Pub Date : 2022-08-31DOI: 10.1109/csi54720.2022.9924137
Swati Goel
A novel procedure to compute service criticality level in a safety and business critical systems is presented. The presented “Service Composability Count (SCC)” procedure can be used to schedule the outage of services, according to their criticality level. This SCC plays an important role in modelling planned outage, as continuous availability of some critical services is highly desirable in service-based systems (SBS). This paper focuses on accurately scheduling the outage on the basis of service criticality. Results indicate that highly composable services are critical, whereas atomic services are very less critical. That means, while scheduling the outage of services for maintenance purposes -Atomic services should be scheduled first while composable services should be scheduled at last.
{"title":"Proposed SCC Metric: Criticality Measure in modeling scheduled outage","authors":"Swati Goel","doi":"10.1109/csi54720.2022.9924137","DOIUrl":"https://doi.org/10.1109/csi54720.2022.9924137","url":null,"abstract":"A novel procedure to compute service criticality level in a safety and business critical systems is presented. The presented “Service Composability Count (SCC)” procedure can be used to schedule the outage of services, according to their criticality level. This SCC plays an important role in modelling planned outage, as continuous availability of some critical services is highly desirable in service-based systems (SBS). This paper focuses on accurately scheduling the outage on the basis of service criticality. Results indicate that highly composable services are critical, whereas atomic services are very less critical. That means, while scheduling the outage of services for maintenance purposes -Atomic services should be scheduled first while composable services should be scheduled at last.","PeriodicalId":221137,"journal":{"name":"2022 International Conference on Connected Systems & Intelligence (CSI)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115867294","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 : 2022-08-31DOI: 10.1109/CSI54720.2022.9923954
Nikhitha Mani, Sandhya Harikumar
With expedition of huge number of research articles published in each domain, retrieval of relevant articles based on researcher's interest and requirement have become challenging. Further, there are circumstances where a researcher may not get the specific information sought after, regardless of whether it is there in the document collection. This is since the article in database is structured only based on title and citation data while the conceptual information and algorithms described inside an article are disregarded. In this work, we propose a useful methodology for re-structuring documents in the database by considering the document as a whole and representing the keywords and key-phrases extracted from the article data using Knowledge Graphs. Clustering of nodes of graph segregate articles into different domains and sub-domains. Knowledge graph is further explored to identify most important documents using modularity and then the documents are sorted based on relevance. Thus, proper structuring of the documents helps the researchers to recognize applicable content from a large database in short span of time since database associated with the query system is improvised. This technique is beneficial to all the researchers who are trying to resolve a problem by identifying apt documents for information need.
{"title":"A Knowledge Graph Approach towards Re-structuring of Scientific Articles","authors":"Nikhitha Mani, Sandhya Harikumar","doi":"10.1109/CSI54720.2022.9923954","DOIUrl":"https://doi.org/10.1109/CSI54720.2022.9923954","url":null,"abstract":"With expedition of huge number of research articles published in each domain, retrieval of relevant articles based on researcher's interest and requirement have become challenging. Further, there are circumstances where a researcher may not get the specific information sought after, regardless of whether it is there in the document collection. This is since the article in database is structured only based on title and citation data while the conceptual information and algorithms described inside an article are disregarded. In this work, we propose a useful methodology for re-structuring documents in the database by considering the document as a whole and representing the keywords and key-phrases extracted from the article data using Knowledge Graphs. Clustering of nodes of graph segregate articles into different domains and sub-domains. Knowledge graph is further explored to identify most important documents using modularity and then the documents are sorted based on relevance. Thus, proper structuring of the documents helps the researchers to recognize applicable content from a large database in short span of time since database associated with the query system is improvised. This technique is beneficial to all the researchers who are trying to resolve a problem by identifying apt documents for information need.","PeriodicalId":221137,"journal":{"name":"2022 International Conference on Connected Systems & Intelligence (CSI)","volume":"185 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123108244","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 : 2022-08-31DOI: 10.1109/CSI54720.2022.9923984
Ranjith J., V. M
This paper provides a high-level review and evaluation of the present situation regarding IoT Securities. An Internet of Things design seeks to connect everyone with anything they want, whenever they want it. The perception layer, network layer, the application layer is the most important three layers which mainly make the Internet of Things. To achieve a stable IoT reality, a range of safety precautions must be imposed at each tier. The only way to assure the future of the IoT infrastructure is to address and resolve the security issues that it has. Many researchers have attempted to address the security concerns specific to IoT layers and devices by applying suitable measures. This paper provides a top level view of safety ideas, technology and security problems, possible remedies, and the IoT's future directions for securing and also in this paper, a depth evaluation of the safety associated demanding situations and raw sources of danger with inside the IoT programs have been showcased. Latter discussion on the safety issue, diverse rising and present technology centered for attaining an excessive diploma of agree with inside the IoT programs are discussed.
