Pub Date : 2022-12-14DOI: 10.1109/IC3I56241.2022.10072302
A. Pandey, Amit Barve
The skin is the body’s outermost layer, concealing/covering numerous biological organs, muscles, and other innumerable body parts. The study found that the body’s exposure to ultraviolet radiation is the main contributor to skin cancer (UV). There are several layers to the skin, but the epidermis and dermis are where cancer first appears. Changes in your skin or the appearance of a mole in many locations on your body are the most common warning signs. The only way to prevent cancer is to stay as far away from UV rays as you can, which would stop your skin from coming into contact with the disease. According to statistics, cases of this cancer have not only increased but are increasing swiftly as a result of the ozone layer’s deterioration, which causes it to stop emitting dangerous light and, as a result, come into contact with our skin. For the following issue, numerous different strategies include machine learning, DL, and TL are being used. Naive Bayes, logistic regression, random forest, decision tree, artificial NN, and convolutional NN are just a few of the numerous techniques used. The study makes an effort to put both TL and DL techniques to use in order to provide a result that shows which performs better for the next challenge.
{"title":"Skin Cancer Prediction Comparative Analysis using TL and NNs","authors":"A. Pandey, Amit Barve","doi":"10.1109/IC3I56241.2022.10072302","DOIUrl":"https://doi.org/10.1109/IC3I56241.2022.10072302","url":null,"abstract":"The skin is the body’s outermost layer, concealing/covering numerous biological organs, muscles, and other innumerable body parts. The study found that the body’s exposure to ultraviolet radiation is the main contributor to skin cancer (UV). There are several layers to the skin, but the epidermis and dermis are where cancer first appears. Changes in your skin or the appearance of a mole in many locations on your body are the most common warning signs. The only way to prevent cancer is to stay as far away from UV rays as you can, which would stop your skin from coming into contact with the disease. According to statistics, cases of this cancer have not only increased but are increasing swiftly as a result of the ozone layer’s deterioration, which causes it to stop emitting dangerous light and, as a result, come into contact with our skin. For the following issue, numerous different strategies include machine learning, DL, and TL are being used. Naive Bayes, logistic regression, random forest, decision tree, artificial NN, and convolutional NN are just a few of the numerous techniques used. The study makes an effort to put both TL and DL techniques to use in order to provide a result that shows which performs better for the next challenge.","PeriodicalId":274660,"journal":{"name":"2022 5th International Conference on Contemporary Computing and Informatics (IC3I)","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114768266","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 : 2022-12-14DOI: 10.1109/IC3I56241.2022.10072843
Supreet Kaur, Vinit Grewal
Through the Wireless Sensor Network (WSN), researchers have made every effort in advancing sensing technology worldwide. However, the essence of communication is deeply affected by the limited battery operating nature of sensor nodes. A lot of research efforts are reported that deal with this concern. Besides, the routing algorithms that tend to promise energy-efficient and optimized routing still fail to achieve optimized performance. Henceforth, sink mobility is one of the eminent solutions that tend to optimize the network through energy-saving routing strategies. In this paper, we have reviewed the sink mobility-based routing algorithms that are proposed for the Year 2022. We believe this review will help the readers to improvise the routing strategy by identifying the research gaps in the existing techniques.
{"title":"Review on sink mobility-based routing algorithms in WSN proposed in the Year 2022","authors":"Supreet Kaur, Vinit Grewal","doi":"10.1109/IC3I56241.2022.10072843","DOIUrl":"https://doi.org/10.1109/IC3I56241.2022.10072843","url":null,"abstract":"Through the Wireless Sensor Network (WSN), researchers have made every effort in advancing sensing technology worldwide. However, the essence of communication is deeply affected by the limited battery operating nature of sensor nodes. A lot of research efforts are reported that deal with this concern. Besides, the routing algorithms that tend to promise energy-efficient and optimized routing still fail to achieve optimized performance. Henceforth, sink mobility is one of the eminent solutions that tend to optimize the network through energy-saving routing strategies. In this paper, we have reviewed the sink mobility-based routing algorithms that are proposed for the Year 2022. We believe this review will help the readers to improvise the routing strategy by identifying the research gaps in the existing techniques.","PeriodicalId":274660,"journal":{"name":"2022 5th International Conference on Contemporary Computing and Informatics (IC3I)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115855251","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}
Bandpass filters, which only transmit frequencies that fall inside the transmission band and reject all other frequencies, are necessary for wireless communication systems. As 5G is set to be implemented, there will be a greater need for filters that operate in new frequency ranges. In 2022, the initial use is anticipated. The two key standards for filters that are used to build mobile applications are size and performance. With a repetition range of 26–28 GHz, the passband channel is designed in this proposal, simulated, and constructed. For downconversion of mmWave signals to microwave frequencies between 2 and 18 GHz, this channel can be employed as the front end of the apparatus. The size and efficiency of channels must be taken into account when planning new portable communications applications. Small, high-performance filters made by merging two components can be used in future mmWave applications like 5G. High-quality channels with minimal imprint are preferred for mmWave applications like 5G.
