Pub Date : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9908759
Atharva Pawar, Chirag Fatnani, Rajani Sonavane, Riya Waghmare, Sarang A. Saoji
Q-R codes are utilised for a variety of purposes, including accessing online web-pages and making a settlement. The Internet facilitates a wide range of illegal acts, including unsolicited e-marketing, financial embezzlement, and malicious distribution. Even though all the users identify the presence of Q-R codes visually, the information stored in those codes can only be accessed through an allocated Q-R code decoder. Q-R codes have also been shown to be used as an effective attack vector, For example techniques include social engineering, phishing, pharming, etc. Harmful codes are distributed under false pretences in congested areas, or malicious Q-R codes are pasted over current ones on billboards. Finally, consumers rely on decoder operating system to determine a random Q-R code is whether malicious or benign.For the purpose of this report, we consider the identification of malicious Q-R codes as a two-way classification problem in this research, and we test the effectiveness of many well-known M-L algorithms, including namely K-Nearest Neighbour, Random Forest, Binary LSTM and Support Vector Machine. This implies that the proposed method might be deemed an optimal and user-friendly QR code security solution. We created a prototype to test our recommendations and found it to be secure and usable in protecting users from harmful QR Codes.
{"title":"Secure QR Code Scanner to Detect Malicious URL using Machine Learning","authors":"Atharva Pawar, Chirag Fatnani, Rajani Sonavane, Riya Waghmare, Sarang A. Saoji","doi":"10.1109/ASIANCON55314.2022.9908759","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9908759","url":null,"abstract":"Q-R codes are utilised for a variety of purposes, including accessing online web-pages and making a settlement. The Internet facilitates a wide range of illegal acts, including unsolicited e-marketing, financial embezzlement, and malicious distribution. Even though all the users identify the presence of Q-R codes visually, the information stored in those codes can only be accessed through an allocated Q-R code decoder. Q-R codes have also been shown to be used as an effective attack vector, For example techniques include social engineering, phishing, pharming, etc. Harmful codes are distributed under false pretences in congested areas, or malicious Q-R codes are pasted over current ones on billboards. Finally, consumers rely on decoder operating system to determine a random Q-R code is whether malicious or benign.For the purpose of this report, we consider the identification of malicious Q-R codes as a two-way classification problem in this research, and we test the effectiveness of many well-known M-L algorithms, including namely K-Nearest Neighbour, Random Forest, Binary LSTM and Support Vector Machine. This implies that the proposed method might be deemed an optimal and user-friendly QR code security solution. We created a prototype to test our recommendations and found it to be secure and usable in protecting users from harmful QR Codes.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"147 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116106444","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 : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9908975
D. Tripathi, S. Biswas, S. Reshmi, Arpita Nath Boruah, B. Purkayastha
Diabetes is an incurable disease which is due to a high level of sugar in the blood over a long period of time. Hence, early prediction is required to reduce its severity significantly. Now-a-days Machine Learning (ML) community has been working on diabetes prediction and much research has been done for decades for its prediction. Keeping in view of its severity, this paper proposes a model, named Diabetes Expert System using Machine Learning Analytics (DESMLA) to explore the diabetes data to predict the disease more effectively. The Diabetes Dataset (DD) is imbalanced in nature; therefore, the DESMLA model uses the 5 most prominent oversampling techniques namely SMOTE, Borderline SMOTE, ADASYN SMOTE, K-Means SMOTE and Gaussian SMOTE to get rid of this class imbalance problem of the diabetes dataset. DESMLA model also performs feature selection to determine only the significant features for diabetes prediction as DD may contain some irrelevant and redundant features. DESMLA shows the comparison between filter and wrapper approaches for feature selection. From the experimental results, it is observed that DESMLA with wrapper approach produces better performance than that of filter approach. The performance improvement of DESMLA with class imbalance treatment and feature selection is observed which is promising and significant.