{"title":"Security Challenges Prospective Measures In The Current Status of Internet of Things (IoT)","authors":"Ranjith J., V. M","doi":"10.1109/CSI54720.2022.9923984","DOIUrl":"https://doi.org/10.1109/CSI54720.2022.9923984","url":null,"abstract":"This paper provides a high-level review and evaluation of the present situation regarding IoT Securities. An Internet of Things design seeks to connect everyone with anything they want, whenever they want it. The perception layer, network layer, the application layer is the most important three layers which mainly make the Internet of Things. To achieve a stable IoT reality, a range of safety precautions must be imposed at each tier. The only way to assure the future of the IoT infrastructure is to address and resolve the security issues that it has. Many researchers have attempted to address the security concerns specific to IoT layers and devices by applying suitable measures. This paper provides a top level view of safety ideas, technology and security problems, possible remedies, and the IoT's future directions for securing and also in this paper, a depth evaluation of the safety associated demanding situations and raw sources of danger with inside the IoT programs have been showcased. Latter discussion on the safety issue, diverse rising and present technology centered for attaining an excessive diploma of agree with inside the IoT programs are discussed.","PeriodicalId":221137,"journal":{"name":"2022 International Conference on Connected Systems & Intelligence (CSI)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126796388","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 : 2022-08-31DOI: 10.1109/CSI54720.2022.9924085
Jayashree Nair, Akhila Krishnan, Vrinda S
Speech and spoken words have always played a key role in everyday life. Speech synthesis is a means of artificially synthesizing speech, whereas text-to-speech (TTS) is a technology that converts written text in a human language into an analogous spoken waveform [speech form].The written form is represented by the text, a sequence of characters, whereas the verbal form is represented by the speech. TTS synthesizers are computer-based systems that read text out loud. The TTS system is divided into two phases: text processing and speech creation. Despite the availability of several TTS systems in various languages, Indian languages continue to lag behind in terms of producing high-quality speech. Acceptability and intelligibility are used to rate the quality of speech. The main objective of this paper is to perform a study on available text-to-speech technologies in Indian languages.
{"title":"Indian Text to Speech Systems: A Short Survey","authors":"Jayashree Nair, Akhila Krishnan, Vrinda S","doi":"10.1109/CSI54720.2022.9924085","DOIUrl":"https://doi.org/10.1109/CSI54720.2022.9924085","url":null,"abstract":"Speech and spoken words have always played a key role in everyday life. Speech synthesis is a means of artificially synthesizing speech, whereas text-to-speech (TTS) is a technology that converts written text in a human language into an analogous spoken waveform [speech form].The written form is represented by the text, a sequence of characters, whereas the verbal form is represented by the speech. TTS synthesizers are computer-based systems that read text out loud. The TTS system is divided into two phases: text processing and speech creation. Despite the availability of several TTS systems in various languages, Indian languages continue to lag behind in terms of producing high-quality speech. Acceptability and intelligibility are used to rate the quality of speech. The main objective of this paper is to perform a study on available text-to-speech technologies in Indian languages.","PeriodicalId":221137,"journal":{"name":"2022 International Conference on Connected Systems & Intelligence (CSI)","volume":"2009 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127323823","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 : 2022-08-31DOI: 10.1109/CSI54720.2022.9923957
Mohd. Belal, Ghufran Ullah, Abdullah Ahmad Khan
Segregating Islamophobic hate speech from other instances of offensive language is a serious hurdle for automatic hate-speech detection on social media platforms such as Twitter. Because lexical detection methods classify all messages containing particular terms like hate speech, previous work using supervised learning has failed to differentiate between these categories. This task is complex due to the level of difficulty in natural language constructs. We have worked on a transfer learning approach using Universal Language Model Fine-tuning (ULMFIT), an efficient method that can be applied to classification tasks. Our method gave more than 80 percent accuracy and the confusion matrix thus formed was successfully able to classify those datasets proportionally into each block. The use of Deep learning in text classification has been underutilized. This method will contribute to solving the spread of Islamophobia which hasn't been taken into consideration when taking action against online hate
