Pub Date : 2021-09-23DOI: 10.1109/CCGE50943.2021.9776372
K. Vengatesan, Abhishek Kumar, Ankit Kumar, K. Kharade, S. Kharade, R. K. Kamat
In the Indian economy, the stock market and bonds play a significant role in predicting any specific company's economic rate or growth rate. There are a lot of parameters that need to be considered for predicting the value of any stock. Stocks are certificates of ownership of a company that describe the rights to the company's profits. Finally, we will get a share or ownership from the company based on the growth rate for every period. A bond is a type of investment from which a user will get monthly or yearly interest from the company based on the profit. Both share and bond will provide guaranteed returns to the customers. In this proposed work, we have taken hardware-based company stock data set. Using time-series data analytics techniques, we will study the value of every stock based on the historical data and estimate which company can have a high scope in the future based on the parameters like opening value and closing value of stock.
{"title":"Stock Market Analysis using Time Series Data Analytics Techniques","authors":"K. Vengatesan, Abhishek Kumar, Ankit Kumar, K. Kharade, S. Kharade, R. K. Kamat","doi":"10.1109/CCGE50943.2021.9776372","DOIUrl":"https://doi.org/10.1109/CCGE50943.2021.9776372","url":null,"abstract":"In the Indian economy, the stock market and bonds play a significant role in predicting any specific company's economic rate or growth rate. There are a lot of parameters that need to be considered for predicting the value of any stock. Stocks are certificates of ownership of a company that describe the rights to the company's profits. Finally, we will get a share or ownership from the company based on the growth rate for every period. A bond is a type of investment from which a user will get monthly or yearly interest from the company based on the profit. Both share and bond will provide guaranteed returns to the customers. In this proposed work, we have taken hardware-based company stock data set. Using time-series data analytics techniques, we will study the value of every stock based on the historical data and estimate which company can have a high scope in the future based on the parameters like opening value and closing value of stock.","PeriodicalId":130452,"journal":{"name":"2021 International Conference on Computing, Communication and Green Engineering (CCGE)","volume":"313 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114413968","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}
The primary goal of any electric power generation system is to provide a sufficient amount of electricity to consumers without jeopardizing the system's economic viability. The modernization of the power grid has resulted in a significant rise in power demand, which has increased the cost of producing electrical energy. When the cost of output rises, so does the cost of transferring energy to the end consumer. As a result, the output of energy at various stages of a power system must be optimized. As a result, the cost per unit of thermal energy output is reduced while load demand requirements and transmission losses are maintained. These complex non-linear quadratic functions with Multiple Fuels lead to a non-Convex problem for steam thermal generating systems, according to previous studies. Perfect Economic Load Dispatch (ELD) modelling for steam thermal generating units is possible with multiple fuels. Because acute variations and disruptions in the incremental cost function are possible, it is difficult to simplify the non-convex problem using existing techniques. Oppositional Teaching Learning Based Optimization (OTLBO) is used to address the ELD problem in this research. Under various load demands, the proposed solution was applied to a 6-unit test system, a 10-unit test system, and a 14-unit test system, and the results were evaluated using the Teaching Learning Based Optimization (TLBO) algorithm.
{"title":"A Nonconvex Constrained based Optimal Load Scheduling of Generators with Multiple Fuels using meta-heuristic Algorithms","authors":"D. Rao, Chiranjeevi Tulluri, Bharath Kumar Narukullapati, Haqqani Arshad, Raju Mv","doi":"10.1109/CCGE50943.2021.9776402","DOIUrl":"https://doi.org/10.1109/CCGE50943.2021.9776402","url":null,"abstract":"The primary goal of any electric power generation system is to provide a sufficient amount of electricity to consumers without jeopardizing the system's economic viability. The modernization of the power grid has resulted in a significant rise in power demand, which has increased the cost of producing electrical energy. When the cost of output rises, so does the cost of transferring energy to the end consumer. As a result, the output of energy at various stages of a power system must be optimized. As a result, the cost per unit of thermal energy output is reduced while load demand requirements and transmission losses are maintained. These complex non-linear quadratic functions with Multiple Fuels lead to a non-Convex problem for steam thermal generating systems, according to previous studies. Perfect Economic Load Dispatch (ELD) modelling for steam thermal generating units is possible with multiple fuels. Because acute variations and disruptions in the incremental cost function are possible, it is difficult to simplify the non-convex problem using existing techniques. Oppositional Teaching Learning Based Optimization (OTLBO) is used to address the ELD problem in this research. Under various load demands, the proposed solution was applied to a 6-unit test system, a 10-unit test system, and a 14-unit test system, and the results were evaluated using the Teaching Learning Based Optimization (TLBO) algorithm.","PeriodicalId":130452,"journal":{"name":"2021 International Conference on Computing, Communication and Green Engineering (CCGE)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126458193","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 : 2021-09-23DOI: 10.1109/CCGE50943.2021.9776401
Abha Sharma, Prasenjit Das, R. B. Patel
A lot of progression has been observed in Mobile Wireless Sensor Network (MWSN) in modern era due to its applications in vicinity. Recent trends show how it is very challenging to retain stability in the network in terms of delay, packet delivery ratio and stability. Nodes mobilization is vital in stabilizing the network, and various routing protocols are used to maintain connectivity, throughput, coverage, minimal energy cost. As a matter of fact it is really demanding that a single routing protocol will be able to cover up for numerous circumstances altogether. Several routing protocols have been worked upon for a variety of network scenarios. The categorization of routing protocols is countered on the basis of type of network structure, information status, network efficiency and mobility. In this paper Hierarchal routing protocols classification is presented which can help enhance network life and save energy consumption.
