Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037435
Abhimanyu Singh, M. Nene
Object Detection (OD) in natural images has made tremendous strides during the last ten years. However, the outcomes are infrequently adequate when the natural image OD approach is used straight to Satellite Images (SI). This results from the intrinsic differences in object scale and orientation introduced by the omniscient viewpoint of the SI. Detecting objects is a challenging task especially when small object areas and complicated backgrounds appear in satellite images under analysis. Occlusion and intense item overlap have a further negative effect on the detection performance. The self-attention mechanisms are proposed to search for minute details in an image. However such searches mechanism come with complexity or high computational cost due to uncertainty induced in visual resolutions. The study in this research paper addresses the problems experienced in the accuracy and precision and the efficacy of the proposed model is demonstrated with the result in this paper.
{"title":"Detection of Multiclass Objects in Satellite Images Using an Improved Algorithmic Approach","authors":"Abhimanyu Singh, M. Nene","doi":"10.1109/IBSSC56953.2022.10037435","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037435","url":null,"abstract":"Object Detection (OD) in natural images has made tremendous strides during the last ten years. However, the outcomes are infrequently adequate when the natural image OD approach is used straight to Satellite Images (SI). This results from the intrinsic differences in object scale and orientation introduced by the omniscient viewpoint of the SI. Detecting objects is a challenging task especially when small object areas and complicated backgrounds appear in satellite images under analysis. Occlusion and intense item overlap have a further negative effect on the detection performance. The self-attention mechanisms are proposed to search for minute details in an image. However such searches mechanism come with complexity or high computational cost due to uncertainty induced in visual resolutions. The study in this research paper addresses the problems experienced in the accuracy and precision and the efficacy of the proposed model is demonstrated with the result in this paper.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134150609","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 : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037421
D. Pawade, Avani M. Sakhapara, Rujuta Ashtekar, Diya Bakhai, Shruti Tyagi
Voice operated devices are becoming popular nowadays. For this it is necessary that voice authentication is secure. In this paper, we address some known attacks like replay, personification and attacks using AI voice bots and limitations like text and language dependency of human voice authentication systems. We have also developed an interactive system to tackle these problems. The system verifies the user by performing voice matching as well as on an intellectual level by asking questions which only humans are able to answer and not any AI bot. In the system, an average user requires around 35 seconds for registration and around 25 seconds for authentication. The system's accuracy comes out to be 97.8% for English speakers and 95% for Hindi speakers.
{"title":"Voice Based Authentication Using Mel-Frequency Cepstral Coefficients and Gaussian Mixture Model","authors":"D. Pawade, Avani M. Sakhapara, Rujuta Ashtekar, Diya Bakhai, Shruti Tyagi","doi":"10.1109/IBSSC56953.2022.10037421","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037421","url":null,"abstract":"Voice operated devices are becoming popular nowadays. For this it is necessary that voice authentication is secure. In this paper, we address some known attacks like replay, personification and attacks using AI voice bots and limitations like text and language dependency of human voice authentication systems. We have also developed an interactive system to tackle these problems. The system verifies the user by performing voice matching as well as on an intellectual level by asking questions which only humans are able to answer and not any AI bot. In the system, an average user requires around 35 seconds for registration and around 25 seconds for authentication. The system's accuracy comes out to be 97.8% for English speakers and 95% for Hindi speakers.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133379312","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 : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037493
Shruti Jain, Jash Jain, Garima Merani, Jash Patel
Investments and the knowledge surrounding them have started to pick up the pace in India. As more and more people learn about it and experiment, opinions have begun to formulate around it. In this paper, we conducted a survey to get the idea of people's perspectives with regards to their knowledge on Investment and the different types of Investments people are making. The major purpose of this study is to analyze the survey responses and draw conclusions using data analysis techniques. We performed Sentiment Analysis on the survey participants to identify why people invest, why some do not invest, and what people think about NFT's and Crypto. With this paper, we hope to have a clearer idea on the Investment and Stocks pattern of Indians.
