Pub Date : 2021-12-01DOI: 10.1109/ICCS54944.2021.00036
K. Raghavendar, Isha Batra, Arun Malik
In the actual world, AI is being used to address issues of class inequality. This is especially true when the information is not just unbalanced, but also multidimensional. When there is a class imbalance, a large dimensionality of datasets is always present, and both difficulties must be considered jointly. When using examples to evaluate each component, standard element picking algorithms usually provide equal weights to tests from different classes. As a result, they are unable to operate effectively with unbalanced data. When the costs of misclassification of different classes are different, cost-effective learning procedures are typically used. Different processes in writing have been established to deal with concerns related to class discomfort.
{"title":"Novel Framework for Resources Optimization to Solve Class Imbalance Problems","authors":"K. Raghavendar, Isha Batra, Arun Malik","doi":"10.1109/ICCS54944.2021.00036","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00036","url":null,"abstract":"In the actual world, AI is being used to address issues of class inequality. This is especially true when the information is not just unbalanced, but also multidimensional. When there is a class imbalance, a large dimensionality of datasets is always present, and both difficulties must be considered jointly. When using examples to evaluate each component, standard element picking algorithms usually provide equal weights to tests from different classes. As a result, they are unable to operate effectively with unbalanced data. When the costs of misclassification of different classes are different, cost-effective learning procedures are typically used. Different processes in writing have been established to deal with concerns related to class discomfort.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124563376","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 : 2021-12-01DOI: 10.1109/ICCS54944.2021.00052
Kamarthi Lava Kumar, B. E. Reddy
Cardiac disease is defined as abnormal heart function caused by a variety of factors. Heart Failure (HF), Coronary Artery Disease (CAD), and Cardiovascular Disease (CV) are the three most frequent forms of heart disease. Coronary artery blockage or narrowing is the leading cause of heart failure. Many researchers have created various methods for the automated diagnosis of heart failure. The recently suggested techniques increases the accuracy of heart failure diagnosis on both testing and training the model. In this proposed system, supervised learning i.e., gradient boosting technique is used to detect the heart failure. The proposed diagnostic system uses gradient boosting algorithm (GB) for training & testing the model. Gradient boosting classifier is used to extract the features of heart diagnosis. In this experiment, the detection of heart failure disease by using Cleveland Dataset. The proposed system, achieves an accuracy of 97.10% which compares with an other methods.
{"title":"Heart Disease Detection System Using Gradient Boosting Technique","authors":"Kamarthi Lava Kumar, B. E. Reddy","doi":"10.1109/ICCS54944.2021.00052","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00052","url":null,"abstract":"Cardiac disease is defined as abnormal heart function caused by a variety of factors. Heart Failure (HF), Coronary Artery Disease (CAD), and Cardiovascular Disease (CV) are the three most frequent forms of heart disease. Coronary artery blockage or narrowing is the leading cause of heart failure. Many researchers have created various methods for the automated diagnosis of heart failure. The recently suggested techniques increases the accuracy of heart failure diagnosis on both testing and training the model. In this proposed system, supervised learning i.e., gradient boosting technique is used to detect the heart failure. The proposed diagnostic system uses gradient boosting algorithm (GB) for training & testing the model. Gradient boosting classifier is used to extract the features of heart diagnosis. In this experiment, the detection of heart failure disease by using Cleveland Dataset. The proposed system, achieves an accuracy of 97.10% which compares with an other methods.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116405334","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}
We received a total of 180 submissions for the conference, of which 27 were selected for the final program. The low acceptance rate (15%) reflects the growing number of submissions to ISCA, since the number of accepted papers was not significantly different than in the past. Each paper was reviewed by at least four reviewers, and at least two of those reviews were by program committee members. A total of 300 people helped review papers. All reviewing was double-blind. The review process was slightly different than in the past; authors were allowed to see and respond to reviews prior to the program committee meeting.