{"title":"A Millimeter Wave Filter for 5G Applications","authors":"Pradosh Kumar Sharma, A. Rana, Smita Sharma, Manish Sharma, Mesay Mengstie, Annam Takshitha Rao","doi":"10.1109/IC3I56241.2022.10072819","DOIUrl":"https://doi.org/10.1109/IC3I56241.2022.10072819","url":null,"abstract":"Bandpass filters, which only transmit frequencies that fall inside the transmission band and reject all other frequencies, are necessary for wireless communication systems. As 5G is set to be implemented, there will be a greater need for filters that operate in new frequency ranges. In 2022, the initial use is anticipated. The two key standards for filters that are used to build mobile applications are size and performance. With a repetition range of 26–28 GHz, the passband channel is designed in this proposal, simulated, and constructed. For downconversion of mmWave signals to microwave frequencies between 2 and 18 GHz, this channel can be employed as the front end of the apparatus. The size and efficiency of channels must be taken into account when planning new portable communications applications. Small, high-performance filters made by merging two components can be used in future mmWave applications like 5G. High-quality channels with minimal imprint are preferred for mmWave applications like 5G.","PeriodicalId":274660,"journal":{"name":"2022 5th International Conference on Contemporary Computing and Informatics (IC3I)","volume":"355 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115287411","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 : 2022-12-14DOI: 10.1109/IC3I56241.2022.10072528
M. Dhinakaran, P. Krishnapriya, Joel Alanya-Beltran, Vaibhav Gandhi, Sushma Jaiswal, D. P. Singh
The community health area comprises an enormous measure of data, and specific methodologies are utilized to deal with that data. One of the most common approaches is handling as well as processing. This technique forecasts the likely consequences of cardiovascular disease. The result of this strategy is to foresee the former heart disease. The work controls IOT utilizing a sensor (a heartbeat sensor to screen beats) and Arduino, and the outcomes might be seen on a successive screen. IFTTT is utilized to break down sensor readings in Google Sheets, which are accordingly changed over into CSV go-like information. The datasets utilized are characterized by treatment boundaries, in addition to being used for data preparation and testing. This technique assesses those boundaries utilizing the data arrangement request strategy. With artificial intelligence calculations and order work. The dataset is first taken apart, analyzed, and screened, after which the accumulated information is handled in Python programming utilizing AI Calculations, specifically Choice Tree Calculation with Irregular Woodlands Arrangement Calculation. SVM (Backing vector machine) creates the best outcomes concerning identifying coronary illness. Thus, the recommended worldview is demonstrated to be a solid one for foreseeing past coronary illness. The recommended equipment and programming innovation help patients in anticipating heart illness in its underlying stages.