{"title":"Islamophobic Tweet Detection using Transfer Learning","authors":"Mohd. Belal, Ghufran Ullah, Abdullah Ahmad Khan","doi":"10.1109/CSI54720.2022.9923957","DOIUrl":"https://doi.org/10.1109/CSI54720.2022.9923957","url":null,"abstract":"Segregating Islamophobic hate speech from other instances of offensive language is a serious hurdle for automatic hate-speech detection on social media platforms such as Twitter. Because lexical detection methods classify all messages containing particular terms like hate speech, previous work using supervised learning has failed to differentiate between these categories. This task is complex due to the level of difficulty in natural language constructs. We have worked on a transfer learning approach using Universal Language Model Fine-tuning (ULMFIT), an efficient method that can be applied to classification tasks. Our method gave more than 80 percent accuracy and the confusion matrix thus formed was successfully able to classify those datasets proportionally into each block. The use of Deep learning in text classification has been underutilized. This method will contribute to solving the spread of Islamophobia which hasn't been taken into consideration when taking action against online hate","PeriodicalId":221137,"journal":{"name":"2022 International Conference on Connected Systems & Intelligence (CSI)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129226169","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 : 2022-08-31DOI: 10.1109/CSI54720.2022.9924023
Abhishek Jain, Neena Goveas
Internet of Things (IoT) based solutions requiring real time results from intensive computation tasks or having large scale data analysis have traditionally been designed with offloading of the work to cloud infrastructure. This has been found to be not an ideal solution due to several issues related to network uncertainties, cost of cloud usage etc. This is especially true for systems with both hard time constraints and large amount of data. Edge computing, with its hierarchical configuration has been proposed to solve these issues. This has led to researchers proposing several algorithms to optimise offloading of computation to the layers of this hierarchy. In this work we propose the use of an actor-critic based reinforcement learning mechanism to solve the offloading planning for a general hierarchical system with multiple end nodes and multiple edge servers. Our simulation based results shows that the proposed method improves the performance of the system as compared to the existing benchmark offloading policies.
{"title":"Enhanced edge offloading using Reinforcement learning","authors":"Abhishek Jain, Neena Goveas","doi":"10.1109/CSI54720.2022.9924023","DOIUrl":"https://doi.org/10.1109/CSI54720.2022.9924023","url":null,"abstract":"Internet of Things (IoT) based solutions requiring real time results from intensive computation tasks or having large scale data analysis have traditionally been designed with offloading of the work to cloud infrastructure. This has been found to be not an ideal solution due to several issues related to network uncertainties, cost of cloud usage etc. This is especially true for systems with both hard time constraints and large amount of data. Edge computing, with its hierarchical configuration has been proposed to solve these issues. This has led to researchers proposing several algorithms to optimise offloading of computation to the layers of this hierarchy. In this work we propose the use of an actor-critic based reinforcement learning mechanism to solve the offloading planning for a general hierarchical system with multiple end nodes and multiple edge servers. Our simulation based results shows that the proposed method improves the performance of the system as compared to the existing benchmark offloading policies.","PeriodicalId":221137,"journal":{"name":"2022 International Conference on Connected Systems & Intelligence (CSI)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132172972","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 : 2022-08-31DOI: 10.1109/CSI54720.2022.9923983
Aashish, Aditya Thakkar, Shubham Yadav, Sandeep Saini, K. Lata
Indian agriculture is quite diverse and plays a vital role in the country's economic growth. The tools and techniques used in agriculture are no more primitive. With the gradual evolution of the population, this sector is under severe pressure to produce at high efficiency. One of the significant factors in improving crop harvest is the timely detection of crop diseases. The farmers use scouting to monitor their crops, which requires extensive labor and is time-consuming. Image processing-based disease identification makes the process faster and more accurate. Recently, deep learning techniques have been deployed for automatic plant disease identification. Researchers have used Convolutional Neural Networks (CNN) to predict the type of diseases in different crops accurately. Considering the advantages of autoencoders and CNN, we have proposed and developed a hybrid deep learning model based on CNN and Autoencoders to detect multiple plant diseases. The proposed architecture is fine-tuned to detect diseases of numerous crops. The proposed model provides higher accuracy when compared with similar systems. We have tested our model using the Plant village dataset containing almost 15 different types of crops.