{"title":"Message from the Program Chair","authors":"D. Grunwald","doi":"10.1109/pact.2011.76","DOIUrl":"https://doi.org/10.1109/pact.2011.76","url":null,"abstract":"We received a total of 180 submissions for the conference, of which 27 were selected for the final program. The low acceptance rate (15%) reflects the growing number of submissions to ISCA, since the number of accepted papers was not significantly different than in the past. Each paper was reviewed by at least four reviewers, and at least two of those reviews were by program committee members. A total of 300 people helped review papers. All reviewing was double-blind. The review process was slightly different than in the past; authors were allowed to see and respond to reviews prior to the program committee meeting.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"201 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123024100","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 : 2021-12-01DOI: 10.1109/ICCS54944.2021.00035
Tanima Thakur, Isha Batra, Arun Malik
A vast number of people are affected by cancer. It is a scary disease that requires timely detection and treatment. There are numerous ways through which it can be predicted. One such way is using gene expression. With the help of the genes, it can be found that a person is going to suffer from disease or not. In the literature, there are varieties of ML and DL methods that may be used to train and test computers on diverse datasets. This paper will provide an overview of various ML and DL models, as well as an evaluation of their performance on a specific dataset. Preprocessing of the data is done and then machines are trained and tested using various DL and ML models. Then the accuracy of the models is calculated. Lastly, the models are compared to find which model of ML and DL performs better for the given dataset.
{"title":"A Comparative Analysis of various Machine Learning and Deep Learning Models for Gene Expression","authors":"Tanima Thakur, Isha Batra, Arun Malik","doi":"10.1109/ICCS54944.2021.00035","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00035","url":null,"abstract":"A vast number of people are affected by cancer. It is a scary disease that requires timely detection and treatment. There are numerous ways through which it can be predicted. One such way is using gene expression. With the help of the genes, it can be found that a person is going to suffer from disease or not. In the literature, there are varieties of ML and DL methods that may be used to train and test computers on diverse datasets. This paper will provide an overview of various ML and DL models, as well as an evaluation of their performance on a specific dataset. Preprocessing of the data is done and then machines are trained and tested using various DL and ML models. Then the accuracy of the models is calculated. Lastly, the models are compared to find which model of ML and DL performs better for the given dataset.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130211683","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 : 2021-12-01DOI: 10.1109/iccs54944.2021.00002
{"title":"Title Page","authors":"","doi":"10.1109/iccs54944.2021.00002","DOIUrl":"https://doi.org/10.1109/iccs54944.2021.00002","url":null,"abstract":"","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"141 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127576140","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 : 2021-12-01DOI: 10.1109/iccs54944.2021.00001
{"title":"Title Page","authors":"","doi":"10.1109/iccs54944.2021.00001","DOIUrl":"https://doi.org/10.1109/iccs54944.2021.00001","url":null,"abstract":"","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126312902","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 : 2021-12-01DOI: 10.1109/ICCS54944.2021.00040
R. Ruchi, Jimmy Singla
The prime objective of this study is to recognize and segment Lower Lumber Spine from the collected sample and then perform classification to separate affected and non-affected regions by lower lumbar spine disease. The proposed model first of all performs identification and separation of regions from the sample. This was performed by converting RGB cell image into gray colour scale. Background subtraction algorithm was applied to extract only cell structures from the image by eliminating the background completely and region-props. In second phase, features from the segmented regions were extracted. These features include homogeneity, contrast, energy, correlation and some hybrid features. In the third phase, digital differential analyzer optimization(DDAO) algorithm was applied to select the significant features. In the final phase, different classifiers were used to validate the performance of proposed optimization approach. The proposedmodel was applied on well-known benchmarked dataset. The obtained results corresponding to identification and separation were 92, 88 and 80% of segmentation accuracy, sensitivity and specificity, respectively. This result was best among other published papers worked on same dataset. Classification accuracy was notably higher as compared to other models not following DDA optimization algorithm. Validation of results was further extended through feature reduction ratio and still remarkable results in terms classification accuracy of 90% was achieved.
