Pub Date : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10141166
Koushik K S, Ankita Mahale, Shobha Rani N
One of the most important tasks in the realm of document analysis and recognition is the detection of equations in documents that were acquired using a camera. The procedure includes several steps, including pre-processing of the images, segmentation, feature extraction, and classification. The suggested method comprises taking a user-provided input expression image and classifying it into one of three types of equations: simple, complex, and highly complex. By choosing a decision boundary set off from the initial hyperplane, the SVR algorithm encodes the image, producing a model that fits the data better. The result is then obtained by character-wise segmenting the image and comparing it with trained models. Two recurrent neural networks make up the RNN encoder-decoder that is used. One RNN creates a fixed-length vector representation from a sequence of symbols, and a different RNN decodes that representation into a different sequence of symbols. 1900 images containing various equations made up the dataset utilized for training, validating, and testing the SVR and RNN. The accuracy of the system was about 93.64%.
{"title":"Equation Detection in the Camera Captured Handwritten Document","authors":"Koushik K S, Ankita Mahale, Shobha Rani N","doi":"10.1109/ICAAIC56838.2023.10141166","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10141166","url":null,"abstract":"One of the most important tasks in the realm of document analysis and recognition is the detection of equations in documents that were acquired using a camera. The procedure includes several steps, including pre-processing of the images, segmentation, feature extraction, and classification. The suggested method comprises taking a user-provided input expression image and classifying it into one of three types of equations: simple, complex, and highly complex. By choosing a decision boundary set off from the initial hyperplane, the SVR algorithm encodes the image, producing a model that fits the data better. The result is then obtained by character-wise segmenting the image and comparing it with trained models. Two recurrent neural networks make up the RNN encoder-decoder that is used. One RNN creates a fixed-length vector representation from a sequence of symbols, and a different RNN decodes that representation into a different sequence of symbols. 1900 images containing various equations made up the dataset utilized for training, validating, and testing the SVR and RNN. The accuracy of the system was about 93.64%.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"405 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127597687","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140470
Arun Biradar, M. Chandan, Y. Raghavendra, K. Chidambarathanu, I. Thamarai, Anuj Raturi
The advent of software-defined networking and virtualization of network functions has brought numerous advantages; however, to achieve the flexibility and programmability envisaged in these technologies, new components in the control and management planes were introduced. Such components require fast recovery because without management the entire data plane is inoperable. To deal with flaws in these plans, the self-healing technique is used, explored in the work that is summarized in this document. The results prove the self-healing efficiency in network slices with strict quality requirements and also demonstrate that the introduced framework is capable of self-healing, that is, healing the degraded environment as well as healing itself.
{"title":"Self-Healing for Software Defined Networking","authors":"Arun Biradar, M. Chandan, Y. Raghavendra, K. Chidambarathanu, I. Thamarai, Anuj Raturi","doi":"10.1109/ICAAIC56838.2023.10140470","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140470","url":null,"abstract":"The advent of software-defined networking and virtualization of network functions has brought numerous advantages; however, to achieve the flexibility and programmability envisaged in these technologies, new components in the control and management planes were introduced. Such components require fast recovery because without management the entire data plane is inoperable. To deal with flaws in these plans, the self-healing technique is used, explored in the work that is summarized in this document. The results prove the self-healing efficiency in network slices with strict quality requirements and also demonstrate that the introduced framework is capable of self-healing, that is, healing the degraded environment as well as healing itself.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"121 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132702953","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10141462
A. Lakshmanarao, M.Naveen Kumar, K.S.V. Ratnakar, Y. Satwika
India's economy is heavily dependent on agriculture, and this study report tries to increase agricultural productivity by forecasting crop yields for a range of crops farmed there. This study is unique in that it forecasts agricultural yields for any chosen time period throughout the year by using simple factors like, district, area, season and State. The article forecasts agricultural production using modern regression techniques including Lasso, Kernel Ridge, and Elastic-Net Regression designs. The idea of Stacking Regression is also used to improve the performance of the designs and provide more accurate forecasts. This research provides a positive breakthrough for India's agricultural industry, with the potential to deliver major advantages for farmers and the larger economy. This study provides a useful tool for improving crop yield projections and eventually increasing agricultural output in the nation by employing cutting-edge analytical methodologies and simple input parameters. Informed decisions regarding crop cultivation, fertilization, and harvest may be made by farmers with the help of technology and data-driven insights, resulting in higher yields and more favorable economic consequences.