{"title":"CNN and Autoencoders based Hybrid Deep Learning Model for Crop Disease Detection","authors":"Aashish, Aditya Thakkar, Shubham Yadav, Sandeep Saini, K. Lata","doi":"10.1109/CSI54720.2022.9923983","DOIUrl":"https://doi.org/10.1109/CSI54720.2022.9923983","url":null,"abstract":"Indian agriculture is quite diverse and plays a vital role in the country's economic growth. The tools and techniques used in agriculture are no more primitive. With the gradual evolution of the population, this sector is under severe pressure to produce at high efficiency. One of the significant factors in improving crop harvest is the timely detection of crop diseases. The farmers use scouting to monitor their crops, which requires extensive labor and is time-consuming. Image processing-based disease identification makes the process faster and more accurate. Recently, deep learning techniques have been deployed for automatic plant disease identification. Researchers have used Convolutional Neural Networks (CNN) to predict the type of diseases in different crops accurately. Considering the advantages of autoencoders and CNN, we have proposed and developed a hybrid deep learning model based on CNN and Autoencoders to detect multiple plant diseases. The proposed architecture is fine-tuned to detect diseases of numerous crops. The proposed model provides higher accuracy when compared with similar systems. We have tested our model using the Plant village dataset containing almost 15 different types of crops.","PeriodicalId":221137,"journal":{"name":"2022 International Conference on Connected Systems & Intelligence (CSI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128652138","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 : 2022-08-31DOI: 10.1109/CSI54720.2022.9923985
A. Kurup, G. Sajeev, Swaminathan J
Crowdsourcing is an information system that provides a cost-effective way of solving computationally challenging problems. However, it is potentially vulnerable to adversarial attacks as the service provider cannot manage workers' behavior. Malicious workers provide unreliable answers to manipulate the system. These attacks affect the truth inference process and thus leads to wrong answers for a targeted set of tasks. Eventually, this reduces the accuracy of aggregated results. Existing works have proposed various types of attacks in crowdsourcing systems and indicate that truth inference is the most affected one. So, we propose methods for defending these attacks for improving the truth inference process. We empirically evaluate the proposed truth inference method on a real and synthetic dataset. The performance of the proposed method is verified, and the results show that it is robust to adversarial attacks with comparable accuracy.
{"title":"Truth Inference in Crowdsourcing Under Adversarial Attacks","authors":"A. Kurup, G. Sajeev, Swaminathan J","doi":"10.1109/CSI54720.2022.9923985","DOIUrl":"https://doi.org/10.1109/CSI54720.2022.9923985","url":null,"abstract":"Crowdsourcing is an information system that provides a cost-effective way of solving computationally challenging problems. However, it is potentially vulnerable to adversarial attacks as the service provider cannot manage workers' behavior. Malicious workers provide unreliable answers to manipulate the system. These attacks affect the truth inference process and thus leads to wrong answers for a targeted set of tasks. Eventually, this reduces the accuracy of aggregated results. Existing works have proposed various types of attacks in crowdsourcing systems and indicate that truth inference is the most affected one. So, we propose methods for defending these attacks for improving the truth inference process. We empirically evaluate the proposed truth inference method on a real and synthetic dataset. The performance of the proposed method is verified, and the results show that it is robust to adversarial attacks with comparable accuracy.","PeriodicalId":221137,"journal":{"name":"2022 International Conference on Connected Systems & Intelligence (CSI)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121838438","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 : 2022-08-31DOI: 10.1109/CSI54720.2022.9923950
Shweta Rajput, Resham Chawra, Palash Shirish Wani, S. Nanda
Due to the low energy attenuation of an acoustic wave in water, the side-scan sonar imaging technique is popularly used for underwater exploration. The images collected in this process contain a high amount of noise, which poses a challenge to accurately detecting underwater objects. In this paper, the de-noising of such images is carried out through a non-local means filtering algorithm. The obtained denoised images are further segmented to effectively determine the object, shadow, and background. The segmentation task is formulated as a clustering problem, and a recently reported nature-inspired algorithm known as Reptile Search Algorithm (RSA) is used. The RSA is based on the hunting behavior of crocodiles in a specific region. The Davies-Bouldin index is used as the fitness function to perform the clustering. The performance of the proposed method is evaluated on four plane and four-ship images collected from the benchmark KLSG-II dataset. The obtained results are compared with the image segmentation performed by particle swarm optimization and genetic algorithm. Comparative results reveal that the proposed RSA-based model obtained better results in de-noising and effectively segmenting the eight images.