{"title":"Diabetes Prediction Using Machine Learning Analytics: Ensemble Learning Techniques","authors":"D. Tripathi, S. Biswas, S. Reshmi, Arpita Nath Boruah, B. Purkayastha","doi":"10.1109/ASIANCON55314.2022.9908975","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9908975","url":null,"abstract":"Diabetes is an incurable disease which is due to a high level of sugar in the blood over a long period of time. Hence, early prediction is required to reduce its severity significantly. Now-a-days Machine Learning (ML) community has been working on diabetes prediction and much research has been done for decades for its prediction. Keeping in view of its severity, this paper proposes a model, named Diabetes Expert System using Machine Learning Analytics (DESMLA) to explore the diabetes data to predict the disease more effectively. The Diabetes Dataset (DD) is imbalanced in nature; therefore, the DESMLA model uses the 5 most prominent oversampling techniques namely SMOTE, Borderline SMOTE, ADASYN SMOTE, K-Means SMOTE and Gaussian SMOTE to get rid of this class imbalance problem of the diabetes dataset. DESMLA model also performs feature selection to determine only the significant features for diabetes prediction as DD may contain some irrelevant and redundant features. DESMLA shows the comparison between filter and wrapper approaches for feature selection. From the experimental results, it is observed that DESMLA with wrapper approach produces better performance than that of filter approach. The performance improvement of DESMLA with class imbalance treatment and feature selection is observed which is promising and significant.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124670081","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 : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9909485
Sneha Annappanavar, P. Chavan
Voting is a process of choosing leader who we trust and will work for our benefit. The process of choosing the leader is currently through EVM having the facts to be considered like Voter Authentication, Security of vote casted and Vote anonymity along with Vote verification. The very first thing which requires to improve is moving from EVM voting to Secure Electronic Voting where digitization plays an important role in providing the access to vote from anywhere which meets the current need for keeping us updated with technology and serving the citizens with the advanced Technology. Keeping Electronic voting in mind here we propose the very first step towards new way of voting that is voter authentication. Using visual cryptography schemes, we can here implement the voter authentication mechanism using fingerprint matching. Fingerprint matching is one of the biometric which can be considered as one of the key steps towards voter authentication along with visual cryptography.
{"title":"(2,2) Visual Cryptography Based Biometric Authentication Mechanism for Online Elections","authors":"Sneha Annappanavar, P. Chavan","doi":"10.1109/ASIANCON55314.2022.9909485","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9909485","url":null,"abstract":"Voting is a process of choosing leader who we trust and will work for our benefit. The process of choosing the leader is currently through EVM having the facts to be considered like Voter Authentication, Security of vote casted and Vote anonymity along with Vote verification. The very first thing which requires to improve is moving from EVM voting to Secure Electronic Voting where digitization plays an important role in providing the access to vote from anywhere which meets the current need for keeping us updated with technology and serving the citizens with the advanced Technology. Keeping Electronic voting in mind here we propose the very first step towards new way of voting that is voter authentication. Using visual cryptography schemes, we can here implement the voter authentication mechanism using fingerprint matching. Fingerprint matching is one of the biometric which can be considered as one of the key steps towards voter authentication along with visual cryptography.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123033238","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 : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9909267
Dinisuru Nisal Gunaratna, Pumudu Fernando
Endoscopy is the most widely adhered medical procedure used to examine the gastrointestinal tract of a person. Accurate pathology detection during the endoscopic procedure is crucial as misidentifications or miss rates could reduce the chance of survival for the patient. After the successful collaboration of artificial intelligence with medicine, researchers around the world have tried different techniques in using this for gastroenterology. Our study demonstrates an extensive survey on existing pathology detection methodologies in endoscopic images using the publicly available datasets. The paper also discusses the content of the recently released datasets, preprocessing techniques tried on these datasets and how they affected the performance of the machine learning models. Furthermore, this study discusses how changing architectures of convolutional neural networks could affect the accuracy of models in relation to different datasets. Finally, the paper presents the results of each reviewed literature along with a brief discussion on the gaps that were identified.
{"title":"A Systematic Literature Review of Machine Learning based Approaches on Pathology Detection in Gastrointestinal Endoscopy","authors":"Dinisuru Nisal Gunaratna, Pumudu Fernando","doi":"10.1109/ASIANCON55314.2022.9909267","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9909267","url":null,"abstract":"Endoscopy is the most widely adhered medical procedure used to examine the gastrointestinal tract of a person. Accurate pathology detection during the endoscopic procedure is crucial as misidentifications or miss rates could reduce the chance of survival for the patient. After the successful collaboration of artificial intelligence with medicine, researchers around the world have tried different techniques in using this for gastroenterology. Our study demonstrates an extensive survey on existing pathology detection methodologies in endoscopic images using the publicly available datasets. The paper also discusses the content of the recently released datasets, preprocessing techniques tried on these datasets and how they affected the performance of the machine learning models. Furthermore, this study discusses how changing architectures of convolutional neural networks could affect the accuracy of models in relation to different datasets. Finally, the paper presents the results of each reviewed literature along with a brief discussion on the gaps that were identified.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123868204","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 : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9909153
Shivam Wadhwa, Shailesh Mishra
In this paper, various real time indoor scenarios have been analyzed to provide the on-site practical solutions for efficient communication link establishment for futuristic 5G indoor application. The presented real time indoor scenario contains a transmitting and a receiving antenna with resonant frequency of 60GHz with 4.39 GHz bandwidth. The user equipment (UE) is considered as a receiving antenna which is placed at various coordinates in the room and a 5G transmitting antenna is placed at different positions in the room. The result analysis is carried out to find the best orientation of transmitting antenna such that it gives maximum power at the receiver antenna placed at different positions in the room. The orientation of the transmitting antenna can be implemented electronically using beamforming technique to established the efficient link.
