Pub Date : 2023-03-03DOI: 10.1109/RAIT57693.2023.10127101
Khushboo Jha, Arun Jain, S. Srivastava
This work proposes an efficient cepstral-frequency domain based acoustic feature as a speaker identification solution for reliable biometric access control system. The Convolutional Neural Network (CNN) trained for this purpose uses the amalgamation of cepstral-frequency domain based acoustic features such as Power Normalized Cepstral Coefficients (PNCC) and Formant as PNCC-F. The PNCC-F with CNN classifier demonstrates an increase in identification efficacy. The speaker identification accuracy in clean, as well as noisy environment, has been used to evaluate the effectiveness of PNCC alone and in tandem with the formant feature. This work has been executed in a Python 3.8.8 environment using the standard database with 43 speakers called VidTIMIT. The efficiency of the PNCC-F feature was further evaluated in a real-time noisy environment by mixing babble, factory, and machine gun noises from NOISEX-92 database to speech samples with 0 to 20 dB of distortion. The proposed PNCC-F feature surpassed the conventional PNCC feature in a clean environment by 2.34%, and outperformed at all SNR levels for all different noises.
{"title":"An Efficient Speaker Identification Approach for Biometric Access Control System","authors":"Khushboo Jha, Arun Jain, S. Srivastava","doi":"10.1109/RAIT57693.2023.10127101","DOIUrl":"https://doi.org/10.1109/RAIT57693.2023.10127101","url":null,"abstract":"This work proposes an efficient cepstral-frequency domain based acoustic feature as a speaker identification solution for reliable biometric access control system. The Convolutional Neural Network (CNN) trained for this purpose uses the amalgamation of cepstral-frequency domain based acoustic features such as Power Normalized Cepstral Coefficients (PNCC) and Formant as PNCC-F. The PNCC-F with CNN classifier demonstrates an increase in identification efficacy. The speaker identification accuracy in clean, as well as noisy environment, has been used to evaluate the effectiveness of PNCC alone and in tandem with the formant feature. This work has been executed in a Python 3.8.8 environment using the standard database with 43 speakers called VidTIMIT. The efficiency of the PNCC-F feature was further evaluated in a real-time noisy environment by mixing babble, factory, and machine gun noises from NOISEX-92 database to speech samples with 0 to 20 dB of distortion. The proposed PNCC-F feature surpassed the conventional PNCC feature in a clean environment by 2.34%, and outperformed at all SNR levels for all different noises.","PeriodicalId":281845,"journal":{"name":"2023 5th International Conference on Recent Advances in Information Technology (RAIT)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125730356","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}
DNA microarrays can simultaneously measure the expression level of thousands of gene within a particular mRNA sample that provide information about the state of cells and tissues. Though these expressive values are useful in cancer classification, and understand the mechanisms involved in the genesis of disease processes, however, only a few genes out of these thousands of genes contribute towards disease classification. On this basis, usage of feature selection algorithm is favourable, as the main goal of feature selection algorithm is to identify the relevant features (here genes) efficiently. In this paper, we have applied four filter Feature Selection (FS) methods, namely, Mutual Information (MI), Pearson Correlation Coefficient (PCC), Chi2, ReliefF along with three well-known classifiers, namely, Random Forest (RF), Decision Tree (DT), and K-Nearest Neighbour (KNN) on six microarray datasets (both binary and multi-class) namely, Leukemia, Lung, Lymphoma and Leukemia, Gastric, SRBCT and Childhood Tumor and recorded the accuracies.