{"title":"Crop Yield Prediction using Regression Models in Machine Learning","authors":"A. Lakshmanarao, M.Naveen Kumar, K.S.V. Ratnakar, Y. Satwika","doi":"10.1109/ICAAIC56838.2023.10141462","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10141462","url":null,"abstract":"India's economy is heavily dependent on agriculture, and this study report tries to increase agricultural productivity by forecasting crop yields for a range of crops farmed there. This study is unique in that it forecasts agricultural yields for any chosen time period throughout the year by using simple factors like, district, area, season and State. The article forecasts agricultural production using modern regression techniques including Lasso, Kernel Ridge, and Elastic-Net Regression designs. The idea of Stacking Regression is also used to improve the performance of the designs and provide more accurate forecasts. This research provides a positive breakthrough for India's agricultural industry, with the potential to deliver major advantages for farmers and the larger economy. This study provides a useful tool for improving crop yield projections and eventually increasing agricultural output in the nation by employing cutting-edge analytical methodologies and simple input parameters. Informed decisions regarding crop cultivation, fertilization, and harvest may be made by farmers with the help of technology and data-driven insights, resulting in higher yields and more favorable economic consequences.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"56 15","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133086663","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10141450
B. Rakesh, D. Ragavi, M. K. Reddy, G. L. Sumalata
Microvascular leakage within the retina causes the illness known as diabetic retinopathy (DR) in the eye. For people with diabetes mellitus (DM), diabetic retinopathy is the main reason for vision loss. This Disease is a global health issue, as the condition can lead to long-term disability and decreased quality of life for affected individuals. As a result, It causes microvascular issues and irreversible vision loss due to increase in sugar levels. Unfortunately, the accuracy of existing approaches is limited because of issues such as inadequate contrast, imaging quality, and lesion unpredictability. We propose a VGG-19 convolutional neural network technique for the identification and classification of NPDR in this research. Overcoming these obstacles, our goal is to design a system that can detect and classify NPDR from retinal pictures. Our findings show that our proposed technique is effective in reaching high accuracy and might potentially contribute to the early identification and treatment of NPDR. We also created a user interface for classification and detection of the severity of the disease.
{"title":"Detection and Classification of Non-Proliferation Diabetic Retinopathy using VGG-19 CNN Algorithm","authors":"B. Rakesh, D. Ragavi, M. K. Reddy, G. L. Sumalata","doi":"10.1109/ICAAIC56838.2023.10141450","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10141450","url":null,"abstract":"Microvascular leakage within the retina causes the illness known as diabetic retinopathy (DR) in the eye. For people with diabetes mellitus (DM), diabetic retinopathy is the main reason for vision loss. This Disease is a global health issue, as the condition can lead to long-term disability and decreased quality of life for affected individuals. As a result, It causes microvascular issues and irreversible vision loss due to increase in sugar levels. Unfortunately, the accuracy of existing approaches is limited because of issues such as inadequate contrast, imaging quality, and lesion unpredictability. We propose a VGG-19 convolutional neural network technique for the identification and classification of NPDR in this research. Overcoming these obstacles, our goal is to design a system that can detect and classify NPDR from retinal pictures. Our findings show that our proposed technique is effective in reaching high accuracy and might potentially contribute to the early identification and treatment of NPDR. We also created a user interface for classification and detection of the severity of the disease.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133599688","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140561
Dileep Kumar Boyapati, Jagathi Gottipati, Vinod Kattula, S. Yelisetti
Sentiment analysis, commonly referred to as opinion mining, reveals the attitudes and feelings of consumers about specific goods or services. The sentiment polarity classification, which identifies whether a review is favourable, negative, or neutral, is the fundamental issue with sentiment analysis. There are still some study gaps, as some studies only investigate the positive, neutral, and negative sentiment classes; none of these studies considered more than three classes; also, none of these studies considered the individual and combined effects of the sentiment polarity aspects. No prior method took into account the verb, adverb, adjective, and their combinations, as well as the five sentiment classes and three sentiment polarity traits. This study, provides a method for categorizing online reviews of Instant Videos based on their sentiment. Proposed study makes use of a substantial data set of 500,000 internet reviews. This review-level categorization process Adjective, verb, and two polarity traits are taken into account additionally as well as their pairings with various senses.