{"title":"Investigation of On-site Channel Model for 5G Indoor Applications","authors":"Shivam Wadhwa, Shailesh Mishra","doi":"10.1109/ASIANCON55314.2022.9909153","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9909153","url":null,"abstract":"In this paper, various real time indoor scenarios have been analyzed to provide the on-site practical solutions for efficient communication link establishment for futuristic 5G indoor application. The presented real time indoor scenario contains a transmitting and a receiving antenna with resonant frequency of 60GHz with 4.39 GHz bandwidth. The user equipment (UE) is considered as a receiving antenna which is placed at various coordinates in the room and a 5G transmitting antenna is placed at different positions in the room. The result analysis is carried out to find the best orientation of transmitting antenna such that it gives maximum power at the receiver antenna placed at different positions in the room. The orientation of the transmitting antenna can be implemented electronically using beamforming technique to established the efficient link.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125568849","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 : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9909208
Arjonel M. Mendoza, Rowell M. Hernandez, Ryndel V. Amorado, Myrna A. Coliat, Poul Isaac C. De Chavez
Network attacks have become necessary in today’s time due to increased network traffic. To determine whether network traffic is normal or anomalous a supervised machine learning system is developed. A network intrusion detection system (IDS) is a must-have piece of a security system. This proposed study aims to discover new patterns automatically from substantial quantities of network data, reducing time manually compiling intrusion and normal behavior patterns. The best model in terms of detection success rate was discovered using a supervised learning algorithm and feature selection method. AdaBoost outperforms Neural Network, kNN, and Naive Bayes in supervised machine learning with feature selection in this study, with a detection accuracy of 100.00%, 99.30%, 91.60%, and 99.70%, respectively. The Network Intrusion Detection dataset is used to classify network intrusions to evaluate the study and it has also been used in past studies. On the other hand, the proposed model proved to be more effective than other studies in terms of intrusion detection. The proposed approach can be used in various fields, including finance, health, and transportation. Furthermore, additional parameter tuning could be added, and different feature selection techniques could be used to improve the performance of the classifiers.
{"title":"Network Data Feature Selection in Detecting Network Intrusion using Supervised Machine Learning Techniques","authors":"Arjonel M. Mendoza, Rowell M. Hernandez, Ryndel V. Amorado, Myrna A. Coliat, Poul Isaac C. De Chavez","doi":"10.1109/ASIANCON55314.2022.9909208","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9909208","url":null,"abstract":"Network attacks have become necessary in today’s time due to increased network traffic. To determine whether network traffic is normal or anomalous a supervised machine learning system is developed. A network intrusion detection system (IDS) is a must-have piece of a security system. This proposed study aims to discover new patterns automatically from substantial quantities of network data, reducing time manually compiling intrusion and normal behavior patterns. The best model in terms of detection success rate was discovered using a supervised learning algorithm and feature selection method. AdaBoost outperforms Neural Network, kNN, and Naive Bayes in supervised machine learning with feature selection in this study, with a detection accuracy of 100.00%, 99.30%, 91.60%, and 99.70%, respectively. The Network Intrusion Detection dataset is used to classify network intrusions to evaluate the study and it has also been used in past studies. On the other hand, the proposed model proved to be more effective than other studies in terms of intrusion detection. The proposed approach can be used in various fields, including finance, health, and transportation. Furthermore, additional parameter tuning could be added, and different feature selection techniques could be used to improve the performance of the classifiers.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129700560","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 : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9909360
Mithra Venkatesan, A. Kulkarni, Radhika Menon, Shashikant Prasad
There is a drastic increase in the number of users who subscribe to the mobile broadband every year. On the other hand, 4G networks have reached the theoretical limits on the data rate and therefore it is not sufficient to accommodate the above increasing traffic. To overcome this problem, new Generation of mobile communication known as fifth generation (5G) comes into the picture. Large network capacity, ultra-low latency and heterogeneous device support are the important features in 5G Technology. Massive MIMO in 5G Technology is built on multi-tier architecture using several low power Base Stations (BSs) inside small cell. Simultaneous usage of the same spectrum causes interference which further reduces the system throughput and network capacity. Thus resource management is an integral part of 5G Heterogeneous Networks (HetNets) so that interference between several base stations and different devices can be minimized. Proposed scheme introduces feedback on the existing cell association and antenna allocation algorithms and also introduces the evolutionary game theory for interference mitigation in HetNets as Game theory can be efficiently modelled for a competitive and compatible environment. Impact of feedback and game theory into RATs on data rate experienced by users and revenue generated by base station from users respectively are observed. Feedback mechanism along with Game theory approach enables to make efficient and effective resource allocation decisions. This facilitates the existing Cell Association algorithms to maximize the data rate of users in different classes and the antenna allocation algorithm to maximize the total profit of the Base station. Both users and base stations are self-interested to maximize their own benefits in terms of data rate and revenue.