{"title":"Statistical and Sentiment Analysis On Investment Pattern of Indians","authors":"Shruti Jain, Jash Jain, Garima Merani, Jash Patel","doi":"10.1109/IBSSC56953.2022.10037493","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037493","url":null,"abstract":"Investments and the knowledge surrounding them have started to pick up the pace in India. As more and more people learn about it and experiment, opinions have begun to formulate around it. In this paper, we conducted a survey to get the idea of people's perspectives with regards to their knowledge on Investment and the different types of Investments people are making. The major purpose of this study is to analyze the survey responses and draw conclusions using data analysis techniques. We performed Sentiment Analysis on the survey participants to identify why people invest, why some do not invest, and what people think about NFT's and Crypto. With this paper, we hope to have a clearer idea on the Investment and Stocks pattern of Indians.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132305532","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 : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037392
Purnima Ahirao, Shubham Joshi
Due to rapid technology improvements, an increasing number of people are being connected to the digital world. Considering other parts of life, the internet has grown increasingly vital to Indians. More than 4.39 billion people use the internet, and almost 70% of them use social media on smartphones, tablets, laptops, and other computers. Management, staff, and users all play an important role in information security. As a result, the human has become the weakest point in the digital environment. Human understanding and behavior are essential for successful and efficient usage of technology. The human aspect can be divided into two categories: one in which humans are directly involved in the system in some way, and the other in which they are not. The end users' lack of understanding, belief, conduct, and inappropriate use of technology are the other specific variables. Users desire security, flexibility, and simplicity of use all at the same time. Finding a balance between all these criteria is extremely difficult for any business or service provider. Users must be willing to give all information, including personal and sensitive information, in order to be online. The future is expected to be data-centric and data-driven. Data, according to researchers, is the new fuel that will drive technology forward. As a result, service providers have become accustomed to organizing the data for computational and surveillance purposes. It is time for data to be protected and confidentiality to be respected. The most significant aspect of this issue is the user's concern in deciding whether to share the data. If exchanging data is required, should the customer be able to track who has accessed the data? After that, the user should have a say in who gets access to the information and who does not. So, these are the numerous issues that require research and the development of a robust solution that places data control in the hands of the user. One of the regulations that deals with data protection and user privacy is the GDPR. This regulation must be made mandatory in all countries, including India. Using blockchain technology, the authors discuss various approaches for protecting user privacy and controlling data access. The authors also attempt to propose a different strategy to resolving the privacy and security issues using blockchain technology in a modified and enhanced manner.
{"title":"Social media users privacy protection from social surveillance using Blockchain Technology","authors":"Purnima Ahirao, Shubham Joshi","doi":"10.1109/IBSSC56953.2022.10037392","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037392","url":null,"abstract":"Due to rapid technology improvements, an increasing number of people are being connected to the digital world. Considering other parts of life, the internet has grown increasingly vital to Indians. More than 4.39 billion people use the internet, and almost 70% of them use social media on smartphones, tablets, laptops, and other computers. Management, staff, and users all play an important role in information security. As a result, the human has become the weakest point in the digital environment. Human understanding and behavior are essential for successful and efficient usage of technology. The human aspect can be divided into two categories: one in which humans are directly involved in the system in some way, and the other in which they are not. The end users' lack of understanding, belief, conduct, and inappropriate use of technology are the other specific variables. Users desire security, flexibility, and simplicity of use all at the same time. Finding a balance between all these criteria is extremely difficult for any business or service provider. Users must be willing to give all information, including personal and sensitive information, in order to be online. The future is expected to be data-centric and data-driven. Data, according