{"title":"Mobile Wireless Sensor Networks And Hierarchical Routing Protocols: A Review","authors":"Abha Sharma, Prasenjit Das, R. B. Patel","doi":"10.1109/CCGE50943.2021.9776401","DOIUrl":"https://doi.org/10.1109/CCGE50943.2021.9776401","url":null,"abstract":"A lot of progression has been observed in Mobile Wireless Sensor Network (MWSN) in modern era due to its applications in vicinity. Recent trends show how it is very challenging to retain stability in the network in terms of delay, packet delivery ratio and stability. Nodes mobilization is vital in stabilizing the network, and various routing protocols are used to maintain connectivity, throughput, coverage, minimal energy cost. As a matter of fact it is really demanding that a single routing protocol will be able to cover up for numerous circumstances altogether. Several routing protocols have been worked upon for a variety of network scenarios. The categorization of routing protocols is countered on the basis of type of network structure, information status, network efficiency and mobility. In this paper Hierarchal routing protocols classification is presented which can help enhance network life and save energy consumption.","PeriodicalId":130452,"journal":{"name":"2021 International Conference on Computing, Communication and Green Engineering (CCGE)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132091825","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 : 2021-09-23DOI: 10.1109/CCGE50943.2021.9776318
Pratiksha Sahane, S. Pangaonkar, Shridhar Khandekar
Vast industrial growth has increased the demand of automatic speech recognition for various automation and human machine interaction application. Performance of various artificial intelligence based approaches is limited because of the speech disability caused due to communication disorders, neurogenic speech disorder or psychological speech disorders. The dysarthric disorder is neurogenic speech disorder that limits the human voice articulation capability. This paper presents, dysarthric speech detection using Multi-Taper Mel Frequency Cepstral coefficients (MTMFCC) that is capable to smallest variation over the dysarthric speech. The efficiency of the proposed algorithm is estimated using the K-Nearest Neighbor (KNN) classifier and support vector machine (SVM) based on accuracy, sensitivity and specificity. The system has shown 99.04 % and 96.00 % accuracy for the MTMFCC+KNN and MTMFCC+SVM which is superior to traditional MFCC.
{"title":"Dysarthric Speech Recognition using Multi-Taper Mel Frequency Cepstrum Coefficients","authors":"Pratiksha Sahane, S. Pangaonkar, Shridhar Khandekar","doi":"10.1109/CCGE50943.2021.9776318","DOIUrl":"https://doi.org/10.1109/CCGE50943.2021.9776318","url":null,"abstract":"Vast industrial growth has increased the demand of automatic speech recognition for various automation and human machine interaction application. Performance of various artificial intelligence based approaches is limited because of the speech disability caused due to communication disorders, neurogenic speech disorder or psychological speech disorders. The dysarthric disorder is neurogenic speech disorder that limits the human voice articulation capability. This paper presents, dysarthric speech detection using Multi-Taper Mel Frequency Cepstral coefficients (MTMFCC) that is capable to smallest variation over the dysarthric speech. The efficiency of the proposed algorithm is estimated using the K-Nearest Neighbor (KNN) classifier and support vector machine (SVM) based on accuracy, sensitivity and specificity. The system has shown 99.04 % and 96.00 % accuracy for the MTMFCC+KNN and MTMFCC+SVM which is superior to traditional MFCC.","PeriodicalId":130452,"journal":{"name":"2021 International Conference on Computing, Communication and Green Engineering (CCGE)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131347064","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 : 2021-09-23DOI: 10.1109/CCGE50943.2021.9776391
Meenakshi Sharma, R. Kaushal, Amit Sharma
Blockchain, a futuristic technology has great potential to provide a revolutionary boom in the healthcare industry by providing a secure, decentralized and network-based peer-to-peer solution to reinvent the way patients store and share their electronic clinical information. Blockchain is most booming technologies on the planet for the next three decades. Purpose of the study is to explore the present literature on blockchain in the field of healthcare and identify the applications, challenges, and open research questions related to electronic health records on blockchain technology. A systematic literature review method is used to support and facilitate understanding of this ever-growing accounting technology Many reputable articles reviewed in this document were also accessed which resulted in adoption challenges, technical issues, and some research questions formed in conjunction with interoperability standards. To this end, even more research is needed to understand the technical side and utility of blockchain in the domain of healthcare.