{"title":"Interference Mitigation Approach using Massive MIMO towards 5G networks","authors":"Mithra Venkatesan, A. Kulkarni, Radhika Menon, Shashikant Prasad","doi":"10.1109/ASIANCON55314.2022.9909360","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9909360","url":null,"abstract":"There is a drastic increase in the number of users who subscribe to the mobile broadband every year. On the other hand, 4G networks have reached the theoretical limits on the data rate and therefore it is not sufficient to accommodate the above increasing traffic. To overcome this problem, new Generation of mobile communication known as fifth generation (5G) comes into the picture. Large network capacity, ultra-low latency and heterogeneous device support are the important features in 5G Technology. Massive MIMO in 5G Technology is built on multi-tier architecture using several low power Base Stations (BSs) inside small cell. Simultaneous usage of the same spectrum causes interference which further reduces the system throughput and network capacity. Thus resource management is an integral part of 5G Heterogeneous Networks (HetNets) so that interference between several base stations and different devices can be minimized. Proposed scheme introduces feedback on the existing cell association and antenna allocation algorithms and also introduces the evolutionary game theory for interference mitigation in HetNets as Game theory can be efficiently modelled for a competitive and compatible environment. Impact of feedback and game theory into RATs on data rate experienced by users and revenue generated by base station from users respectively are observed. Feedback mechanism along with Game theory approach enables to make efficient and effective resource allocation decisions. This facilitates the existing Cell Association algorithms to maximize the data rate of users in different classes and the antenna allocation algorithm to maximize the total profit of the Base station. Both users and base stations are self-interested to maximize their own benefits in terms of data rate and revenue.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129599432","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 : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9908931
G. P, Sanjay Kumar, Jambi Ratna Raja Kumar, Saju Raj T
Wireless Sensor Networks (WSNs) is main IoT module that gathers data from the environment and sends it to the destinations. The IOT may contain a broad variety of devices. Interconnecting several singly operating IoT devices via internet presents various issues, security that is remains a major concern given the large and frequently unknown audience. For resource-constrained nodes, most known techniques are very recursive. The sensor node’s resource will be severely shortened, compromising communication and security. However, the opponents' behaviour in WSNs has never been studied. The IoT network and its applications need a sophisticated security architecture to protect both gateway and sensor nodes from attacks. A secure communication system is the major goal of this work.
{"title":"Design of Secure Communication Methodologies for WSN Assisted IoT Applications","authors":"G. P, Sanjay Kumar, Jambi Ratna Raja Kumar, Saju Raj T","doi":"10.1109/ASIANCON55314.2022.9908931","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9908931","url":null,"abstract":"Wireless Sensor Networks (WSNs) is main IoT module that gathers data from the environment and sends it to the destinations. The IOT may contain a broad variety of devices. Interconnecting several singly operating IoT devices via internet presents various issues, security that is remains a major concern given the large and frequently unknown audience. For resource-constrained nodes, most known techniques are very recursive. The sensor node’s resource will be severely shortened, compromising communication and security. However, the opponents' behaviour in WSNs has never been studied. The IoT network and its applications need a sophisticated security architecture to protect both gateway and sensor nodes from attacks. A secure communication system is the major goal of this work.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128504618","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 : 2022-08-26DOI: 10.1109/ASIANCON55314.2022.9909083
Avani Sharma, Sumit Dhariwal
With the advent of time series prediction in multidisciplinary domains, Multivariate Chaotic Time Series (MCTS) prediction has become a popular topic of re-search. Manifold applications like weather forecasting, stocks prediction, medical support, etc., deploy such kind prediction approach to predict the future of the time series based on past observations. In literature, various solutions have been explored and proposed to forecast future values in time series data. Significant efforts have been made to utilize various Neural Networks for time series prediction considering their applicability for future data prediction. However, a comprehensive evaluation of such existing methods is missing which demands attention for accurate and efficient prediction of time series data. In this paper, we have applied and evaluated various deep learning techniques on different dynamically generated data sets. Further, a comprehensive comparison of different techniques have been presented referencing loss observed with performance matrix Mean Absolute Error.
