the network and predicts the abnormal behavior of the sensor nodes as well. But built model by the proposed system various approaches in detecting these kinds of intrusions in any wireless sensor networks in the past few years. The proposed system methodology gives a phenomenon control over the wireless sensor network in detecting the inclusions in its early stages itself. The Data set pre-processing is done by a method of applying the minimum number of features for intrusion detection systems using a machine learning algorithm. The main scope of this article is to improve the prediction of intrusion in a wireless sensor network using AI- based algorithms. This also includes the finest feature selection methodologies to increase the performance of the built model using the selected classifier, which is the Bayes category algorithm. Performance accuracy in the prediction of different attacks in wireless sensor networks is attained at nearly 95.8% for six selected attributes, a Precision level of 0.958, and the receiver operating characteristics or the area under the curve is equal to 0.989.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":"1 1","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"69473536","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}
Sumaiya Thaseen Ikram, P. V, Shourya Chambial, D. Sood, Arulkumar V
In the current era, many fake videos and images are created with the help of various software and new AI (Artificial Intelligence) technologies, which leave a few hints of manipulation. There are many unethical ways videos can be used to threaten, fight, or create panic among people. It is important to ensure that such methods are not used to create fake videos. An AI-based technique for the synthesis of human images is called Deep Fake. They are created by combining and superimposing existing videos onto the source videos. In this paper, a system is developed that uses a hybrid Convolutional Neural Network (CNN) consisting of InceptionResnet v2 and Xception to extract frame-level features. Experimental analysis is performed using the DFDC deep fake detection challenge on Kaggle. These deep learning-based methods are optimized to increase accuracy and decrease training time by using this dataset for training and testing. We achieved a precision of 0.985, a recall of 0.96, an f1-score of 0.98, and support of 0.968.
{"title":"A Performance Enhancement of Deepfake Video Detection through the use of a Hybrid CNN Deep Learning Model","authors":"Sumaiya Thaseen Ikram, P. V, Shourya Chambial, D. Sood, Arulkumar V","doi":"10.32985/ijeces.14.2.6","DOIUrl":"https://doi.org/10.32985/ijeces.14.2.6","url":null,"abstract":"In the current era, many fake videos and images are created with the help of various software and new AI (Artificial Intelligence) technologies, which leave a few hints of manipulation. There are many unethical ways videos can be used to threaten, fight, or create panic among people. It is important to ensure that such methods are not used to create fake videos. An AI-based technique for the synthesis of human images is called Deep Fake. They are created by combining and superimposing existing videos onto the source videos. In this paper, a system is developed that uses a hybrid Convolutional Neural Network (CNN) consisting of InceptionResnet v2 and Xception to extract frame-level features. Experimental analysis is performed using the DFDC deep fake detection challenge on Kaggle. These deep learning-based methods are optimized to increase accuracy and decrease training time by using this dataset for training and testing. We achieved a precision of 0.985, a recall of 0.96, an f1-score of 0.98, and support of 0.968.","PeriodicalId":41912,"journal":{"name":"International Journal of Electrical and Computer Engineering Systems","volume":" ","pages":""},"PeriodicalIF":1.7,"publicationDate":"2023-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43529449","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：