Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037360
Shankey Garg, Pradeep Singh
Colorectal cancer is the most common type of cancer after breast cancer in women and third in men after lungs and prostrate cancer. The disease rank third in incidence and second in terms of mortality, hence early diagnosis is necessary for the correct line of treatment. Knowledge distillation based models boost the performance of small neural network and are performing efficiently for various image classification based tasks. In this work, a novel knowledge distillation based technique is developed to efficiently classify colorectal cancer histology images. Unlike traditional distillation, out method performs distillation in parts. Instead of supervising the student with a converged knowledge of teacher, the proposed method is fetching the teacher's knowledge at regular intervals and providing these knowledge to the student model during student training process. Through this multi-part distillation technique student can effectively learn the intermediate representational knowledge rather than the abstract knowledge of the teacher and hence boost the overall performance of the model. The the proposed model has achived 92.10% accuracy.
{"title":"Multi-Part Knowledge Distillation for the Efficient Classification of Colorectal Cancer Histology Images","authors":"Shankey Garg, Pradeep Singh","doi":"10.1109/IBSSC56953.2022.10037360","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037360","url":null,"abstract":"Colorectal cancer is the most common type of cancer after breast cancer in women and third in men after lungs and prostrate cancer. The disease rank third in incidence and second in terms of mortality, hence early diagnosis is necessary for the correct line of treatment. Knowledge distillation based models boost the performance of small neural network and are performing efficiently for various image classification based tasks. In this work, a novel knowledge distillation based technique is developed to efficiently classify colorectal cancer histology images. Unlike traditional distillation, out method performs distillation in parts. Instead of supervising the student with a converged knowledge of teacher, the proposed method is fetching the teacher's knowledge at regular intervals and providing these knowledge to the student model during student training process. Through this multi-part distillation technique student can effectively learn the intermediate representational knowledge rather than the abstract knowledge of the teacher and hence boost the overall performance of the model. The the proposed model has achived 92.10% accuracy.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"6 4 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":"126910041","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/IBSSC56953.2022.10037261
Shraddha Gore, Shital Patil, Vivek Khalane
A change in agricultural practices is necessary to prevent future food shortages caused by global overpopulation. With the Internet of Things (IoT) and low-power and low-cost devices, the agriculture industry can automate irrigation systems to efficiently use water resources by monitoring farm fields. Low Power Wide Area Networks (LPWAN), along with IoT, can solve bandwidth, coverage and power problems which are the main drawbacks of other wireless communication technologies. Long Range Wide Area Network (LoRaWAN) protocol is known as LoRa in LPWAN space. This protocol provides additional benefits like security, scalability, and robustness. In this paper, a smart agriculture model is proposed to assist in farmers' decision-making and help them to get more productive results. The result of this paper is a prototype equipment for measuring humidity and soil moisture content done by combining the data obtained from the sensors via a LoRaWAN network. This model sends sensor Data such as temperature (degree Celsius), soil moisture (percentage), and humidity (percentage) from the transmitter node to the receiver node using the LoRa communication method. The readings from these nodes are transmitted and then forwarded to the network server through a single gateway. The Wi-Fi-enabled receiving node track data daily on the ThingSpeak platform. The primary goal of this paper is to help farmers monitor their farms more effectively.
{"title":"Intelligent Farm Monitoring System using LoRa Enabled IoT","authors":"Shraddha Gore, Shital Patil, Vivek Khalane","doi":"10.1109/IBSSC56953.2022.10037261","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037261","url":null,"abstract":"A change in agricultural practices is necessary to prevent future food shortages caused by global overpopulation. With the Internet of Things (IoT) and low-power and low-cost devices, the agriculture industry can automate irrigation systems to efficiently use water resources by monitoring farm fields. Low Power Wide Area Networks (LPWAN), along with IoT, can solve bandwidth, coverage and power problems which are the main drawbacks of other wireless communication technologies. Long Range Wide Area Network (LoRaWAN) protocol is known as LoRa in LPWAN space. This protocol provides additional benefits like security, scalability, and robustness. In this paper, a smart agriculture model is proposed to assist in farmers' decision-making and help them to get more productive results. The result of this paper is a prototype equipment for measuring humidity and soil moisture content done by combining the data obtained from the sensors via a LoRaWAN network. This model sends sensor Data such as temperature (degree Celsius), soil moisture (percentage), and humidity (percentage) from the transmitter node to the receiver node using the LoRa communication method. The readings from these nodes are transmitted and then forwarded to the network server through a single gateway. The Wi-Fi-enabled receiving node track data daily on the ThingSpeak platform. The primary goal of this paper is to help farmers monitor their farms more effectively.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"26 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":"131697713","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}
Early detection of brain tumors is very crucial as they grow extremely fast. To extend patients' life expectancy, correct treatment planning and precise diagnoses are critical. Manual diagnosis can be prone to errors and is a time-consuming and complex task for radiologists because of how minute variations in the tumor could lead to a completely different diagnosis. The proposed method is focused on creating an automated way of classifying brain MRI images by using SOTA models like VGG-16 and InceptionV3 and building on them. The brain MRI images are classified into four classes by extracting significant features and experimented with and without pre-processing. The experimental results have shown that the VGG-16 model used, although without any image augmentation, has given a high validation accuracy of 74%. The inceptionV3 model without image augmentation techniques reported a worse validation accuracy of 69%, defining VGG-16 to be the better classifier.