{"title":"An Innovative Internet of Things (IoT) Computing-Based Health Monitoring System with the Aid of Machine Learning Approach","authors":"M. Dhinakaran, P. Krishnapriya, Joel Alanya-Beltran, Vaibhav Gandhi, Sushma Jaiswal, D. P. Singh","doi":"10.1109/IC3I56241.2022.10072528","DOIUrl":"https://doi.org/10.1109/IC3I56241.2022.10072528","url":null,"abstract":"The community health area comprises an enormous measure of data, and specific methodologies are utilized to deal with that data. One of the most common approaches is handling as well as processing. This technique forecasts the likely consequences of cardiovascular disease. The result of this strategy is to foresee the former heart disease. The work controls IOT utilizing a sensor (a heartbeat sensor to screen beats) and Arduino, and the outcomes might be seen on a successive screen. IFTTT is utilized to break down sensor readings in Google Sheets, which are accordingly changed over into CSV go-like information. The datasets utilized are characterized by treatment boundaries, in addition to being used for data preparation and testing. This technique assesses those boundaries utilizing the data arrangement request strategy. With artificial intelligence calculations and order work. The dataset is first taken apart, analyzed, and screened, after which the accumulated information is handled in Python programming utilizing AI Calculations, specifically Choice Tree Calculation with Irregular Woodlands Arrangement Calculation. SVM (Backing vector machine) creates the best outcomes concerning identifying coronary illness. Thus, the recommended worldview is demonstrated to be a solid one for foreseeing past coronary illness. The recommended equipment and programming innovation help patients in anticipating heart illness in its underlying stages.","PeriodicalId":274660,"journal":{"name":"2022 5th International Conference on Contemporary Computing and Informatics (IC3I)","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123410638","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 : 2022-12-14DOI: 10.1109/IC3I56241.2022.10072838
S. Kayalvizhi, S. Mythili
The recent electronic Medical monitoring system focuses on maintaining the database of patients with their medical history and up-to-date prescriptions and treatment maintained with high degree of confidentiality. The E-Medical services where Wireless Body Area Network (WBAN) is integrated face certain challenges such as data sharing, network reliability, database management and so on. The greatest threat in WBAN remained the data security and integrity. Several new methodologies supports to overcome these threats with secure hash function, digital signature with effective routing protocols to provide enhanced solutions for secure data maintenance and sharing in WBAN. The security system based on Block Chain Technology(BCT) are the most common trust building measure for WBAN, where the data are stored in distributed ledgers. This technology faces problems on minimal storage size and unauthorized access of the distributed data. This was overcome by the technique called sequential aggregate signature scheme with designated verifier with block chain based cloud transaction in WBAN in recent trends. The above scheme faces problem of wastage of storage space in the ledgers and it could be rectified by the proposed system of Encrypted Multi map block chain in WBAN which supports high degree of data access with reduced computational complexity of storage. This proposed system aims to simulate the WBAN with improved access of distributed authentication key of users in ledger pools and provides confidential sharing of data. This system deals with the comparative study of different encryption used in block chain storage to analyze their performance in WBAN
{"title":"A Survey on Encryption Algorithm with Hybrid Block Chain in Wireless Body Area Network","authors":"S. Kayalvizhi, S. Mythili","doi":"10.1109/IC3I56241.2022.10072838","DOIUrl":"https://doi.org/10.1109/IC3I56241.2022.10072838","url":null,"abstract":"The recent electronic Medical monitoring system focuses on maintaining the database of patients with their medical history and up-to-date prescriptions and treatment maintained with high degree of confidentiality. The E-Medical services where Wireless Body Area Network (WBAN) is integrated face certain challenges such as data sharing, network reliability, database management and so on. The greatest threat in WBAN remained the data security and integrity. Several new methodologies supports to overcome these threats with secure hash function, digital signature with effective routing protocols to provide enhanced solutions for secure data maintenance and sharing in WBAN. The security system based on Block Chain Technology(BCT) are the most common trust building measure for WBAN, where the data are stored in distributed ledgers. This technology faces problems on minimal storage size and unauthorized access of the distributed data. This was overcome by the technique called sequential aggregate signature scheme with designated verifier with block chain based cloud transaction in WBAN in recent trends. The above scheme faces problem of wastage of storage space in the ledgers and it could be rectified by the proposed system of Encrypted Multi map block chain in WBAN which supports high degree of data access with reduced computational complexity of storage. This proposed system aims to simulate the WBAN with improved access of distributed authentication key of users in ledger pools and provides confidential sharing of data. This system deals with the comparative study of different encryption used in block chain storage to analyze their performance in WBAN","PeriodicalId":274660,"journal":{"name":"2022 5th International Conference on Contemporary Computing and Informatics (IC3I)","volume":"335 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123512444","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 : 2022-12-14DOI: 10.1109/IC3I56241.2022.10072980
Vishal Suthar, V. Bansal, C. Reddy, J. L. Arias-Gonzáles, Devendra Singh, D. P. Singh
The development of blockchain technology (BT) in recent years has made it a distinctive, revolutionary, and popular innovation. Information security and confidentiality are prioritised by the decentralised database in BT. Additionally, the consensus process in it ensures the validity and security of the data. However, it brings up fresh security concerns including majority assault and the double expenditures. Data analytics using cryptocurrency sensitive data are needed to address the aforementioned problems. These dataset' analytics highlight the value of recently developed techniques such as machine learning (ML). ML uses a reasonable quantity of data to generate accurate predictions. In ML, data exchange and dependability are essential to enhancing the precision of outcomes. Results from the fusion of these two technologies (ML and BT) may be quite exact. In this research, we give a thorough investigation into the use of machine learning (ML) to strengthen the security of BT-based intelligent systems. The assaults on a blockchain-based network may be analysed using a variety of classic machine learning (ML) approaches, including Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Clustering, Bagging, and Support Vector Machines (SVM) (LSTM).We also discuss how the two technologies may be used together in a number of advanced areas, including smart urban, the national grid, medicine, and autonomous aerial vehicles (UAVs). The difficulties and concerns facing future research are then examined. Finally, a study based with a thorough analysis is offered.