to researchers, is the new fuel that will drive technology forward. As a result, service providers have become accustomed to organizing the data for computational and surveillance purposes. It is time for data to be protected and confidentiality to be respected. The most significant aspect of this issue is the user's concern in deciding whether to share the data. If exchanging data is required, should the customer be able to track who has accessed the data? After that, the user should have a say in who gets access to the information and who does not. So, these are the numerous issues that require research and the development of a robust solution that places data control in the hands of the user. One of the regulations that deals with data protection and user privacy is the GDPR. This regulation must be made mandatory in all countries, including India. Using blockchain technology, the authors discuss various approaches for protecting user privacy and controlling data access. The authors also attempt to propose a different strategy to resolving the privacy and security issues using blockchain technology in a modified and enhanced manner.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130957483","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 : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037558
Manisha Joshi, Savita Bhosale, V. Vyawahare
Fractional calculus has been adopted in the modelling of many scientific processes and systems. Due to the inherent feature of long term memory of fractional derivatives, it has been used in the learning process of neural networks. A fractional order derivative based back propagation learning algorithm in neural networks is proposed in this paper. Specifically, Riemann-Liouville (R-L), Caputo (C) and Caputo Fabrizio (CF) fractional Derivative based on the back propagation algorithms in a three layer feed-forward neural network employed. To get a faster learning rate without oscillation, momentum factor is incorporated. The effect of fractional order and momentum factor is investigated and compared. The performance of these fractional derivatives based algorithms with integer derivatives based algorithm in terms of mean square error (MSE), particularly the salary based on years of experience is predicted. Results demonstrate that fractional derivative based learning algorithms outperform the integer derivatives.
{"title":"Using fractional derivative in learning algorithm for artificial neural network: Application for salary prediction","authors":"Manisha Joshi, Savita Bhosale, V. Vyawahare","doi":"10.1109/IBSSC56953.2022.10037558","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037558","url":null,"abstract":"Fractional calculus has been adopted in the modelling of many scientific processes and systems. Due to the inherent feature of long term memory of fractional derivatives, it has been used in the learning process of neural networks. A fractional order derivative based back propagation learning algorithm in neural networks is proposed in this paper. Specifically, Riemann-Liouville (R-L), Caputo (C) and Caputo Fabrizio (CF) fractional Derivative based on the back propagation algorithms in a three layer feed-forward neural network employed. To get a faster learning rate without oscillation, momentum factor is incorporated. The effect of fractional order and momentum factor is investigated and compared. The performance of these fractional derivatives based algorithms with integer derivatives based algorithm in terms of mean square error (MSE), particularly the salary based on years of experience is predicted. Results demonstrate that fractional derivative based learning algorithms outperform the integer derivatives.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133288583","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 : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037499
Anupama Jawale, Ganesh M. Magar
There exist many techniques for feature selection and reduction to reduce dimensions of the large sensor dataset. For real time data processing, compressed and prominent feature of highest significance is desirable for efficient way of resource optimization and computation cost reduction. The goal of this research study is to highlight most significant feature of the dataset and to generate compressed time series by highlighting it. The highlighted feature of accelerometer sensor dataset is extracted, and a more compressed form of time series is generated using statistical and clustering methods like k-means, Partition around Medoids (PAM), Max-Value, 95% Confidence Interval values and Ceil Function calculations. As a result, around 80 % reduction in dataset with the similar pattern as of original time series is achieved. The original time series is compared with generated output series using Dynamic Time Warping method, where, we have obtained normalized error distance of 0.02. (Accuracy 98%)
为了对大型传感器数据集进行降维，存在许多特征选择和降维技术。在实时数据处理中，最重要的压缩和突出特征是优化资源和降低计算成本的有效途径。本研究的目的是突出数据集的最重要特征，并通过突出数据集来生成压缩时间序列。提取加速度计传感器数据集的突出特征，并使用k-means, Partition around mediids (PAM)， Max-Value, 95%置信区间值和Ceil函数计算等统计和聚类方法生成更压缩的时间序列。结果表明，具有与原始时间序列相似模式的数据集的概率降低了80%左右。使用Dynamic time Warping方法将原始时间序列与生成的输出序列进行比较，得到归一化误差距离为0.02。(精度为98%)