{"title":"Classification Of Brain Images For Identification Of Tumors","authors":"Jayashree Shetty, Manjula K Shenoy, Vedant Rishi Das, Mahek Mishra, Rohan Prasad, Sarthak Seth","doi":"10.1109/IBSSC56953.2022.10037548","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037548","url":null,"abstract":"Early detection of brain tumors is very crucial as they grow extremely fast. To extend patients' life expectancy, correct treatment planning and precise diagnoses are critical. Manual diagnosis can be prone to errors and is a time-consuming and complex task for radiologists because of how minute variations in the tumor could lead to a completely different diagnosis. The proposed method is focused on creating an automated way of classifying brain MRI images by using SOTA models like VGG-16 and InceptionV3 and building on them. The brain MRI images are classified into four classes by extracting significant features and experimented with and without pre-processing. The experimental results have shown that the VGG-16 model used, although without any image augmentation, has given a high validation accuracy of 74%. The inceptionV3 model without image augmentation techniques reported a worse validation accuracy of 69%, defining VGG-16 to be the better classifier.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"1 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":"133617335","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}
Melanoma, a type of skin cancer, occurs when melanocytes become cancerous and is a common cause of death in adults. The presence of melanoma can be conclusively proved through biopsies, but these lap reports often take time. Early detection of melanoma could improve mortality rates and reduce costs. AI-based assistive tools can aid early detection. Most studies focus on detection either in dermoscopic images or in non-dermoscopic images, not both. In this paper, we propose a novel generalised framework which can detect melanoma in both dermoscopic and non-dermoscopic images. The framework includes a preprocessing pipeline, data augmentation and resolving class imbalances, followed by a VGG-16 model. The model gives a sensitivity (for melanoma cases) of 87% on non-dermoscopic images and 91 % on dermoscopic images.
{"title":"Robust deep learning framework for the detection of melanoma in images","authors":"Trisha Sarkar, Anushka Khare, Mohit Parekh, Param Mehta, Avani Bhuva","doi":"10.1109/IBSSC56953.2022.10037456","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037456","url":null,"abstract":"Melanoma, a type of skin cancer, occurs when melanocytes become cancerous and is a common cause of death in adults. The presence of melanoma can be conclusively proved through biopsies, but these lap reports often take time. Early detection of melanoma could improve mortality rates and reduce costs. AI-based assistive tools can aid early detection. Most studies focus on detection either in dermoscopic images or in non-dermoscopic images, not both. In this paper, we propose a novel generalised framework which can detect melanoma in both dermoscopic and non-dermoscopic images. The framework includes a preprocessing pipeline, data augmentation and resolving class imbalances, followed by a VGG-16 model. The model gives a sensitivity (for melanoma cases) of 87% on non-dermoscopic images and 91 % on dermoscopic images.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"33 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":"132910079","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/IBSSC56953.2022.10037317
Devarsh Patel, Nicole D'Souza, Riddhi Gawande
Information generation and its dissemination increases day by day on a very large scale as the count of users increase on social media. These platforms are a stage for the people to exchange their ideas and opinions. Social media microblogging platform (ex. Twitter) is the go-to place in case of discussion about any important event. Information spreads at a lightning pace on twitter. This leads to rapid spread of false information i.e. rumours which can cause a feeling of unrest among the people. Hence, it is crucial to analyze and verify the degree of truthfulness of such content. The automatic detection of rumours in its initial stages is a challenge because of the complexity of the text. In this paper, we have implemented and compared different existing machine learning algorithms on the PHEME dataset to identify and detect the rumours. The performance of the models has been analyzed.