{"title":"Machine Learning Adoption in Blockchain-Based Smart Applications","authors":"Vishal Suthar, V. Bansal, C. Reddy, J. L. Arias-Gonzáles, Devendra Singh, D. P. Singh","doi":"10.1109/IC3I56241.2022.10072980","DOIUrl":"https://doi.org/10.1109/IC3I56241.2022.10072980","url":null,"abstract":"The development of blockchain technology (BT) in recent years has made it a distinctive, revolutionary, and popular innovation. Information security and confidentiality are prioritised by the decentralised database in BT. Additionally, the consensus process in it ensures the validity and security of the data. However, it brings up fresh security concerns including majority assault and the double expenditures. Data analytics using cryptocurrency sensitive data are needed to address the aforementioned problems. These dataset' analytics highlight the value of recently developed techniques such as machine learning (ML). ML uses a reasonable quantity of data to generate accurate predictions. In ML, data exchange and dependability are essential to enhancing the precision of outcomes. Results from the fusion of these two technologies (ML and BT) may be quite exact. In this research, we give a thorough investigation into the use of machine learning (ML) to strengthen the security of BT-based intelligent systems. The assaults on a blockchain-based network may be analysed using a variety of classic machine learning (ML) approaches, including Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Clustering, Bagging, and Support Vector Machines (SVM) (LSTM).We also discuss how the two technologies may be used together in a number of advanced areas, including smart urban, the national grid, medicine, and autonomous aerial vehicles (UAVs). The difficulties and concerns facing future research are then examined. Finally, a study based with a thorough analysis is offered.","PeriodicalId":274660,"journal":{"name":"2022 5th International Conference on Contemporary Computing and Informatics (IC3I)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121728257","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 : 2022-12-14DOI: 10.1109/IC3I56241.2022.10073318
M. Meenaakumari, P. Jayasuriya, Nasa Dhanraj, Seema Sharma, Geetha Manoharan, M. Tiwari
Banks serves the basic necessities of everyone next to hospitals and schools. People reach out to banks for various purposes. But one of the most common services offered by banks is loans. However, many common people are not completely aware of the banking procedures and eligibility criteria for loans. This study aims to develop a Machine Learning (ML) model which is capable of predicting whether the person is eligible for a health loan or not by analyzing some basic values entered by the user. For this process, a dataset consisting of all necessary parameters for a loan application is collected from Kaggle. The collected dataset is then preprocessed by two methods namely the null value elimination method and encoding. Simultaneously, three ML models were developed using three different algorithms. They are the Random Forest (RF), Naive Bayes (NB), and Linear Regression (LR). The preprocessed data will next be used to train the models. Following that, a comparison of a few parameters will be used to assess the models' effectiveness. The results of the analysis prove that the RF algorithm is the best in terms of both accuracy and error. The accuracy of the RF algorithm is 91% and it also predicts loan eligibility with lesser error values. The LR model has the lowest accuracy values and the highest error value making it the least efficient algorithm that can be used in loan prediction.