{"title":"Highlighting Prominent Features for Size Reduction in Time Series Data using Clustering Techniques","authors":"Anupama Jawale, Ganesh M. Magar","doi":"10.1109/IBSSC56953.2022.10037499","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037499","url":null,"abstract":"There exist many techniques for feature selection and reduction to reduce dimensions of the large sensor dataset. For real time data processing, compressed and prominent feature of highest significance is desirable for efficient way of resource optimization and computation cost reduction. The goal of this research study is to highlight most significant feature of the dataset and to generate compressed time series by highlighting it. The highlighted feature of accelerometer sensor dataset is extracted, and a more compressed form of time series is generated using statistical and clustering methods like k-means, Partition around Medoids (PAM), Max-Value, 95% Confidence Interval values and Ceil Function calculations. As a result, around 80 % reduction in dataset with the similar pattern as of original time series is achieved. The original time series is compared with generated output series using Dynamic Time Warping method, where, we have obtained normalized error distance of 0.02. (Accuracy 98%)","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133497805","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 : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037536
B. Dora, S. Bhat, Sudip Halder, Ishan Srivastava
This paper proposes a hybrid metaheuristic algorithm to solve the decade Generation Expansion Planning (GEP)problem. In this proposed hybrid approach, the mutualism phase of Symbiotic Organism Search (SOS) is implemented in the Dwarf Mongoose Optimization Algorithm (DMOA) to improve the local search capability of the DMOA. In this hybrid algorithm, global search is taken care by the DMOA, and the local search is taken care by the mutualism phase SOS algorithm, which will help in solving nonlinear and nonconvex optimization problems. In recent decade every country aims to decarbonize its economy by implementing policies that increase the penetration of Renewable Energy Sources (RES) in its power generation capacity. This paper also presents a multidimensional framework of GEP based on the increasing penetration level of RES with the help of Enhanced Dwarf Mongoose Optimization Algorithm (EDMOA). The simulation results are discussed in the result section and compared with many previously published algorithms. The statistical study confirms the hybrid algorithm's effectiveness and resilience.
{"title":"A Solution to the Techno-Economic Generation Expansion Planning using Enhanced Dwarf Mongoose Optimization Algorithm","authors":"B. Dora, S. Bhat, Sudip Halder, Ishan Srivastava","doi":"10.1109/IBSSC56953.2022.10037536","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037536","url":null,"abstract":"This paper proposes a hybrid metaheuristic algorithm to solve the decade Generation Expansion Planning (GEP)problem. In this proposed hybrid approach, the mutualism phase of Symbiotic Organism Search (SOS) is implemented in the Dwarf Mongoose Optimization Algorithm (DMOA) to improve the local search capability of the DMOA. In this hybrid algorithm, global search is taken care by the DMOA, and the local search is taken care by the mutualism phase SOS algorithm, which will help in solving nonlinear and nonconvex optimization problems. In recent decade every country aims to decarbonize its economy by implementing policies that increase the penetration of Renewable Energy Sources (RES) in its power generation capacity. This paper also presents a multidimensional framework of GEP based on the increasing penetration level of RES with the help of Enhanced Dwarf Mongoose Optimization Algorithm (EDMOA). The simulation results are discussed in the result section and compared with many previously published algorithms. The statistical study confirms the hybrid algorithm's effectiveness and resilience.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"113 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114313425","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}
Many industries today have benefited from developing new technologies, particularly data science, machine learning, artificial intelligence, and deep learning. This includes agriculture. Previous research have shown that plant leaf diseases are losing productivity at an increasing pace, which accounts for 40-42% of agricultural production losses (Cost: 12.42 billion euros; Source: United Nations Food and Agriculture Organization (FAO)). This big issue may be resolved by employing this method for recognizing plant leaf disease from the input photographs. This technique involves steps including feature extraction, image segmentation, and image preprocessing. Next, a convolutional neural network-based classification approach is applied. The suggested implementation was 98.3% accurate in predicting plant leaf diseases.