{"title":"A Review on Sustainability of Blockchain in Electronic Health Records","authors":"Meenakshi Sharma, R. Kaushal, Amit Sharma","doi":"10.1109/CCGE50943.2021.9776391","DOIUrl":"https://doi.org/10.1109/CCGE50943.2021.9776391","url":null,"abstract":"Blockchain, a futuristic technology has great potential to provide a revolutionary boom in the healthcare industry by providing a secure, decentralized and network-based peer-to-peer solution to reinvent the way patients store and share their electronic clinical information. Blockchain is most booming technologies on the planet for the next three decades. Purpose of the study is to explore the present literature on blockchain in the field of healthcare and identify the applications, challenges, and open research questions related to electronic health records on blockchain technology. A systematic literature review method is used to support and facilitate understanding of this ever-growing accounting technology Many reputable articles reviewed in this document were also accessed which resulted in adoption challenges, technical issues, and some research questions formed in conjunction with interoperability standards. To this end, even more research is needed to understand the technical side and utility of blockchain in the domain of healthcare.","PeriodicalId":130452,"journal":{"name":"2021 International Conference on Computing, Communication and Green Engineering (CCGE)","volume":"159 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123021102","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 : 2021-09-23DOI: 10.1109/ccge50943.2021.9776478
Monitoramento Participativo, aGENDAS tRANSVERSAIS, aGENDAS tRANSVERSAIS, Monitoramento Participativo, Plano Mais Brasil, Relatório DE Monitoramento, Miriam Belchior, Secretária Executiva, Eva Maria Cella, Augusto da Silva Lima
{"title":"[Agendas]","authors":"Monitoramento Participativo, aGENDAS tRANSVERSAIS, aGENDAS tRANSVERSAIS, Monitoramento Participativo, Plano Mais Brasil, Relatório DE Monitoramento, Miriam Belchior, Secretária Executiva, Eva Maria Cella, Augusto da Silva Lima","doi":"10.1109/ccge50943.2021.9776478","DOIUrl":"https://doi.org/10.1109/ccge50943.2021.9776478","url":null,"abstract":"","PeriodicalId":130452,"journal":{"name":"2021 International Conference on Computing, Communication and Green Engineering (CCGE)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114569314","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 : 2021-09-23DOI: 10.1109/CCGE50943.2021.9776377
R. Mittal, A. Mittal, Jaiteg Singh, Vikas Rattan, Varun Malik
Store choice is a function of store image which in turn comprises of store attributes. Different store attributes are evaluated differently by shoppers. For researchers and managers, it is not easy to understand how shoppers assess the multiple attributes that a store has. The high number of attributes needs to be reduced to a more manageable number and this can be done using the data mining technique of feature selection or factor analysis. Once this data mining technique is applied, the emerging factors can be processed to understand shoppers store choice criteria much better. This study assesses 23 store attributes evaluated by 197 shoppers of hypermarkets in India which were reduced to seven factors. The seven factors were ranked. Price / Value related factor was ranked highest.