{"title":"Microarray Data Analysis for Diagnosis of Cancer Diseases by Machine Learning algorithm","authors":"Shemim Begum, Swaraj Samanta, Salauddin Ahmed, Debasis Chakraborty","doi":"10.1109/RAIT57693.2023.10127091","DOIUrl":"https://doi.org/10.1109/RAIT57693.2023.10127091","url":null,"abstract":"DNA microarrays can simultaneously measure the expression level of thousands of gene within a particular mRNA sample that provide information about the state of cells and tissues. Though these expressive values are useful in cancer classification, and understand the mechanisms involved in the genesis of disease processes, however, only a few genes out of these thousands of genes contribute towards disease classification. On this basis, usage of feature selection algorithm is favourable, as the main goal of feature selection algorithm is to identify the relevant features (here genes) efficiently. In this paper, we have applied four filter Feature Selection (FS) methods, namely, Mutual Information (MI), Pearson Correlation Coefficient (PCC), Chi2, ReliefF along with three well-known classifiers, namely, Random Forest (RF), Decision Tree (DT), and K-Nearest Neighbour (KNN) on six microarray datasets (both binary and multi-class) namely, Leukemia, Lung, Lymphoma and Leukemia, Gastric, SRBCT and Childhood Tumor and recorded the accuracies.","PeriodicalId":281845,"journal":{"name":"2023 5th International Conference on Recent Advances in Information Technology (RAIT)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114582235","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 : 2023-03-03DOI: 10.1109/RAIT57693.2023.10126590
Amitesh Das, Sayanti Ghosh, A. Bhowmick, Sanjay Dhar Roy, S. Kundu
This study investigates the performance of a relay-assisted cognitive radio (CR) enabled device-to-device (D2D) communication with Kernelized Energy Detection (KED). A D2D user uses KED technique for sensing the cellular user (CU) channel and uses the same while it is found to be idle. The D2D communication system uses multiple-input and multiple-output (MIMO), and non-orthogonal multiple access (NOMA) techniques to reduce the bad impact of fading and improve spectrum efficiency. The relay forwards the data received from a D2D source to a D2D destination and at each destination device, the received signals are combined using the maximal ratio combining (MRC) technique. The outage probability is studied as a performance metric. An analytical model of the outage probability for the considered network scenario is developed. A simulation framework has been developed and validated with the analytical framework.
{"title":"Outage Analysis of a D2D Network for MIMO-NOMA-based Downlink Transmission","authors":"Amitesh Das, Sayanti Ghosh, A. Bhowmick, Sanjay Dhar Roy, S. Kundu","doi":"10.1109/RAIT57693.2023.10126590","DOIUrl":"https://doi.org/10.1109/RAIT57693.2023.10126590","url":null,"abstract":"This study investigates the performance of a relay-assisted cognitive radio (CR) enabled device-to-device (D2D) communication with Kernelized Energy Detection (KED). A D2D user uses KED technique for sensing the cellular user (CU) channel and uses the same while it is found to be idle. The D2D communication system uses multiple-input and multiple-output (MIMO), and non-orthogonal multiple access (NOMA) techniques to reduce the bad impact of fading and improve spectrum efficiency. The relay forwards the data received from a D2D source to a D2D destination and at each destination device, the received signals are combined using the maximal ratio combining (MRC) technique. The outage probability is studied as a performance metric. An analytical model of the outage probability for the considered network scenario is developed. A simulation framework has been developed and validated with the analytical framework.","PeriodicalId":281845,"journal":{"name":"2023 5th International Conference on Recent Advances in Information Technology (RAIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122095218","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 : 2023-03-03DOI: 10.1109/RAIT57693.2023.10127066
Ashok Ajad, Taniya Saini, K. M. Niranjan
Nowadays Covid-19 is prevailing across the world, it has affected millions of populations across the world. The exponential increase in covid cases makes the health care system overwhelmed. Many testing methods are used for covid-19 detection like Rapid antigen test, RT-PCR test, etc. These tests have certain limitations, sometimes people got confused between respiratory infection and covid-19infection, as many symptoms are similar. So for confirming the disease, a chest x-ray is preferred. Covid-19 has similar symptoms of pneumonia, consolidation, and ground-glass opacities, in our approach we consider them as covid. In this paper, images are acquired from reputed hospitals and various online datasets used in Covidnet architecture. After accumulation, the dataset is verified by experienced radiologists. In our approach, we trained our models on various symptoms of covid19 like pneumonia, consolidation, ggopacities and finally on covid-19 dataset images. In our research, we have used single as well as ensemble models for classification. Models like densenet, efficient net, resnet, etc are used. Certain preprocessing techniques are used before passing the image dataset into training like adaptive histogram equalization, data augmentation methods, etc. Finally, a approach based on Deep Learning used for identification of covid 19. We are claiming 95% plus testing accuracy and 99% training accuracy. Beyond classification, we further generate the reports and localize the covid virus on Xray using various visualization methods. Further results are classified based on single and ensemble models on the in-house dataset.
