Pub Date : 2020-12-01DOI: 10.1109/CSCI51800.2020.00097
Hyoung-Jin Oh, Donghoon Kim, Won-gyum Kim, Doosung Hwang
Tor (The Onion Router) ensures network anonymity by encrypting contents through multiple relay nodes. Recent studies on website fingerprinting (WF) showed that websites can be identified with high accuracy by analyzing traffic data. However, websites are changing over time by updating contents, which can significantly reduce the accuracy of WF attacks. This study analyzes the performance over time by using ensemble models with excellent WF attack performance. The experiment are conducted in two cases with the initial model. The not updated analyzes the accuracy of models made from initial data over time, whereas the updated adds data that has changed over time to update the model to analyzes the accuracy. The average accuracy of the initial ensemble models is over 90.0% and the Rotation Forest algorithm shows high performance of 93.5%. Comparing the models trained after 30 days with the initial model, the classification performance dropped in both cases; the not updated dropped by more than 30.0% and the updated dropped by about 10.0%. The experimental results suggest that WF using machine learning may require model learning on a regular basis.
Tor (The Onion Router)通过多个中继节点对内容进行加密，保证了网络的匿名性。近年来对网站指纹技术的研究表明，通过对流量数据的分析，可以较准确地识别出网站。然而，随着时间的推移，网站会不断更新内容，这大大降低了WF攻击的准确性。本研究通过使用具有优异WF攻击性能的集成模型来分析性能随时间的变化。用初始模型进行了两种情况下的实验。未更新的模型分析基于初始数据的模型随时间的准确性，而更新的模型添加随时间变化的数据来更新模型以分析准确性。初始集成模型的平均精度超过90.0%，旋转森林算法的性能达到93.5%。将30天后训练的模型与初始模型进行比较，两种情况下的分类性能都有所下降;未更新的降幅超过30.0%，更新的降幅约为10.0%。实验结果表明，使用机器学习的WF可能需要定期进行模型学习。
{"title":"Performance Analysis of Tor Website Fingerprinting over Time using Tree Ensemble Models","authors":"Hyoung-Jin Oh, Donghoon Kim, Won-gyum Kim, Doosung Hwang","doi":"10.1109/CSCI51800.2020.00097","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00097","url":null,"abstract":"Tor (The Onion Router) ensures network anonymity by encrypting contents through multiple relay nodes. Recent studies on website fingerprinting (WF) showed that websites can be identified with high accuracy by analyzing traffic data. However, websites are changing over time by updating contents, which can significantly reduce the accuracy of WF attacks. This study analyzes the performance over time by using ensemble models with excellent WF attack performance. The experiment are conducted in two cases with the initial model. The not updated analyzes the accuracy of models made from initial data over time, whereas the updated adds data that has changed over time to update the model to analyzes the accuracy. The average accuracy of the initial ensemble models is over 90.0% and the Rotation Forest algorithm shows high performance of 93.5%. Comparing the models trained after 30 days with the initial model, the classification performance dropped in both cases; the not updated dropped by more than 30.0% and the updated dropped by about 10.0%. The experimental results suggest that WF using machine learning may require model learning on a regular basis.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132201020","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00122
Marina Villalba Carballo, Byeong Kil Lee
Deep neural networks (DNNs) have several technical issues on computational complexity, redundancy, and the parameter size – especially when applied in embedded devices. Among those issues, lots of parameters require high memory capacity which causes migration problem to embedded devices. Many pruning techniques are proposed to reduce the network size in deep neural networks, but there are still various issues that exist for applying pruning techniques to DNNs. In this paper, we propose a simple-yet-efficient scheme, accuracy-aware structured pruning based on the characterization of each convolutional layer. We investigate the accuracy and compression rate of individual layer with a fixed pruning ratio and re-order the pruning priority depending on the accuracy of each layer. To achieve a further compression rate, we also add quantization to the linear layers. Our results show that the order of the layers pruned does affect the final accuracy of the deep neural network. Based on our experiments, the pruned AlexNet and VGG16 models’ parameter size is compressed up to 47.28x and 35.21x with less than 1% accuracy drop with respect to the original model.