{"title":"Automatic Twitter Rumour Detection using Machine Learning","authors":"Devarsh Patel, Nicole D'Souza, Riddhi Gawande","doi":"10.1109/IBSSC56953.2022.10037317","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037317","url":null,"abstract":"Information generation and its dissemination increases day by day on a very large scale as the count of users increase on social media. These platforms are a stage for the people to exchange their ideas and opinions. Social media microblogging platform (ex. Twitter) is the go-to place in case of discussion about any important event. Information spreads at a lightning pace on twitter. This leads to rapid spread of false information i.e. rumours which can cause a feeling of unrest among the people. Hence, it is crucial to analyze and verify the degree of truthfulness of such content. The automatic detection of rumours in its initial stages is a challenge because of the complexity of the text. In this paper, we have implemented and compared different existing machine learning algorithms on the PHEME dataset to identify and detect the rumours. The performance of the models has been analyzed.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"7 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":"133151052","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/IBSSC56953.2022.10037549
Sharisha Shanbhog M, Jeevan M
A sound mental health has its benefits for the overall well-being of an individual. The decline in mental health conditions has a critical impression on other vital functionalities of the human system both psychologically and physiologically. And a student's well-being is largely contributed by the level of perceived stress and overall quality of nighttime sleep which might have evolved by various external factors over a while. The main objective of this study is to understand the correlation between Perceived Stress Scale (PSS) scores and Pittsburgh Sleep Quality Index (PSQI) global scores from StudentLife, a publicly available dataset over the period, and classify the well-being factor as ‘Good’ ‘Average’ and ‘Bad’ The linear regression model significantly demonstrated the association between PSS scores and Pittsburgh Sleep Quality Index (PSQI) scores. Machine Learning techniques like Decision Trees (DT), Support Vector Machine (SVM), and K-nearest neighbors(K-NN) were implemented on both Pre-Test and Post-test questionnaire data. While SVM resulted in better accuracy for Pre-test data, the K-NN classifier resulted in best accuracy for Post-test data, and the performance was evaluated using performance metrics like accuracy Precision, recall, and F1 score.
{"title":"Prediction of Student's Wellbeing from Stress and Sleep Questionnaire data using Machine Learning Approach","authors":"Sharisha Shanbhog M, Jeevan M","doi":"10.1109/IBSSC56953.2022.10037549","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037549","url":null,"abstract":"A sound mental health has its benefits for the overall well-being of an individual. The decline in mental health conditions has a critical impression on other vital functionalities of the human system both psychologically and physiologically. And a student's well-being is largely contributed by the level of perceived stress and overall quality of nighttime sleep which might have evolved by various external factors over a while. The main objective of this study is to understand the correlation between Perceived Stress Scale (PSS) scores and Pittsburgh Sleep Quality Index (PSQI) global scores from StudentLife, a publicly available dataset over the period, and classify the well-being factor as ‘Good’ ‘Average’ and ‘Bad’ The linear regression model significantly demonstrated the association between PSS scores and Pittsburgh Sleep Quality Index (PSQI) scores. Machine Learning techniques like Decision Trees (DT), Support Vector Machine (SVM), and K-nearest neighbors(K-NN) were implemented on both Pre-Test and Post-test questionnaire data. While SVM resulted in better accuracy for Pre-test data, the K-NN classifier resulted in best accuracy for Post-test data, and the performance was evaluated using performance metrics like accuracy Precision, recall, and F1 score.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"17 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":"133402348","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}
Every company in today's time intends to maximize their profits. As we are in a fast-growing world, the need for products has increased drastically. Price of products depends on majorly two things, the cost of production (including transportation costs) and profit margins. Transportation problems are used to either minimize transportation cost or to maximize profits on shipping commodities. In this paper, we focus on how to minimize transportation cost while fulfilling the supply and demand conditions. The problem taken by us contain 3 sources and 3 destinations based in Nigeria and the Cost matrix was converted from Nigerian currency (Naira) to Dollars (conversion done according to rates on 15th October, 2022) for ease of calculation. Using Vogel's Approximation Method (VAM), Least Cost Method (LCM) and the North-West Corner Method (NWCM), we found that VAM was the most optimal. Self-written Python codes were used to verify the manual solutions. The unavailable or forbidden routes have also been considered in the code. The output displayed the allocations for the 3x3 matrix, and printed the total cost. After this, MS-Excel and Excel-QM software were used for verification. We found that VAM is 25.11 percent better than the LCM and 52.65 percent better than the NWCM for this problem. In every enterprise, generating higher revenue remains one of the most essential objectives. If codes for such methods are made universally available, enterprises would benefit highly. Use of transportation problems for optimal solutions have great potential, if one has knowledge about them.