{"title":"Sentiment Polarity Categorization of Product Reviews using Twitter Data","authors":"Dileep Kumar Boyapati, Jagathi Gottipati, Vinod Kattula, S. Yelisetti","doi":"10.1109/ICAAIC56838.2023.10140561","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140561","url":null,"abstract":"Sentiment analysis, commonly referred to as opinion mining, reveals the attitudes and feelings of consumers about specific goods or services. The sentiment polarity classification, which identifies whether a review is favourable, negative, or neutral, is the fundamental issue with sentiment analysis. There are still some study gaps, as some studies only investigate the positive, neutral, and negative sentiment classes; none of these studies considered more than three classes; also, none of these studies considered the individual and combined effects of the sentiment polarity aspects. No prior method took into account the verb, adverb, adjective, and their combinations, as well as the five sentiment classes and three sentiment polarity traits. This study, provides a method for categorizing online reviews of Instant Videos based on their sentiment. Proposed study makes use of a substantial data set of 500,000 internet reviews. This review-level categorization process Adjective, verb, and two polarity traits are taken into account additionally as well as their pairings with various senses.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131841828","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140622
Raksha Pandey, A. Kushwaha, Suraj Sharma, Ankit Anand, Suraj Kumar
With the increase in sharing of videos worldwide over social networks, presence of high-quality fakes is on increase. Forged videos affect the authenticity and integrity of the video as a whole. This can lead to serious implications. For example, in case of video to be used in courts as an evidence, presence of forgery can implicate innocents or help criminal to escape justice. This calls for the detection mechanisms to counter. This leads to the discovery of several different approaches to detect copy-move forgery by analysing the side effects due to tempering. One of the most common approaches is copy-move video forgery which consists of duplicating area of frame. Traditional approach detects for patterns related to duplication manually which is not so successful. In contrast, methods related to deep learning gives better results. Therefore, this research follows deep learning model using pertained architecture to detect copy-move video forgery.
{"title":"Intra-frame Copy-move Video Forgery Detection","authors":"Raksha Pandey, A. Kushwaha, Suraj Sharma, Ankit Anand, Suraj Kumar","doi":"10.1109/ICAAIC56838.2023.10140622","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140622","url":null,"abstract":"With the increase in sharing of videos worldwide over social networks, presence of high-quality fakes is on increase. Forged videos affect the authenticity and integrity of the video as a whole. This can lead to serious implications. For example, in case of video to be used in courts as an evidence, presence of forgery can implicate innocents or help criminal to escape justice. This calls for the detection mechanisms to counter. This leads to the discovery of several different approaches to detect copy-move forgery by analysing the side effects due to tempering. One of the most common approaches is copy-move video forgery which consists of duplicating area of frame. Traditional approach detects for patterns related to duplication manually which is not so successful. In contrast, methods related to deep learning gives better results. Therefore, this research follows deep learning model using pertained architecture to detect copy-move video forgery.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133843011","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140405
D. K, D. M, Mangaladharsini L. G, Devipriya R, V. V.
The skin, being the largest organ in the human body, plays a crucial role in protecting and covering the body while performing various functions. However, skin diseases, such as vitiligo, can result in changes to the skin's appearance, leading to white patches. Vitiligo is a prevalent skin disorder affecting millions of individuals worldwide. Despite the lack of a cure for vitiligo, early detection and treatment can prevent its dissemination to other body parts. To address this issue, an innovative system has been developed to enable users to check their skin condition for the presence of vitiligo in a user-friendly manner. This system comprises both hardware and software components. Specifically, a color sensor is utilized to gather RGB values of the user's skin surface, which are subsequently analyzed using a machine learning algorithm to ascertain the presence or absence of vitiligo. The device offers an easy-to-use tool for users to monitor their skin condition, which could significantly improve the quality of life for those affected by vitiligo comprehensive data collection and analysis.
