Pub Date : 2021-12-01DOI: 10.1109/ICCS54944.2021.00031
Mrinalini Rana, Jimmy Singla
To achieve efficient rule mining feature selection or preprocessing is need to be handled before the implementing the optimization technique. For these different methods are available. In the proposed model $K$ means clustering is used to generate the clusters. Then PSO-ABC hybrid approach is for feature optimization. For the obtained result, PSO-ABC represent more normalized features as compared to using PSO only.
{"title":"A Pre-processing Model for Feature Extraction Based on K-mean, PSO and ABC","authors":"Mrinalini Rana, Jimmy Singla","doi":"10.1109/ICCS54944.2021.00031","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00031","url":null,"abstract":"To achieve efficient rule mining feature selection or preprocessing is need to be handled before the implementing the optimization technique. For these different methods are available. In the proposed model $K$ means clustering is used to generate the clusters. Then PSO-ABC hybrid approach is for feature optimization. For the obtained result, PSO-ABC represent more normalized features as compared to using PSO only.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"16 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":"127214971","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/ICCS54944.2021.00032
Cherry Khosla, B. Saini
Database Management systems are the crucial components in data intensive applications. For efficiently processing and analyzing the data, the database systems should be at par in terms of performance. Tuning the database is nowadays the major goal to improve DBMS's operations. Optimizing the databases to increase the performance and to meet the needs of an application has surpassed the abilities of the humans. Therefore, the researchers have explored how machine learning can be used to tune the databases automatically. In this paper, we have reviewed and discussed various areas of the database tuning, and how machine learning is used to automatically tune the configurations. We have also discussed the various approaches that can be used to integrate the machine learning with database systems. Lastly, we discussed the open problem in tuning database systems.
{"title":"On a Way Together - Database and Machine Learning for Performance Tuning","authors":"Cherry Khosla, B. Saini","doi":"10.1109/ICCS54944.2021.00032","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00032","url":null,"abstract":"Database Management systems are the crucial components in data intensive applications. For efficiently processing and analyzing the data, the database systems should be at par in terms of performance. Tuning the database is nowadays the major goal to improve DBMS's operations. Optimizing the databases to increase the performance and to meet the needs of an application has surpassed the abilities of the humans. Therefore, the researchers have explored how machine learning can be used to tune the databases automatically. In this paper, we have reviewed and discussed various areas of the database tuning, and how machine learning is used to automatically tune the configurations. We have also discussed the various approaches that can be used to integrate the machine learning with database systems. Lastly, we discussed the open problem in tuning database systems.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"21 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":"126405295","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/ICCS54944.2021.00059
Pooja Mallam, Ashu Ashu, Baljeet Singh
Business intelligence and data analytics areas identify areas where business is at stake or performing well. A much better view of where business may be dropping short is observed. Business Analytics and Data analytics use evidence accumulated over years and get useful information to know where is deformity in businesses. This means that the info is organized and presented that allows it to be visualized. While raw numbers are important, once information becomes valuable and demonstrate value, making insights more straight forward to understand. This paper gives you a brief review of development of business intelligence over course of time using Data Analytics.