{"title":"Principal Component Analysis based Feature Selection Driving Store Choice: A Data Mining Approach","authors":"R. Mittal, A. Mittal, Jaiteg Singh, Vikas Rattan, Varun Malik","doi":"10.1109/CCGE50943.2021.9776377","DOIUrl":"https://doi.org/10.1109/CCGE50943.2021.9776377","url":null,"abstract":"Store choice is a function of store image which in turn comprises of store attributes. Different store attributes are evaluated differently by shoppers. For researchers and managers, it is not easy to understand how shoppers assess the multiple attributes that a store has. The high number of attributes needs to be reduced to a more manageable number and this can be done using the data mining technique of feature selection or factor analysis. Once this data mining technique is applied, the emerging factors can be processed to understand shoppers store choice criteria much better. This study assesses 23 store attributes evaluated by 197 shoppers of hypermarkets in India which were reduced to seven factors. The seven factors were ranked. Price / Value related factor was ranked highest.","PeriodicalId":130452,"journal":{"name":"2021 International Conference on Computing, Communication and Green Engineering (CCGE)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126594905","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 : 2021-09-23DOI: 10.1109/CCGE50943.2021.9776386
M. Dadhich, Ruchi Doshi, S. Mathur, Rajesh Meena, Rajat Kumar Gujral, P. Dhotre
One of the most remarkable changes in the academic Diaspora is the international creation of virtual platforms, which has given rise to a new edge system of learning. Covid-19 presents a unique and severe problem on every front. The nationwide shutdown by the administration aims to control the diffusion of Covid-19 at education institutions across the country. Many (local, national, and worldwide) institutions have implemented a reliable and beneficial contactless atmosphere for students and faculties to maintain the continuity of learning. As a result, teachers and students are greatly influenced by the new-age virtual teaching method adopted and implemented. The survey respondents were picked by a combination of online surveys and personality tests, and then the questionnaire they were given included both closed- and open-ended items. The numbers of university and secondary school portals have recently seen an upward trend. So, to better investigate the abilities of teachers and learners to identify the efficiency of dominating content delivery methods, a hybrid approach of the exploratory study was employed. Students and faculty, 140 each who have taken web-based learning at 25 Indian institutions, are sampled using a snowball sampling methodology. The results of the t-test demonstrated a considerable divergence in teaching-learning impressions between faculty and students on three manifests ($mathrm{p} < 0.005$). Learners' responses differed from faculty responses, and statistically significant differences were found, such as scientific material can be taught effectively online, improved technocratic pedagogy is the core part of e-learning, reliance on computers/connectivity.
{"title":"Empirical Study of Awareness towards Blended e-learning Gateways during Covid-19 Lockdown","authors":"M. Dadhich, Ruchi Doshi, S. Mathur, Rajesh Meena, Rajat Kumar Gujral, P. Dhotre","doi":"10.1109/CCGE50943.2021.9776386","DOIUrl":"https://doi.org/10.1109/CCGE50943.2021.9776386","url":null,"abstract":"One of the most remarkable changes in the academic Diaspora is the international creation of virtual platforms, which has given rise to a new edge system of learning. Covid-19 presents a unique and severe problem on every front. The nationwide shutdown by the administration aims to control the diffusion of Covid-19 at education institutions across the country. Many (local, national, and worldwide) institutions have implemented a reliable and beneficial contactless atmosphere for students and faculties to maintain the continuity of learning. As a result, teachers and students are greatly influenced by the new-age virtual teaching method adopted and implemented. The survey respondents were picked by a combination of online surveys and personality tests, and then the questionnaire they were given included both closed- and open-ended items. The numbers of university and secondary school portals have recently seen an upward trend. So, to better investigate the abilities of teachers and learners to identify the efficiency of dominating content delivery methods, a hybrid approach of the exploratory study was employed. Students and faculty, 140 each who have taken web-based learning at 25 Indian institutions, are sampled using a snowball sampling methodology. The results of the t-test demonstrated a considerable divergence in teaching-learning impressions between faculty and students on three manifests ($mathrm{p} < 0.005$). Learners' responses differed from faculty responses, and statistically significant differences were found, such as scientific material can be taught effectively online, improved technocratic pedagogy is the core part of e-learning, reliance on computers/connectivity.","PeriodicalId":130452,"journal":{"name":"2021 International Conference on Computing, Communication and Green Engineering (CCGE)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126612552","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 : 2021-09-23DOI: 10.1109/CCGE50943.2021.9776398
Aparna S. Nagure, S. Shahapure
The rainfall-runoff analysis and modeling have been the subject of a large number of research activities and a range of types of models have been developed in the last few decades, to predict the runoff well in advance to avoid the huge amount of losses due to floods. However, all these research activities are focused on the result and accuracy of models and their comparative study. It often remains unclear which model is best under which conditions. It is necessary to select the appropriate rainfall-runoff model for the watershed area according to its physical/chemical/biological characteristics. In this paper, one of the significant characteristics of the watershed that is the size of the case study area is selected as a parameter to understand how it affects the selection of the model. To understand this, 42 research papers published between 2000 to 2019 have been reviewed and categorized according to the size of the watershed, climatic conditions, and type of models used for rainfall-runoff analysis. The result obtained indicates that for major research work, black box models or data-driven models have been used for the watershed of size ranging between 250 km2 to 10000 km2. Similarly, maximum work is carried out for medium size watershed areas.