{"title":"Plant Leaf Disease Detection And Classification Based On Machine Learning Model","authors":"Aashish Jha, Madhavi Purohit, Vivek Maurya, Amiyakumar Tripathy","doi":"10.1109/IBSSC56953.2022.10037470","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037470","url":null,"abstract":"Many industries today have benefited from developing new technologies, particularly data science, machine learning, artificial intelligence, and deep learning. This includes agriculture. Previous research have shown that plant leaf diseases are losing productivity at an increasing pace, which accounts for 40-42% of agricultural production losses (Cost: 12.42 billion euros; Source: United Nations Food and Agriculture Organization (FAO)). This big issue may be resolved by employing this method for recognizing plant leaf disease from the input photographs. This technique involves steps including feature extraction, image segmentation, and image preprocessing. Next, a convolutional neural network-based classification approach is applied. The suggested implementation was 98.3% accurate in predicting plant leaf diseases.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127942017","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 : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037449
Utkarsh Gupta, Varun Mk, G. Srinivasa
This study analyses road traffic accident data recorded over a period of time to gain insights to the underlying pain points in the infrastructure and policies. Such insight allows us to focus our efforts in the right direction to make the lives of people safer. The data includes various geographical and meteorological factors affecting the severity of these accidents. We use Kernel density estimation (KDE) plots to analyse hotspots of accident-prone areas weighed against severity over years to understand the evolution of these dangerous zones. Furthermore, we use machine learning algorithms to predict the accident severity given certain parameters and to understand the factors that have a major influence on the severity of the accident. We have studied a publicly available dataset of road traffic accidents in the UK as a proof of concept of the pipeline to understand the underlying patterns of accidents occurring in a region of interest.
{"title":"A Comprehensive Study of Road Traffic Accidents: Hotspot Analysis and Severity Prediction Using Machine Learning","authors":"Utkarsh Gupta, Varun Mk, G. Srinivasa","doi":"10.1109/IBSSC56953.2022.10037449","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037449","url":null,"abstract":"This study analyses road traffic accident data recorded over a period of time to gain insights to the underlying pain points in the infrastructure and policies. Such insight allows us to focus our efforts in the right direction to make the lives of people safer. The data includes various geographical and meteorological factors affecting the severity of these accidents. We use Kernel density estimation (KDE) plots to analyse hotspots of accident-prone areas weighed against severity over years to understand the evolution of these dangerous zones. Furthermore, we use machine learning algorithms to predict the accident severity given certain parameters and to understand the factors that have a major influence on the severity of the accident. We have studied a publicly available dataset of road traffic accidents in the UK as a proof of concept of the pipeline to understand the underlying patterns of accidents occurring in a region of interest.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"516 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132380002","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 : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037519
Yada Sai Pranay, Jagadeeshwar Tabjula, Srijith Kanakambaran
Distributed fiber optic sensors are smart replacements to point sensors in monitoring vibrations over long distances with excellent resolution. In this paper, we investigate the use of machine learning models to classify different vibrational events. Spectrograms of vibrational events available on a public database is used for training and testing the machine learning models like Support Vector Machine, Ensemble learning and K-Nearest Neighbour. The best accuracy of 86.1% is obtained for Support Vector classifier after hyperparameter tuning with 5-fold cross validation.
{"title":"Classification Studies on Vibrational Patterns of Distributed Fiber Sensors using Machine Learning","authors":"Yada Sai Pranay, Jagadeeshwar Tabjula, Srijith Kanakambaran","doi":"10.1109/IBSSC56953.2022.10037519","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037519","url":null,"abstract":"Distributed fiber optic sensors are smart replacements to point sensors in monitoring vibrations over long distances with excellent resolution. In this paper, we investigate the use of machine learning models to classify different vibrational events. Spectrograms of vibrational events available on a public database is used for training and testing the machine learning models like Support Vector Machine, Ensemble learning and K-Nearest Neighbour. The best accuracy of 86.1% is obtained for Support Vector classifier after hyperparameter tuning with 5-fold cross validation.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132475731","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：