{"title":"Case Study On Transport Of Petroleum In Nigerian Cities","authors":"Snehee Chheda, Anisha Gharat, Kruttika Abhyankar, Sheetal Gonsalves","doi":"10.1109/IBSSC56953.2022.10037348","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037348","url":null,"abstract":"Every company in today's time intends to maximize their profits. As we are in a fast-growing world, the need for products has increased drastically. Price of products depends on majorly two things, the cost of production (including transportation costs) and profit margins. Transportation problems are used to either minimize transportation cost or to maximize profits on shipping commodities. In this paper, we focus on how to minimize transportation cost while fulfilling the supply and demand conditions. The problem taken by us contain 3 sources and 3 destinations based in Nigeria and the Cost matrix was converted from Nigerian currency (Naira) to Dollars (conversion done according to rates on 15th October, 2022) for ease of calculation. Using Vogel's Approximation Method (VAM), Least Cost Method (LCM) and the North-West Corner Method (NWCM), we found that VAM was the most optimal. Self-written Python codes were used to verify the manual solutions. The unavailable or forbidden routes have also been considered in the code. The output displayed the allocations for the 3x3 matrix, and printed the total cost. After this, MS-Excel and Excel-QM software were used for verification. We found that VAM is 25.11 percent better than the LCM and 52.65 percent better than the NWCM for this problem. In every enterprise, generating higher revenue remains one of the most essential objectives. If codes for such methods are made universally available, enterprises would benefit highly. Use of transportation problems for optimal solutions have great potential, if one has knowledge about them.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"635 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":"115111320","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/IBSSC56953.2022.10037260
Akash Budhrani, Upasna Singh, Bhupendra Singh
The Operating System creates numerous objects to improve its efficiency and user experience and such objects are called artifacts. These artifacts record crucial data about the user activity. Such artifacts are the start point of any investigation as they can be an additional lead to a forensic triage. Prefetch file is one among various objects, presence of which confirms the execution of a particular application. Prefetch gives additional inside for the purpose of investigation. Thus, this paper brings out the forensic value of it, the tools required to decode the information it contains and also look in various caveats in interpreting this artifact to learn its strength and weaknesses to properly incorporate in support of opinion derived by the analyst. In this work, Prefetch is forensically examined to bring out its forensic value, knowledge it contains and all of that in whole or in parts can be used to help advance in investigation. Paper also brings out the difference in format of this artifact among various version of Windows OS.
{"title":"Forensic Analysis of Windows 11 Prefetch Artifact","authors":"Akash Budhrani, Upasna Singh, Bhupendra Singh","doi":"10.1109/IBSSC56953.2022.10037260","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037260","url":null,"abstract":"The Operating System creates numerous objects to improve its efficiency and user experience and such objects are called artifacts. These artifacts record crucial data about the user activity. Such artifacts are the start point of any investigation as they can be an additional lead to a forensic triage. Prefetch file is one among various objects, presence of which confirms the execution of a particular application. Prefetch gives additional inside for the purpose of investigation. Thus, this paper brings out the forensic value of it, the tools required to decode the information it contains and also look in various caveats in interpreting this artifact to learn its strength and weaknesses to properly incorporate in support of opinion derived by the analyst. In this work, Prefetch is forensically examined to bring out its forensic value, knowledge it contains and all of that in whole or in parts can be used to help advance in investigation. Paper also brings out the difference in format of this artifact among various version of Windows OS.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"29 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":"124401079","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/IBSSC56953.2022.10037272
Muskan Goyal, Prachi Tawde
Users of products and services, as human beings have a wide range of personalities. This is being experienced right from the initial days of e-commerce and m-commerce in India. In this research an attempt has been made to predict personalities using MBTI (Myers Briggs Type Indicator) based approach making use of natural language based processing, machine learning and transformer based modelling. As each human being is unique and exhibits different personality trait, therefore it is impractical to offer a generalized treatment for all users. But it is possible to categorize individuals, in terms of their defining characteristics based on MBTI based approach, which groups personalities/users into 16 groups and thus helps in predicting personalities. In this study authors made an attempt to extract social media based information of users through their accounts to characterize users into one of the 16 MBTI personality types. For this prediction and modelling, authors made use of pre-processed data from Kaggle, which was then fed into the transformer for modelling/processing. Based on the information it gets, like comments, post captions, reviews, etc., the transformer is fine-tuned to predict the user's personality. The required qualities of the model were taken into account while coding the transformer's parameters. Additionally, an attempt is also made to compare the outcomes of two trained transformer models. Authors report that the prediction accuracy of their modelling as 64%, outperforming all other models used. The testing data had a 76% precision.