{"title":"Classification of Lumbar Disc Disorder from MRI and CT images using Iterative Differential Approach","authors":"R. Ruchi, Jimmy Singla","doi":"10.1109/ICCS54944.2021.00040","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00040","url":null,"abstract":"The prime objective of this study is to recognize and segment Lower Lumber Spine from the collected sample and then perform classification to separate affected and non-affected regions by lower lumbar spine disease. The proposed model first of all performs identification and separation of regions from the sample. This was performed by converting RGB cell image into gray colour scale. Background subtraction algorithm was applied to extract only cell structures from the image by eliminating the background completely and region-props. In second phase, features from the segmented regions were extracted. These features include homogeneity, contrast, energy, correlation and some hybrid features. In the third phase, digital differential analyzer optimization(DDAO) algorithm was applied to select the significant features. In the final phase, different classifiers were used to validate the performance of proposed optimization approach. The proposedmodel was applied on well-known benchmarked dataset. The obtained results corresponding to identification and separation were 92, 88 and 80% of segmentation accuracy, sensitivity and specificity, respectively. This result was best among other published papers worked on same dataset. Classification accuracy was notably higher as compared to other models not following DDA optimization algorithm. Validation of results was further extended through feature reduction ratio and still remarkable results in terms classification accuracy of 90% was achieved.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126620017","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 : 2021-12-01DOI: 10.1109/ICCS54944.2021.00013
Vipin Kumar, Navneet Malik
Wireless Sensor technology research is becoming more popular in studying the Internet of Things (IoT) area. Sensor technology helps to collect information from the environment that can be used to analyze the system and helps to enhance the system's overall performance. The Internet of Things, a collection of linked devices, has become a new research topic for sensor technology. Quality of service (QoS), data routing, real-time monitoring performance, and connecting heterogeneous devices are difficulties faced by the Internet of Things security concerns. These networks are concerned about the wearable devices' short battery life, limited range, and limited capacity. Because of the wide variety of assaults that may be conducted against them in the real world, the security of Internet of Things devices is a complex problem to solve. As a consequence, particular standards for devices linked to the Internet of Things are needed. The sensor network must choose the most appropriate encryption technique from various choices to allow secure communication between sensor nodes. The proper operation of encrypted communications necessitates the use of keys. As a result, they must be developed and distributed. The present key management method is linked with significant computational overheads since it consumes a lot of energy and takes a long time to finish. As a result of the restricted bandwidth capacity of the sensor nodes in the network, the network is inefficient. IoT controllers are responsible for controlling a group of networks, and this article describes a method for dynamic key management that is both dynamic and scalable. Packet loss has been reduced by a significant proportion when compared to a conventional one-hop key management scheme. The suggested approach reduces energy consumption, computational overheads, and latency, all of which help to enhance network performance.
{"title":"Dynamic Group Key Management Technique in Context of Modern IoT Applications","authors":"Vipin Kumar, Navneet Malik","doi":"10.1109/ICCS54944.2021.00013","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00013","url":null,"abstract":"Wireless Sensor technology research is becoming more popular in studying the Internet of Things (IoT) area. Sensor technology helps to collect information from the environment that can be used to analyze the system and helps to enhance the system's overall performance. The Internet of Things, a collection of linked devices, has become a new research topic for sensor technology. Quality of service (QoS), data routing, real-time monitoring performance, and connecting heterogeneous devices are difficulties faced by the Internet of Things security concerns. These networks are concerned about the wearable devices' short battery life, limited range, and limited capacity. Because of the wide variety of assaults that may be conducted against them in the real world, the security of Internet of Things devices is a complex problem to solve. As a consequence, particular standards for devices linked to the Internet of Things are needed. The sensor network must choose the most appropriate encryption technique from various choices to allow secure communication between sensor nodes. The proper operation of encrypted communications necessitates the use of keys. As a result, they must be developed and distributed. The present key management method is linked with significant computational overheads since it consumes a lot of energy and takes a long time to finish. As a result of the restricted bandwidth capacity of the sensor nodes in the network, the network is inefficient. IoT controllers are responsible for controlling a group of networks, and this article describes a method for dynamic key management that is both dynamic and scalable. Packet loss has been reduced by a significant proportion when compared to a conventional one-hop key management scheme. The suggested approach reduces energy consumption, computational overheads, and latency, all of which help to enhance network performance.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126639777","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 : 2021-12-01DOI: 10.1109/ICCS54944.2021.00064
Gaganjot Kaur, J. Kilby
Health care is a main analysis area required instant results. Data storage as well as assessment have become more difficult as a result of the digitalization of information in all fields. As a result, the demand for skilled methodologies for analyzing health information is growing. Predictive analytics is an important issue from the health care field to computer technology researchers in ability to forecast as well as reduce potential health uprisings. Parallel research efforts are being made in many areas to forecast the disease's potential effect on multiple healthcare areas. Even so, those attempts are restricted and do not go as far to produce the desired outcomes. Lately, in the context of information systems, non-contact methodologies have been shown to make a positive contribution to the healthcare profession through improving the accuracy as well as speed of disease diagnosis. As a result, this research analyzed heart rate assessment utilizing non-contact technique to measure disease severity stages.