{"title":"Business Intelligence Techniques Using Data Analytics: An Overview","authors":"Pooja Mallam, Ashu Ashu, Baljeet Singh","doi":"10.1109/ICCS54944.2021.00059","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00059","url":null,"abstract":"Business intelligence and data analytics areas identify areas where business is at stake or performing well. A much better view of where business may be dropping short is observed. Business Analytics and Data analytics use evidence accumulated over years and get useful information to know where is deformity in businesses. This means that the info is organized and presented that allows it to be visualized. While raw numbers are important, once information becomes valuable and demonstrate value, making insights more straight forward to understand. This paper gives you a brief review of development of business intelligence over course of time using Data Analytics.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"45 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":"122255413","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/ICCS54944.2021.00065
Aarsh Agrawal, Vinay Bhardwaj
Social media is widely regarded as one of the most important unstructured data. Analyzing and extracting meaning from such data is a time-consuming process. Because of the enormous data available on social media platforms, sentiment extraction has gotten a lot of attention. Microblogging is a relatively new phenomenon, with Twitter being the most widely utilized. It's one of the most comprehensive free and open data sources available. Today's society sees a lot of differing viewpoints on Twitter. Researchers can use opinion mining to obtain the present emotion and mood of the public. Sentiment Analysis is defined as the technique of extracting and finding the polarity of a given material to get insight into the hidden information, emotion, feeling contained within a text. The ultimate objective of sentiment analysis is to extract meaningful material from various sources of information. The first analysis of tweets was done using the Natural Language Processing (NLP) method. For further analysis of the opinionated data, two approaches are available: the Lexicon Based Approach (LBA) and the Machine Learning Approach (MLA) based on supervised learning. The LBA approach employs a resource dictionary, namely the Hindi SentiWordNet, and a Hybrid Based Approach (HBA) that joins the Lexicon based and Machine learning for categorizing tweets as positive or negative
{"title":"Methods of Sentiment Analysis for Hindi and English Languages","authors":"Aarsh Agrawal, Vinay Bhardwaj","doi":"10.1109/ICCS54944.2021.00065","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00065","url":null,"abstract":"Social media is widely regarded as one of the most important unstructured data. Analyzing and extracting meaning from such data is a time-consuming process. Because of the enormous data available on social media platforms, sentiment extraction has gotten a lot of attention. Microblogging is a relatively new phenomenon, with Twitter being the most widely utilized. It's one of the most comprehensive free and open data sources available. Today's society sees a lot of differing viewpoints on Twitter. Researchers can use opinion mining to obtain the present emotion and mood of the public. Sentiment Analysis is defined as the technique of extracting and finding the polarity of a given material to get insight into the hidden information, emotion, feeling contained within a text. The ultimate objective of sentiment analysis is to extract meaningful material from various sources of information. The first analysis of tweets was done using the Natural Language Processing (NLP) method. For further analysis of the opinionated data, two approaches are available: the Lexicon Based Approach (LBA) and the Machine Learning Approach (MLA) based on supervised learning. The LBA approach employs a resource dictionary, namely the Hindi SentiWordNet, and a Hybrid Based Approach (HBA) that joins the Lexicon based and Machine learning for categorizing tweets as positive or negative","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"7 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":"133131454","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/ICCS54944.2021.00061
Ravi Shanker, Aman Singh
Various packet capturing tools are available to capture packets in the form of datasets and several well-known datasets available now a day for benchmarking the network attacks for intrusion detection system and attracted the researchers to further analyse these attacks for future attacks also. These datasets contain several parameters that can be utilized for identification of attack at cross layers. Here the cross layer refers the data link layer, network layer, transport layer and application layer. These datasets can be used in research for identification and automation of novel attack. This paper concentrates on the attack types and classification of network attacks on data link layer. Post analysis will investigate the possibility of using Snort as intrusion detection tool for identifying attack at data link layer. Snort generally works at network layer and above so not all data link layer attack can be identified by IDS. For the current work a various solution already researched will be put together to analyse attack at data link layer with the help of Snort. This analysis will help in understanding the possibility to put together various attack at data link layer using Snort or provide already suggested solution done by various researcher.