产品和服务的使用者，如同人类一样，有着广泛的个性。这种情况从印度电子商务和移动商务的最初几天就开始了。在这项研究中，利用基于自然语言的处理、机器学习和基于变压器的建模，尝试使用MBTI (Myers Briggs Type Indicator)方法来预测性格。由于每个人都是独特的，表现出不同的个性特征，因此对所有用户提供通用的治疗是不切实际的。但是，根据MBTI的定义特征对个体进行分类是可能的，MBTI将个性/用户分为16组，从而有助于预测个性。在这项研究中，作者试图通过用户的账户提取基于社交媒体的用户信息，将用户定性为16种MBTI人格类型之一。为了进行预测和建模，作者使用了Kaggle的预处理数据，然后将其输入变压器进行建模/处理。根据它获得的信息，比如评论、帖子标题、评论等，转换器会进行微调，以预测用户的个性。在对变压器参数进行编码时，考虑了模型的质量要求。此外，还尝试比较两种训练后的变压器模型的结果。作者报告说，他们的模型的预测精度为64%，优于所有其他使用的模型。测试数据的精确度为76%。
{"title":"A research attempt to predict and model personalities through users' social media details","authors":"Muskan Goyal, Prachi Tawde","doi":"10.1109/IBSSC56953.2022.10037272","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037272","url":null,"abstract":"Users of products and services, as human beings have a wide range of personalities. This is being experienced right from the initial days of e-commerce and m-commerce in India. In this research an attempt has been made to predict personalities using MBTI (Myers Briggs Type Indicator) based approach making use of natural language based processing, machine learning and transformer based modelling. As each human being is unique and exhibits different personality trait, therefore it is impractical to offer a generalized treatment for all users. But it is possible to categorize individuals, in terms of their defining characteristics based on MBTI based approach, which groups personalities/users into 16 groups and thus helps in predicting personalities. In this study authors made an attempt to extract social media based information of users through their accounts to characterize users into one of the 16 MBTI personality types. For this prediction and modelling, authors made use of pre-processed data from Kaggle, which was then fed into the transformer for modelling/processing. Based on the information it gets, like comments, post captions, reviews, etc., the transformer is fine-tuned to predict the user's personality. The required qualities of the model were taken into account while coding the transformer's parameters. Additionally, an attempt is also made to compare the outcomes of two trained transformer models. Authors report that the prediction accuracy of their modelling as 64%, outperforming all other models used. The testing data had a 76% precision.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"26 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":"116245104","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/IBSSC56953.2022.10037374
Arya Shah
Monkeypox is classified as a viral zoonotic disease which is transmitted to humans from animals. The recent outbreak of the Monkeypox virus has affected more than 40 countries. With the rapid spread and ever-growing challenges of provisioning PCR (Polymerase Chain Reaction) Tests in areas with less availability, computer aided methods incorporating Deep Learning techniques for automated detection of skin lesions proves to be a feasible solution. The paper proposes a Transfer Learning based approach to classify Monkeypox skin lesions from chickenpox and normal skin images. A total of 5 Transfer Learning models namely- MobileNetv2, ResNet50, Inceptionv3, EfficientNetB5 and Xception have been trained on a skin lesion image dataset sourced from News reports, public health websites and case studies. A comparison of the trained models is provided to select the best performing model which can be further utilized in any application for quick, automated detection of monkeypox skin lesions in remote areas. MobileNetv2 provided the best model accuracy of 98.78% for classification of monkeypox skin lesion images.
{"title":"Monkeypox Skin Lesion Classification Using Transfer Learning Approach","authors":"Arya Shah","doi":"10.1109/IBSSC56953.2022.10037374","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037374","url":null,"abstract":"Monkeypox is classified as a viral zoonotic disease which is transmitted to humans from animals. The recent outbreak of the Monkeypox virus has affected more than 40 countries. With the rapid spread and ever-growing challenges of provisioning PCR (Polymerase Chain Reaction) Tests in areas with less availability, computer aided methods incorporating Deep Learning techniques for automated detection of skin lesions proves to be a feasible solution. The paper proposes a Transfer Learning based approach to classify Monkeypox skin lesions from chickenpox and normal skin images. A total of 5 Transfer Learning models namely- MobileNetv2, ResNet50, Inceptionv3, EfficientNetB5 and Xception have been trained on a skin lesion image dataset sourced from News reports, public health websites and case studies. A comparison of the trained models is provided to select the best performing model which can be further utilized in any application for quick, automated detection of monkeypox skin lesions in remote areas. MobileNetv2 provided the best model accuracy of 98.78% for classification of monkeypox skin lesion images.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"79 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":"114313364","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：