{"title":"CV-CXR: A Method for Classification and Visualisation of Covid-19 virus using CNN and Heatmap*","authors":"Ashok Ajad, Taniya Saini, K. M. Niranjan","doi":"10.1109/RAIT57693.2023.10127066","DOIUrl":"https://doi.org/10.1109/RAIT57693.2023.10127066","url":null,"abstract":"Nowadays Covid-19 is prevailing across the world, it has affected millions of populations across the world. The exponential increase in covid cases makes the health care system overwhelmed. Many testing methods are used for covid-19 detection like Rapid antigen test, RT-PCR test, etc. These tests have certain limitations, sometimes people got confused between respiratory infection and covid-19infection, as many symptoms are similar. So for confirming the disease, a chest x-ray is preferred. Covid-19 has similar symptoms of pneumonia, consolidation, and ground-glass opacities, in our approach we consider them as covid. In this paper, images are acquired from reputed hospitals and various online datasets used in Covidnet architecture. After accumulation, the dataset is verified by experienced radiologists. In our approach, we trained our models on various symptoms of covid19 like pneumonia, consolidation, ggopacities and finally on covid-19 dataset images. In our research, we have used single as well as ensemble models for classification. Models like densenet, efficient net, resnet, etc are used. Certain preprocessing techniques are used before passing the image dataset into training like adaptive histogram equalization, data augmentation methods, etc. Finally, a approach based on Deep Learning used for identification of covid 19. We are claiming 95% plus testing accuracy and 99% training accuracy. Beyond classification, we further generate the reports and localize the covid virus on Xray using various visualization methods. Further results are classified based on single and ensemble models on the in-house dataset.","PeriodicalId":281845,"journal":{"name":"2023 5th International Conference on Recent Advances in Information Technology (RAIT)","volume":"47 17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124892583","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 : 2023-03-03DOI: 10.1109/RAIT57693.2023.10127096
Ashok Ajad, Taniya Saini, Kumar M Niranjan
Several professional societies have advocated for structured reporting in radiology, citing gains in quality, but some studies have shown that rigid templates and strict adherence may be too distracting and lead to incomplete reports. To gain the advantages of structured reporting while requiring minimal change to a radiologist's work-flow, the present work proposes a two-stage abstractive summarization approach that first finds the key findings in an unstructured report and then generates and organizes descriptions of each finding into a given template. The method uses a large manually annotated dataset and a taxonomy and other domain knowledge that were prepared in consultation with several practising radiologists. It can be used to structure reports dictated by radiologists and as post- and pre-processing steps for machine-learning pipelines. On the subtask of label extraction, the method achieves significantly better performance than previous rule-based approaches and learning-based approaches that were trained on automatically extracted labels. On the task of summarization, the method achieves more than 0.5 BLEU-4 score across 8 of the 10 most common labels and serves as a strong baseline for future experiments.