{"title":"Analysis of Network Attacks at Data Link Layer and its Mitigation","authors":"Ravi Shanker, Aman Singh","doi":"10.1109/ICCS54944.2021.00061","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00061","url":null,"abstract":"Various packet capturing tools are available to capture packets in the form of datasets and several well-known datasets available now a day for benchmarking the network attacks for intrusion detection system and attracted the researchers to further analyse these attacks for future attacks also. These datasets contain several parameters that can be utilized for identification of attack at cross layers. Here the cross layer refers the data link layer, network layer, transport layer and application layer. These datasets can be used in research for identification and automation of novel attack. This paper concentrates on the attack types and classification of network attacks on data link layer. Post analysis will investigate the possibility of using Snort as intrusion detection tool for identifying attack at data link layer. Snort generally works at network layer and above so not all data link layer attack can be identified by IDS. For the current work a various solution already researched will be put together to analyse attack at data link layer with the help of Snort. This analysis will help in understanding the possibility to put together various attack at data link layer using Snort or provide already suggested solution done by various researcher.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"112 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":"117261090","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/ICCS54944.2021.00049
Ahmed Al Ahdal, D. Prashar, Manik Rakhra, Ankita Wadhawan
Heart diseases leading most causes of death globally according to World Health Organization cardiovascular or all heart related disease are responsible for 17.9 million death every year. An early detection and diagnosis of the disease is very important and maybe it's the key of cure. The major challenge is to predict the disease in early stages therefor most of scientists and researches focus on Machine learning techniques which have the capability of detection with accurate result for large and complex data and apply those techniques to help in health care. The purpose of this work is to detect heart diseases at early stage and avoid consequences by implementing different Machine Learning Algorithm for example, KNN Decision Tree (DT), Logistic Regression, SVM, Random Forest (RF), and Naïve Bayes (NB).
{"title":"Machine Learning-Based Heart Patient Scanning, Visualization, and Monitoring","authors":"Ahmed Al Ahdal, D. Prashar, Manik Rakhra, Ankita Wadhawan","doi":"10.1109/ICCS54944.2021.00049","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00049","url":null,"abstract":"Heart diseases leading most causes of death globally according to World Health Organization cardiovascular or all heart related disease are responsible for 17.9 million death every year. An early detection and diagnosis of the disease is very important and maybe it's the key of cure. The major challenge is to predict the disease in early stages therefor most of scientists and researches focus on Machine learning techniques which have the capability of detection with accurate result for large and complex data and apply those techniques to help in health care. The purpose of this work is to detect heart diseases at early stage and avoid consequences by implementing different Machine Learning Algorithm for example, KNN Decision Tree (DT), Logistic Regression, SVM, Random Forest (RF), and Naïve Bayes (NB).","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"29 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":"129416104","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/ICCS54944.2021.00011
Amal Mathew, Kaushik Daiv, Polkumpally Rohan Goud, Piyush Talreja, Sai Sanjana Reddy Vatte
Face identification using ML (machine learning) is well-known. Attendance structures may benefit from this method. Using this method, you may achieve the desired area, as well as beneficial attributes and a dataset, by preparing two sets of data again for test and training phases. To distinguish between a testing set and a test sets, a photograph is used as a testing set. An ensemble classification method is used to sort the test images into categories like “identified” and “unidentified.” This model can't provide reliable findings since it simply divides data into two categories. The development of GLCM was motivated by the need to use texture properties to identify faces. The existence of the query picture is noted once face detection has taken place. In simulation findings, the new model outperforms the baseline models in terms of accuracy. Keywords—Ensemble classifier, GLCM, Face Spoof, SVM, DWT
{"title":"Face Spoof Detection Using Gray Level Co-Occurrence Matrix and Discrete Wavelet Transform Feature Extractor","authors":"Amal Mathew, Kaushik Daiv, Polkumpally Rohan Goud, Piyush Talreja, Sai Sanjana Reddy Vatte","doi":"10.1109/ICCS54944.2021.00011","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00011","url":null,"abstract":"Face identification using ML (machine learning) is well-known. Attendance structures may benefit from this method. Using this method, you may achieve the desired area, as well as beneficial attributes and a dataset, by preparing two sets of data again for test and training phases. To distinguish between a testing set and a test sets, a photograph is used as a testing set. An ensemble classification method is used to sort the test images into categories like “identified” and “unidentified.” This model can't provide reliable findings since it simply divides data into two categories. The development of GLCM was motivated by the need to use texture properties to identify faces. The existence of the query picture is noted once face detection has taken place. In simulation findings, the new model outperforms the baseline models in terms of accuracy. Keywords—Ensemble classifier, GLCM, Face Spoof, SVM, DWT","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"17 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":"127287510","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/ICCS54944.2021.00028
Ravikumar Ch, Isha Batra, A. Malik
Blockchain technology has now spread like wildfire across the internet. Blockchain has emerged as a game-changing technology for complex industrial processes as a result of its openness, availability, and security. In this study, we used a hybrid adaptive crypto cloud framework that combines blockchain technology with multiple authorities and a botnet framework to improve cloud security while reducing computation time. The proposed adaptive crypto cloud system divides the cloud security framework into stages to organize secure data communication and reduce communication latency while detecting internal and external threats. In order to compute authentication using hash mapping and deploy an authentication system to safeguard the various users' authentication information, the whole system placed a major emphasis on block chain technology. The technology not only improves security but also makes the role-based access control system and anonymous authentication system more user-friendly.