{"title":"Non-contact Methods for Heart Rate Measurement: A Review","authors":"Gaganjot Kaur, J. Kilby","doi":"10.1109/ICCS54944.2021.00064","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00064","url":null,"abstract":"Health care is a main analysis area required instant results. Data storage as well as assessment have become more difficult as a result of the digitalization of information in all fields. As a result, the demand for skilled methodologies for analyzing health information is growing. Predictive analytics is an important issue from the health care field to computer technology researchers in ability to forecast as well as reduce potential health uprisings. Parallel research efforts are being made in many areas to forecast the disease's potential effect on multiple healthcare areas. Even so, those attempts are restricted and do not go as far to produce the desired outcomes. Lately, in the context of information systems, non-contact methodologies have been shown to make a positive contribution to the healthcare profession through improving the accuracy as well as speed of disease diagnosis. As a result, this research analyzed heart rate assessment utilizing non-contact technique to measure disease severity stages.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114968087","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 : 2021-12-01DOI: 10.1109/ICCS54944.2021.00025
Korakod Tongkachok, Luigi Pio Leonardo Cavaliere, Sudakshina C, G. Hosamani, Dhiraj Kapila, Samrat Ray
In computing, the Internet of Things (IoT) refers to a networked structure of linked computers gadgets, technical and computerized equipment, things, creatures, or humans who have separate personalities and the ability to transmit information without the need for direct human or computer interaction. As per study, 50 billion items would be linked to the internet, with 35 billion being IoT devices and the other 15 billion being smartphones, tablets, smart objects, and the like. Technologies have had a significant effect on the learning experiences in current history. The Internet of Things (IoT) is greatly benefiting the field of information and communications technology as well as societal growth. With IoT, educational institutions may improve their students' learning experiences by offering more rich learning opportunities. The Internet of Things need expansion, which universities may be able to assist with. Scholars, researchers, and students play a critical role in advancing the research and development of Internet of Things systems, devices, apps, and activities, as well as in identifying new opportunities. On the other hand, the Internet of Things presents major challenges to higher education. As a consequence, this article also offers a point of view on the challenges of IoT in higher ed, which is discussed further below. The Internet of Things (IoT) has a significant impact on the way university's function & improves student education across a broad variety of disciplines and at all stages.
{"title":"Towards a framework for Internet of Things and Its Impact on Performance Management in a Higher Education Institution","authors":"Korakod Tongkachok, Luigi Pio Leonardo Cavaliere, Sudakshina C, G. Hosamani, Dhiraj Kapila, Samrat Ray","doi":"10.1109/ICCS54944.2021.00025","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00025","url":null,"abstract":"In computing, the Internet of Things (IoT) refers to a networked structure of linked computers gadgets, technical and computerized equipment, things, creatures, or humans who have separate personalities and the ability to transmit information without the need for direct human or computer interaction. As per study, 50 billion items would be linked to the internet, with 35 billion being IoT devices and the other 15 billion being smartphones, tablets, smart objects, and the like. Technologies have had a significant effect on the learning experiences in current history. The Internet of Things (IoT) is greatly benefiting the field of information and communications technology as well as societal growth. With IoT, educational institutions may improve their students' learning experiences by offering more rich learning opportunities. The Internet of Things need expansion, which universities may be able to assist with. Scholars, researchers, and students play a critical role in advancing the research and development of Internet of Things systems, devices, apps, and activities, as well as in identifying new opportunities. On the other hand, the Internet of Things presents major challenges to higher education. As a consequence, this article also offers a point of view on the challenges of IoT in higher ed, which is discussed further below. The Internet of Things (IoT) has a significant impact on the way university's function & improves student education across a broad variety of disciplines and at all stages.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115052008","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：