{"title":"IoT based Vitiligo Detection","authors":"D. K, D. M, Mangaladharsini L. G, Devipriya R, V. V.","doi":"10.1109/ICAAIC56838.2023.10140405","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140405","url":null,"abstract":"The skin, being the largest organ in the human body, plays a crucial role in protecting and covering the body while performing various functions. However, skin diseases, such as vitiligo, can result in changes to the skin's appearance, leading to white patches. Vitiligo is a prevalent skin disorder affecting millions of individuals worldwide. Despite the lack of a cure for vitiligo, early detection and treatment can prevent its dissemination to other body parts. To address this issue, an innovative system has been developed to enable users to check their skin condition for the presence of vitiligo in a user-friendly manner. This system comprises both hardware and software components. Specifically, a color sensor is utilized to gather RGB values of the user's skin surface, which are subsequently analyzed using a machine learning algorithm to ascertain the presence or absence of vitiligo. The device offers an easy-to-use tool for users to monitor their skin condition, which could significantly improve the quality of life for those affected by vitiligo comprehensive data collection and analysis.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"764 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116133892","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10141080
Venkata Subba Reddy Gade, M. Sumathi
Speaker recognition depends on identifying the speaker using particular segments of the sound stream. A single speech characteristic only reveals the speaker's identity partially. Current advances in machine learning have considerably enhanced automatic voice recognition and localization systems. Nevertheless, this advantage comes at the expense of requiring complicated models and calculations. Additional microphone arrays will be used, as well as practice data. This study introduces a novel deep convolutional neural network-based end-to-end hybrid identification and localization model (HDCNN). HDCNN are employing a cutting-edge data augmentation strategy. This model can recognize both single- and multi-speaker arrangements and show which speaker is active with outstanding accuracy. HDCNN, a hybrid machine-learning algorithm. The final outcomes of proposed HDCNN model show greatest performance with an accuracy of 98.33%, which is higher than existing model's performance metrics.
{"title":"Hybrid Deep Convolutional Neural Network based Speaker Recognition for Noisy Speech Environments","authors":"Venkata Subba Reddy Gade, M. Sumathi","doi":"10.1109/ICAAIC56838.2023.10141080","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10141080","url":null,"abstract":"Speaker recognition depends on identifying the speaker using particular segments of the sound stream. A single speech characteristic only reveals the speaker's identity partially. Current advances in machine learning have considerably enhanced automatic voice recognition and localization systems. Nevertheless, this advantage comes at the expense of requiring complicated models and calculations. Additional microphone arrays will be used, as well as practice data. This study introduces a novel deep convolutional neural network-based end-to-end hybrid identification and localization model (HDCNN). HDCNN are employing a cutting-edge data augmentation strategy. This model can recognize both single- and multi-speaker arrangements and show which speaker is active with outstanding accuracy. HDCNN, a hybrid machine-learning algorithm. The final outcomes of proposed HDCNN model show greatest performance with an accuracy of 98.33%, which is higher than existing model's performance metrics.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"266 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123293123","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10140446
Kummari Karthik, Alla Lokesh Reddy, Rithesh Kulkarni, Mohd. Javeed Mehdi
Recent studies say that heart diseases are the major threat to humans. The diagnosis of the disease is obtained by making predictions from the patient's medical details. A minor error in predicting or diagnosis the results of heart related diseases can cause several problems. To address the issue, several researchers used the hospital data or patients' information for data mining and statistical tools for helping the health care system in the diagnosis of heart diseases. For making people aware of heart disease, a prediction model is required for early detection. The prediction model uses the training data and predicts the results by using several machine learning techniques. Using this training data, the testing of the other data is done precisely. In this research, for the prediction of the results from the given data, machine learning algorithms are used for model development. The prediction includes accuracy of each algorithm. By using machine learning techniques, the correlation between various features present in the dataset is also identified in the research while performing the experiment. The framework makes use of 13 features, including ones related to age, gender, obesity, blood pressure, cholesterol, and cp as various attributes to generate the classifiers. Using these features the output of these classifiers reveals that the accuracy of each algorithm and assists in predicting risk factors related to heart diseases and gives which is best suitable technique for producing the best predictions.