{"title":"Loan Eligibility Prediction using Machine Learning based on Personal Information","authors":"M. Meenaakumari, P. Jayasuriya, Nasa Dhanraj, Seema Sharma, Geetha Manoharan, M. Tiwari","doi":"10.1109/IC3I56241.2022.10073318","DOIUrl":"https://doi.org/10.1109/IC3I56241.2022.10073318","url":null,"abstract":"Banks serves the basic necessities of everyone next to hospitals and schools. People reach out to banks for various purposes. But one of the most common services offered by banks is loans. However, many common people are not completely aware of the banking procedures and eligibility criteria for loans. This study aims to develop a Machine Learning (ML) model which is capable of predicting whether the person is eligible for a health loan or not by analyzing some basic values entered by the user. For this process, a dataset consisting of all necessary parameters for a loan application is collected from Kaggle. The collected dataset is then preprocessed by two methods namely the null value elimination method and encoding. Simultaneously, three ML models were developed using three different algorithms. They are the Random Forest (RF), Naive Bayes (NB), and Linear Regression (LR). The preprocessed data will next be used to train the models. Following that, a comparison of a few parameters will be used to assess the models' effectiveness. The results of the analysis prove that the RF algorithm is the best in terms of both accuracy and error. The accuracy of the RF algorithm is 91% and it also predicts loan eligibility with lesser error values. The LR model has the lowest accuracy values and the highest error value making it the least efficient algorithm that can be used in loan prediction.","PeriodicalId":274660,"journal":{"name":"2022 5th International Conference on Contemporary Computing and Informatics (IC3I)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122178480","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 : 2022-12-14DOI: 10.1109/IC3I56241.2022.10072927
Nagothi Vaibhav Anjani Kumar, S. Mehrotra
Digital text data is increasing daily in various uses, such as clinical notes, lab test reports, research articles, etc. Most of the mentioned data are unstructured. While searching for information lot of unrelated information is returned against the query. The paper presents a comparative analysis of word embedding techniques and text similarity measures to determine how similar two bits of text are in respective lexical, semantic characteristics, and closeness. The principal aim of this paper is to perform pre-processing process of medical history notes of the patient's data followed by word embedding techniques such as Word2Vec, FastText, and Doc2Vec.
{"title":"A Comparative Analysis of word embedding techniques and text similarity Measures","authors":"Nagothi Vaibhav Anjani Kumar, S. Mehrotra","doi":"10.1109/IC3I56241.2022.10072927","DOIUrl":"https://doi.org/10.1109/IC3I56241.2022.10072927","url":null,"abstract":"Digital text data is increasing daily in various uses, such as clinical notes, lab test reports, research articles, etc. Most of the mentioned data are unstructured. While searching for information lot of unrelated information is returned against the query. The paper presents a comparative analysis of word embedding techniques and text similarity measures to determine how similar two bits of text are in respective lexical, semantic characteristics, and closeness. The principal aim of this paper is to perform pre-processing process of medical history notes of the patient's data followed by word embedding techniques such as Word2Vec, FastText, and Doc2Vec.","PeriodicalId":274660,"journal":{"name":"2022 5th International Conference on Contemporary Computing and Informatics (IC3I)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121412895","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 : 2022-12-14DOI: 10.1109/IC3I56241.2022.10072818
Anupam Singh, Abhishek Kumar, H. M. Salman, Navneet Rawat, Sanjiv Kumar Jain, Annam Takshitha Rao
The ability to identify and categorize bacteria is crucial in modern medicine for disease diagnosis, infection treatment, and epidemic investigation. However, manually identification and categorization of bacteria requires a lot of time and effort from humans. As technology has progressed, computer systems-based techniques are now doing the duty of identifying images captured by digital electron microscopes. On top of that, modern Deep Learning (DL) methods have shown remarkable improvement in the area of image classification. In this research, we explore a method for using a DL model to automate the identification and categorization of bacteria. To develop the DL model, we used a dataset consisting of more than 600 images of 33 distinct bacteria taken with a microscope and the ‘transfer learning’ technique. GoogLeNet and AlexNet are two examples of transfer learning models used in this research. The DL classification accuracy was evaluated using 20% randomly selected and isolated images from the dataset. Experimental findings of prediction obtained an accuracy of roughly 98.67% by GoogLeNet, and both transfer learning models recognized and classified all 33 bacterial species with better success rates.