{"title":"Combining Blockchain Multi Authority and Botnet to Create a Hybrid Adaptive Crypto Cloud Framework","authors":"Ravikumar Ch, Isha Batra, A. Malik","doi":"10.1109/ICCS54944.2021.00028","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00028","url":null,"abstract":"Blockchain technology has now spread like wildfire across the internet. Blockchain has emerged as a game-changing technology for complex industrial processes as a result of its openness, availability, and security. In this study, we used a hybrid adaptive crypto cloud framework that combines blockchain technology with multiple authorities and a botnet framework to improve cloud security while reducing computation time. The proposed adaptive crypto cloud system divides the cloud security framework into stages to organize secure data communication and reduce communication latency while detecting internal and external threats. In order to compute authentication using hash mapping and deploy an authentication system to safeguard the various users' authentication information, the whole system placed a major emphasis on block chain technology. The technology not only improves security but also makes the role-based access control system and anonymous authentication system more user-friendly.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"25 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":"131040164","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/ICCS54944.2021.00039
Aseem Kumar, A. Malik
The Coronavirus pandemic has made Irreversible changes in our society and the business world. Almost all aspects of the business and daily routine have shifted to the digital platforms and various forms of personal, indirect communications and suit the current environment in guarding us against coronavirus. The outbreak also brought a refreshing load of creativity in the people who found new ways to solve everyday problems. The key to solving problems is effective communication. With the help of mobile devices and computers, people were able to change their environment so that their expression of thoughts and their tasks of daily routine got aligned with social media platforms. People express themselves as if they are not going to get another chance to express themselves. They use doodles, poetic tweets, and many other forms of colloquial language. Using mixed language such as Hinglish became a norm for the commoner. In this research work, an attempt has been made to review techniques that can be used to work trust models from which meaning insights can be drawn in times such as covid-19 pandemic. From this study it can be inferred that no single approach of modeling complex scenarios such trust in times of covid-19 can be done. There is an urgent need to take inspiration from multiple techniques and approaches to assess the trust level in the digital society.
{"title":"A Comparative Analysis of Various Models for Assessment of Trust in Digital Age Accelerated by Covid-19","authors":"Aseem Kumar, A. Malik","doi":"10.1109/ICCS54944.2021.00039","DOIUrl":"https://doi.org/10.1109/ICCS54944.2021.00039","url":null,"abstract":"The Coronavirus pandemic has made Irreversible changes in our society and the business world. Almost all aspects of the business and daily routine have shifted to the digital platforms and various forms of personal, indirect communications and suit the current environment in guarding us against coronavirus. The outbreak also brought a refreshing load of creativity in the people who found new ways to solve everyday problems. The key to solving problems is effective communication. With the help of mobile devices and computers, people were able to change their environment so that their expression of thoughts and their tasks of daily routine got aligned with social media platforms. People express themselves as if they are not going to get another chance to express themselves. They use doodles, poetic tweets, and many other forms of colloquial language. Using mixed language such as Hinglish became a norm for the commoner. In this research work, an attempt has been made to review techniques that can be used to work trust models from which meaning insights can be drawn in times such as covid-19 pandemic. From this study it can be inferred that no single approach of modeling complex scenarios such trust in times of covid-19 can be done. There is an urgent need to take inspiration from multiple techniques and approaches to assess the trust level in the digital society.","PeriodicalId":340594,"journal":{"name":"2021 International Conference on Computing Sciences (ICCS)","volume":"5 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":"122361889","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：