{"title":"Noisy Sonar Image Segmentation using Reptile Search Algorithm","authors":"Shweta Rajput, Resham Chawra, Palash Shirish Wani, S. Nanda","doi":"10.1109/CSI54720.2022.9923950","DOIUrl":"https://doi.org/10.1109/CSI54720.2022.9923950","url":null,"abstract":"Due to the low energy attenuation of an acoustic wave in water, the side-scan sonar imaging technique is popularly used for underwater exploration. The images collected in this process contain a high amount of noise, which poses a challenge to accurately detecting underwater objects. In this paper, the de-noising of such images is carried out through a non-local means filtering algorithm. The obtained denoised images are further segmented to effectively determine the object, shadow, and background. The segmentation task is formulated as a clustering problem, and a recently reported nature-inspired algorithm known as Reptile Search Algorithm (RSA) is used. The RSA is based on the hunting behavior of crocodiles in a specific region. The Davies-Bouldin index is used as the fitness function to perform the clustering. The performance of the proposed method is evaluated on four plane and four-ship images collected from the benchmark KLSG-II dataset. The obtained results are compared with the image segmentation performed by particle swarm optimization and genetic algorithm. Comparative results reveal that the proposed RSA-based model obtained better results in de-noising and effectively segmenting the eight images.","PeriodicalId":221137,"journal":{"name":"2022 International Conference on Connected Systems & Intelligence (CSI)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121559961","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 : 2022-08-31DOI: 10.1109/CSI54720.2022.9924139
Bembamba Fulbert, O. T. Frédéric, Malo Sadouanouan, Yougbare Bernadette, O. Dominique
Uncontrolled crossbreeding between zebus and taurine cattle is jeopardizing the genetic heritage of West African taurines and their specific ability to resist trypanosomosis. to achieve any successful conservation policy for this species, it is crucial to accurately identify purebred taurines. Techniques in use today include empirical method and biological analysis. We offer in this paper a supervised Machine Learning approach of pure-bred taurine recognition. Five algorithms were trained using morphological data from hundreds of cows. Each of the models produced promising results. The RBF non linear SVM performs the best with up to 87% accuracy and 0.9308 of AUC. Furthermore, the correlation coefficients allowed to define the most discriminating morphological trait.
{"title":"An intelligent system for taurine breed recognition: preliminary results","authors":"Bembamba Fulbert, O. T. Frédéric, Malo Sadouanouan, Yougbare Bernadette, O. Dominique","doi":"10.1109/CSI54720.2022.9924139","DOIUrl":"https://doi.org/10.1109/CSI54720.2022.9924139","url":null,"abstract":"Uncontrolled crossbreeding between zebus and taurine cattle is jeopardizing the genetic heritage of West African taurines and their specific ability to resist trypanosomosis. to achieve any successful conservation policy for this species, it is crucial to accurately identify purebred taurines. Techniques in use today include empirical method and biological analysis. We offer in this paper a supervised Machine Learning approach of pure-bred taurine recognition. Five algorithms were trained using morphological data from hundreds of cows. Each of the models produced promising results. The RBF non linear SVM performs the best with up to 87% accuracy and 0.9308 of AUC. Furthermore, the correlation coefficients allowed to define the most discriminating morphological trait.","PeriodicalId":221137,"journal":{"name":"2022 International Conference on Connected Systems & Intelligence (CSI)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116072259","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：