{"title":"Rad-Former: Structuring Radiology Reports using Transformers*","authors":"Ashok Ajad, Taniya Saini, Kumar M Niranjan","doi":"10.1109/RAIT57693.2023.10127096","DOIUrl":"https://doi.org/10.1109/RAIT57693.2023.10127096","url":null,"abstract":"Several professional societies have advocated for structured reporting in radiology, citing gains in quality, but some studies have shown that rigid templates and strict adherence may be too distracting and lead to incomplete reports. To gain the advantages of structured reporting while requiring minimal change to a radiologist's work-flow, the present work proposes a two-stage abstractive summarization approach that first finds the key findings in an unstructured report and then generates and organizes descriptions of each finding into a given template. The method uses a large manually annotated dataset and a taxonomy and other domain knowledge that were prepared in consultation with several practising radiologists. It can be used to structure reports dictated by radiologists and as post- and pre-processing steps for machine-learning pipelines. On the subtask of label extraction, the method achieves significantly better performance than previous rule-based approaches and learning-based approaches that were trained on automatically extracted labels. On the task of summarization, the method achieves more than 0.5 BLEU-4 score across 8 of the 10 most common labels and serves as a strong baseline for future experiments.","PeriodicalId":281845,"journal":{"name":"2023 5th International Conference on Recent Advances in Information Technology (RAIT)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133533451","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 : 2023-03-03DOI: 10.1109/RAIT57693.2023.10126579
Lilima Jain, U. Venkanna
Flow table overflow attack on data plane devices is one of the prominent vulnerabilities in the Software Defined Networking (SDN) architecture. Flow table uses limited-sized TCAM to store the flow rules in the data plane. Unfortunately, TCAM based Flow tables are prone to various attacks such as memory saturation attacks, DDoS attacks, cross-plane attacks, Flow table overflow attacks, etc. These attacks lead to the starvation of benign requests, and saturation of network resources. However, the existing solutions are focused on the controller-based attack mitigation mechanism using OpenFlow switches which increases communication overhead between the control plane and data plane. This paper proposes a switch centric based in-network Flow table overflow attack detection and mitigation framework. We introduce IP_SourceGuard which keeps an audit of the flow rules by counting the threat value of a particular port. Mitigating the attack traffic when the threat value exceeds the limit of the warning threshold. Further, IP_SourceGuard blocks the attacker port from further not communicating it to the network. The solution has been implemented using the BMv2 software switch and determined that the solution reduces the Flow table utilization to 88%. From the result, it is observed that our solution mitigates the Flow table overflow attack in a real-time environment.
{"title":"P4 based Switch Centric Flow table Overflow Detection and Mitigation in Data Plane Devices","authors":"Lilima Jain, U. Venkanna","doi":"10.1109/RAIT57693.2023.10126579","DOIUrl":"https://doi.org/10.1109/RAIT57693.2023.10126579","url":null,"abstract":"Flow table overflow attack on data plane devices is one of the prominent vulnerabilities in the Software Defined Networking (SDN) architecture. Flow table uses limited-sized TCAM to store the flow rules in the data plane. Unfortunately, TCAM based Flow tables are prone to various attacks such as memory saturation attacks, DDoS attacks, cross-plane attacks, Flow table overflow attacks, etc. These attacks lead to the starvation of benign requests, and saturation of network resources. However, the existing solutions are focused on the controller-based attack mitigation mechanism using OpenFlow switches which increases communication overhead between the control plane and data plane. This paper proposes a switch centric based in-network Flow table overflow attack detection and mitigation framework. We introduce IP_SourceGuard which keeps an audit of the flow rules by counting the threat value of a particular port. Mitigating the attack traffic when the threat value exceeds the limit of the warning threshold. Further, IP_SourceGuard blocks the attacker port from further not communicating it to the network. The solution has been implemented using the BMv2 software switch and determined that the solution reduces the Flow table utilization to 88%. From the result, it is observed that our solution mitigates the Flow table overflow attack in a real-time environment.","PeriodicalId":281845,"journal":{"name":"2023 5th International Conference on Recent Advances in Information Technology (RAIT)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125908267","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：