{"title":"Analysis of Multivariate Chaotic Time Series using Neural Networks","authors":"Avani Sharma, Sumit Dhariwal","doi":"10.1109/ASIANCON55314.2022.9909083","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9909083","url":null,"abstract":"With the advent of time series prediction in multidisciplinary domains, Multivariate Chaotic Time Series (MCTS) prediction has become a popular topic of re-search. Manifold applications like weather forecasting, stocks prediction, medical support, etc., deploy such kind prediction approach to predict the future of the time series based on past observations. In literature, various solutions have been explored and proposed to forecast future values in time series data. Significant efforts have been made to utilize various Neural Networks for time series prediction considering their applicability for future data prediction. However, a comprehensive evaluation of such existing methods is missing which demands attention for accurate and efficient prediction of time series data. In this paper, we have applied and evaluated various deep learning techniques on different dynamically generated data sets. Further, a comprehensive comparison of different techniques have been presented referencing loss observed with performance matrix Mean Absolute Error.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124774102","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}
Cryptocurrencies are becoming a well-known and commonly acknowledged kind of substitute trade money. Most monetary businesses now include cryptocurrency. Accordingly, cryptocurrency trading is widely regarded as the most of prevalent and capable types of lucrative investments. However, because this financial sector is already known for its extreme volatility and quick price changes, over brief periods of time. For such constantly changing nature of crypto trends and price, it has become a necessary part for traders and crypto enthusiast to get a detailed analysis before investing. Also, the construction of a precise and dependable forecasting model is regarded vital for portfolio management and optimization. In this paper we propose a web system, which will help to understand cryptocurrency in a more statistical way. Proposed system focuses mainly on four coins : Bitcoin, Ethereum, Dogecoin and Shiba Inu performing analysis and forecasting on all the four coins. System will also do statistical comparison between the coins. Analysis and comparison is carried out using python libraries and modules whereas LSTM and ARIMA are used for forecasting. Extensive research was conducted using real-time and historical information, on four key cryptocurrencies, two of which had the greatest market capitalization, notably Bitcoin and Ethereum, while the other, Dogecoin and Shiba Inu, that had a significant growth in market capitalization over the previous year. In comparison to old fully-connected deep neural networks, the suggested model may employ mixed crypto data more proficiently, minimizing overfitting and computing costs.
{"title":"Cryptocurrency Analysis and Forecasting","authors":"Payal Pagariya, Sadhvee Shinde, Rupali Shivpure, Sakshi Patil, Ashwini Jarali","doi":"10.1109/ASIANCON55314.2022.9909168","DOIUrl":"https://doi.org/10.1109/ASIANCON55314.2022.9909168","url":null,"abstract":"Cryptocurrencies are becoming a well-known and commonly acknowledged kind of substitute trade money. Most monetary businesses now include cryptocurrency. Accordingly, cryptocurrency trading is widely regarded as the most of prevalent and capable types of lucrative investments. However, because this financial sector is already known for its extreme volatility and quick price changes, over brief periods of time. For such constantly changing nature of crypto trends and price, it has become a necessary part for traders and crypto enthusiast to get a detailed analysis before investing. Also, the construction of a precise and dependable forecasting model is regarded vital for portfolio management and optimization. In this paper we propose a web system, which will help to understand cryptocurrency in a more statistical way. Proposed system focuses mainly on four coins : Bitcoin, Ethereum, Dogecoin and Shiba Inu performing analysis and forecasting on all the four coins. System will also do statistical comparison between the coins. Analysis and comparison is carried out using python libraries and modules whereas LSTM and ARIMA are used for forecasting. Extensive research was conducted using real-time and historical information, on four key cryptocurrencies, two of which had the greatest market capitalization, notably Bitcoin and Ethereum, while the other, Dogecoin and Shiba Inu, that had a significant growth in market capitalization over the previous year. In comparison to old fully-connected deep neural networks, the suggested model may employ mixed crypto data more proficiently, minimizing overfitting and computing costs.","PeriodicalId":429704,"journal":{"name":"2022 2nd Asian Conference on Innovation in Technology (ASIANCON)","volume":"167 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121806240","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：