{"title":"Algorithm Accuracy Verification in Heart Disease Analysis using Machine Learning","authors":"Kummari Karthik, Alla Lokesh Reddy, Rithesh Kulkarni, Mohd. Javeed Mehdi","doi":"10.1109/ICAAIC56838.2023.10140446","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10140446","url":null,"abstract":"Recent studies say that heart diseases are the major threat to humans. The diagnosis of the disease is obtained by making predictions from the patient's medical details. A minor error in predicting or diagnosis the results of heart related diseases can cause several problems. To address the issue, several researchers used the hospital data or patients' information for data mining and statistical tools for helping the health care system in the diagnosis of heart diseases. For making people aware of heart disease, a prediction model is required for early detection. The prediction model uses the training data and predicts the results by using several machine learning techniques. Using this training data, the testing of the other data is done precisely. In this research, for the prediction of the results from the given data, machine learning algorithms are used for model development. The prediction includes accuracy of each algorithm. By using machine learning techniques, the correlation between various features present in the dataset is also identified in the research while performing the experiment. The framework makes use of 13 features, including ones related to age, gender, obesity, blood pressure, cholesterol, and cp as various attributes to generate the classifiers. Using these features the output of these classifiers reveals that the accuracy of each algorithm and assists in predicting risk factors related to heart diseases and gives which is best suitable technique for producing the best predictions.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124453266","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 : 2023-05-04DOI: 10.1109/ICAAIC56838.2023.10141427
Ananya Singhai, S. Aanjankumar, S. Poonkuntran
Credit cards offer a convenient and efficient option for online transactions; however, their increasing use has led to a rise in credit card fraud, resulting in significant financial losses for both cardholders and financial institutions. This research aims to identify such frauds by considering various criteria, including the availability of public data, high-class disparity statistics, changes in fraudulent processes, and high false alarm rates. With the growth of e-payments, fraudsters have resorted to various tactics such as fake emails and data breaches to steal money during online transactions. Although these methods are inaccurate, cutting-edge machine-learning algorithms must be used to reduce fraud losses. Therefore, this study's primary focus is on the recent advancements in machine learning algorithms for credit card fraud detection. The research paper aims to investigate the application of machine learning algorithms in distinguishing between genuine and fake online transactions. In the paper, KNN is compared to other machine-learning methods for detecting credit card fraud. The proposed approach achieved an accuracy of 99.95%, a precision of 97.2%, a recall of 85.71%, and an F1-score of 90.3%.
{"title":"A Novel Methodology for Credit Card Fraud Detection using KNN Dependent Machine Learning Methodology","authors":"Ananya Singhai, S. Aanjankumar, S. Poonkuntran","doi":"10.1109/ICAAIC56838.2023.10141427","DOIUrl":"https://doi.org/10.1109/ICAAIC56838.2023.10141427","url":null,"abstract":"Credit cards offer a convenient and efficient option for online transactions; however, their increasing use has led to a rise in credit card fraud, resulting in significant financial losses for both cardholders and financial institutions. This research aims to identify such frauds by considering various criteria, including the availability of public data, high-class disparity statistics, changes in fraudulent processes, and high false alarm rates. With the growth of e-payments, fraudsters have resorted to various tactics such as fake emails and data breaches to steal money during online transactions. Although these methods are inaccurate, cutting-edge machine-learning algorithms must be used to reduce fraud losses. Therefore, this study's primary focus is on the recent advancements in machine learning algorithms for credit card fraud detection. The research paper aims to investigate the application of machine learning algorithms in distinguishing between genuine and fake online transactions. In the paper, KNN is compared to other machine-learning methods for detecting credit card fraud. The proposed approach achieved an accuracy of 99.95%, a precision of 97.2%, a recall of 85.71%, and an F1-score of 90.3%.","PeriodicalId":267906,"journal":{"name":"2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125161402","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：