{"title":"Transfer Learning Approach on Bacteria Classification from Microscopic Images","authors":"Anupam Singh, Abhishek Kumar, H. M. Salman, Navneet Rawat, Sanjiv Kumar Jain, Annam Takshitha Rao","doi":"10.1109/IC3I56241.2022.10072818","DOIUrl":"https://doi.org/10.1109/IC3I56241.2022.10072818","url":null,"abstract":"The ability to identify and categorize bacteria is crucial in modern medicine for disease diagnosis, infection treatment, and epidemic investigation. However, manually identification and categorization of bacteria requires a lot of time and effort from humans. As technology has progressed, computer systems-based techniques are now doing the duty of identifying images captured by digital electron microscopes. On top of that, modern Deep Learning (DL) methods have shown remarkable improvement in the area of image classification. In this research, we explore a method for using a DL model to automate the identification and categorization of bacteria. To develop the DL model, we used a dataset consisting of more than 600 images of 33 distinct bacteria taken with a microscope and the ‘transfer learning’ technique. GoogLeNet and AlexNet are two examples of transfer learning models used in this research. The DL classification accuracy was evaluated using 20% randomly selected and isolated images from the dataset. Experimental findings of prediction obtained an accuracy of roughly 98.67% by GoogLeNet, and both transfer learning models recognized and classified all 33 bacterial species with better success rates.","PeriodicalId":274660,"journal":{"name":"2022 5th International Conference on Contemporary Computing and Informatics (IC3I)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122547063","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 : 2022-12-14DOI: 10.1109/IC3I56241.2022.10073416
Alpana Sharma, S. Poojitha, Archana B. Saxena, M. Bhanushali, Priyanka Rawal
Man-made brainpower has been underexplored. Machines with profound learning skills can take advanced showcasing higher than ever with their Man-made consciousness having a significant effect. This paper tries to find discoveries from an investigation of responses across various socioeconomics to robots and their selling powers. It’s also been discovered that software engineers need to work in tandem with digital marketers using machines with deep learning to consider consumer attitudes, behaviors, and preferences while designing the architecture. Because of this, in the future, marketers will have far more access to correct information on customers, which will have enormous advantages for the company. While search engine marketing automation has a lot of potential, marketers know that humans still have an essential role to play in the formulation of abstract strategies. The review that is being given here centres around how promoting offices, media organizations, and sponsors use and utilize ML-driven investigation instruments. The exploration features four central issues, including 1) the meaning of wise scientific apparatuses in the turn of events and execution of promoting strategies; 2) the absence of familiarity with arising advancements, for example, Al (ML) and man-made consciousness (simulated intelligence); 3) the imminent utilization of ML instruments in publicizing; and 4) the low degree of improvement and use of ML-driven logical devices in advertising. To help organizations in distinguishing valuable open doors and completing drives zeroed in on the sending and acknowledgement of quantitative ML devices in computerized showcasing, a system comprised of facilitators and a task plan was laid out.Data collection and analysis will perform using SPSS software, and findings will be drawn from a combination of a fuzzy-approach to determining how best to persuade customers to utilize the machine’s services and a variable-oriented, quantitative examination of the obtained data will consider.
{"title":"A Conceptual Analysis of Machine Learning Towards Digital Marketing Transformation","authors":"Alpana Sharma, S. Poojitha, Archana B. Saxena, M. Bhanushali, Priyanka Rawal","doi":"10.1109/IC3I56241.2022.10073416","DOIUrl":"https://doi.org/10.1109/IC3I56241.2022.10073416","url":null,"abstract":"Man-made brainpower has been underexplored. Machines with profound learning skills can take advanced showcasing higher than ever with their Man-made consciousness having a significant effect. This paper tries to find discoveries from an investigation of responses across various socioeconomics to robots and their selling powers. It’s also been discovered that software engineers need to work in tandem with digital marketers using machines with deep learning to consider consumer attitudes, behaviors, and preferences while designing the architecture. Because of this, in the future, marketers will have far more access to correct information on customers, which will have enormous advantages for the company. While search engine marketing automation has a lot of potential, marketers know that humans still have an essential role to play in the formulation of abstract strategies. The review that is being given here centres around how promoting offices, media organizations, and sponsors use and utilize ML-driven investigation instruments. The exploration features four central issues, including 1) the meaning of wise scientific apparatuses in the turn of events and execution of promoting strategies; 2) the absence of familiarity with arising advancements, for example, Al (ML) and man-made consciousness (simulated intelligence); 3) the imminent utilization of ML instruments in publicizing; and 4) the low degree of improvement and use of ML-driven logical devices in advertising. To help organizations in distinguishing valuable open doors and completing drives zeroed in on the sending and acknowledgement of quantitative ML devices in computerized showcasing, a system comprised of facilitators and a task plan was laid out.Data collection and analysis will perform using SPSS software, and findings will be drawn from a combination of a fuzzy-approach to determining how best to persuade customers to utilize the machine’s services and a variable-oriented, quantitative examination of the obtained data will consider.","PeriodicalId":274660,"journal":{"name":"2022 5th International Conference on Contemporary Computing and Informatics (IC3I)","volume":"288 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122811764","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：