{"title":"Effect of Watershed Characteristics on a Rainfall Runoff Analysis and Hydrological Model Selection - A review","authors":"Aparna S. Nagure, S. Shahapure","doi":"10.1109/CCGE50943.2021.9776398","DOIUrl":"https://doi.org/10.1109/CCGE50943.2021.9776398","url":null,"abstract":"The rainfall-runoff analysis and modeling have been the subject of a large number of research activities and a range of types of models have been developed in the last few decades, to predict the runoff well in advance to avoid the huge amount of losses due to floods. However, all these research activities are focused on the result and accuracy of models and their comparative study. It often remains unclear which model is best under which conditions. It is necessary to select the appropriate rainfall-runoff model for the watershed area according to its physical/chemical/biological characteristics. In this paper, one of the significant characteristics of the watershed that is the size of the case study area is selected as a parameter to understand how it affects the selection of the model. To understand this, 42 research papers published between 2000 to 2019 have been reviewed and categorized according to the size of the watershed, climatic conditions, and type of models used for rainfall-runoff analysis. The result obtained indicates that for major research work, black box models or data-driven models have been used for the watershed of size ranging between 250 km2 to 10000 km2. Similarly, maximum work is carried out for medium size watershed areas.","PeriodicalId":130452,"journal":{"name":"2021 International Conference on Computing, Communication and Green Engineering (CCGE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128961435","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 : 2021-09-23DOI: 10.1109/CCGE50943.2021.9776440
Md. Rekib Ahmed, Neeraj Bhadani, I. Chakraborty
With the emergence of deep neural networks along with high-end computers that can process deep architectures, there has been a lot of research when Computer Vision and Natural Language Processing has been fused into a single problem. To enable students and researchers to deep dive into multimodal deep learning Facebook AI Research team published a dataset on hateful meme classification “The Hateful Meme Challenge Dataset” in May 2020 that gave us the motivation to test ourselves and an opportunity to learn more about the dataset. The rise of communication on the internet with memes as a medium, they have been used to convey incorrect information, political agendas and also has led to cyberbullying, trolling etc. This results in the need of creating an automated tool that can detect such hateful content published on the internet and remove it at the root level before it does any harm. This paper intends to adopt Unimodal Text and Image models using Bert, LSTM and VGG16, Resnet50, SE-Resnet50, XSE-Resnet architectures and combining them into Multimodal models for effective prediction of a hateful meme. The paper compares various architectures both unimodal models and multimodal models on the evaluation metrics AUC-ROC score, F1 score and accuracy score.)
{"title":"Hateful Meme Prediction Model Using Multimodal Deep Learning","authors":"Md. Rekib Ahmed, Neeraj Bhadani, I. Chakraborty","doi":"10.1109/CCGE50943.2021.9776440","DOIUrl":"https://doi.org/10.1109/CCGE50943.2021.9776440","url":null,"abstract":"With the emergence of deep neural networks along with high-end computers that can process deep architectures, there has been a lot of research when Computer Vision and Natural Language Processing has been fused into a single problem. To enable students and researchers to deep dive into multimodal deep learning Facebook AI Research team published a dataset on hateful meme classification “The Hateful Meme Challenge Dataset” in May 2020 that gave us the motivation to test ourselves and an opportunity to learn more about the dataset. The rise of communication on the internet with memes as a medium, they have been used to convey incorrect information, political agendas and also has led to cyberbullying, trolling etc. This results in the need of creating an automated tool that can detect such hateful content published on the internet and remove it at the root level before it does any harm. This paper intends to adopt Unimodal Text and Image models using Bert, LSTM and VGG16, Resnet50, SE-Resnet50, XSE-Resnet architectures and combining them into Multimodal models for effective prediction of a hateful meme. The paper compares various architectures both unimodal models and multimodal models on the evaluation metrics AUC-ROC score, F1 score and accuracy score.)","PeriodicalId":130452,"journal":{"name":"2021 International Conference on Computing, Communication and Green Engineering (CCGE)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121943102","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：