{"title":"Accuracy-aware Structured Filter Pruning for Deep Neural Networks","authors":"Marina Villalba Carballo, Byeong Kil Lee","doi":"10.1109/CSCI51800.2020.00122","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00122","url":null,"abstract":"Deep neural networks (DNNs) have several technical issues on computational complexity, redundancy, and the parameter size – especially when applied in embedded devices. Among those issues, lots of parameters require high memory capacity which causes migration problem to embedded devices. Many pruning techniques are proposed to reduce the network size in deep neural networks, but there are still various issues that exist for applying pruning techniques to DNNs. In this paper, we propose a simple-yet-efficient scheme, accuracy-aware structured pruning based on the characterization of each convolutional layer. We investigate the accuracy and compression rate of individual layer with a fixed pruning ratio and re-order the pruning priority depending on the accuracy of each layer. To achieve a further compression rate, we also add quantization to the linear layers. Our results show that the order of the layers pruned does affect the final accuracy of the deep neural network. Based on our experiments, the pruned AlexNet and VGG16 models’ parameter size is compressed up to 47.28x and 35.21x with less than 1% accuracy drop with respect to the original model.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132858440","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00095
L. Ismail, Huned Materwala
Diabetes is one of the top 10 causes of death worldwide. Health professionals are aiming for machine learning models to support the prognosis of diabetes for better healthcare and to put in place an effective prevention plan. In this paper, we conduct a comparative analysis of the most used machine learning models in the literature to predict the prevalence of diabetes mellitus type 2. We evaluate the models in terms of accuracy, F-measure and execution time with and without feature selection using a real-life diabetes dataset. The detailed analysis is in the paper.
{"title":"Comparative Analysis of Machine Learning Models for Diabetes Mellitus Type 2 Prediction","authors":"L. Ismail, Huned Materwala","doi":"10.1109/CSCI51800.2020.00095","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00095","url":null,"abstract":"Diabetes is one of the top 10 causes of death worldwide. Health professionals are aiming for machine learning models to support the prognosis of diabetes for better healthcare and to put in place an effective prevention plan. In this paper, we conduct a comparative analysis of the most used machine learning models in the literature to predict the prevalence of diabetes mellitus type 2. We evaluate the models in terms of accuracy, F-measure and execution time with and without feature selection using a real-life diabetes dataset. The detailed analysis is in the paper.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"105 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122426806","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00054
Thanh-Huong Le
Text sentiment analysis is target-oriented, aiming to identify the opinion or attitude from a piece of natural language text toward topics or entities, whether it is negative, positive or neutral using natural language processing and computational methods. With the growth of the internet, numerous business websites have been deployed to support shopping products, booking services online as well as to allow online reviewing and commenting the services in forms of either business forums or social networks. Use of text sentiment analysis for automatically mining opinion from the feedbacks on such emerging internet platforms is not only useful for customers seeking for advice, but also necessary for business to study customers’ attitudes toward brands, products, services, or events, and has become an increasingly dominant trend in business strategic management. Current state-of-the-art approaches for text sentiment analysis include lexicon based and machine learning based methods. In this research, we proposed a method that utilizes deep learning with attention word embedding. We showed that our method outperformed popular lexicon and embedding based methods.
{"title":"An attention-based deep learning method for text sentiment analysis","authors":"Thanh-Huong Le","doi":"10.1109/CSCI51800.2020.00054","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00054","url":null,"abstract":"Text sentiment analysis is target-oriented, aiming to identify the opinion or attitude from a piece of natural language text toward topics or entities, whether it is negative, positive or neutral using natural language processing and computational methods. With the growth of the internet, numerous business websites have been deployed to support shopping products, booking services online as well as to allow online reviewing and commenting the services in forms of either business forums or social networks. Use of text sentiment analysis for automatically mining opinion from the feedbacks on such emerging internet platforms is not only useful for customers seeking for advice, but also necessary for business to study customers’ attitudes toward brands, products, services, or events, and has become an increasingly dominant trend in business strategic management. Current state-of-the-art approaches for text sentiment analysis include lexicon based and machine learning based methods. In this research, we proposed a method that utilizes deep learning with attention word embedding. We showed that our method outperformed popular lexicon and embedding based methods.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116908389","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00309
XU Yingheng, Zhong Yueqi
Instance segmentation of clothing images is a task that fashion analysts are paying more and more attention to in recent years. The segmentation of different clothing components allows designers to better design new fashion items and allows consumers to better understand design concepts. Most of the current methods are based on deep convolutional neural networks (DCNN). However, most of the current instance segmentation neural networks are limited by the size of the receptive field and cannot capture the global dependence. For clothing images, the use of contextual information between different clothing and collocations can obtain fine-grained and higher clothing segmentation images. Previous studies have shown that attention-based methods can obtain non-local dependencies of the whole image and are mostly used for panoramic segmentation of aerial images. For instance segmentation of clothing images, we propose a new dual-branch attention module based on the Non-local attention mechanism, called Multiple Attention MaskRCNN (HAMaskRCNN). Specifically, for the attention module, we use two branches: position attention and channel attention. After feature fusion, the FPN module and the attention module are connected in parallel to form a multiple attention module. We use the Imaterialist-fashion (2019) dataset to conduct experiments and compare with the benchmark to prove the effectiveness of our HAMaskRCNN.
{"title":"Multiple Attention Mechanism Neural Network in Garment Image Segmentation","authors":"XU Yingheng, Zhong Yueqi","doi":"10.1109/CSCI51800.2020.00309","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00309","url":null,"abstract":"Instance segmentation of clothing images is a task that fashion analysts are paying more and more attention to in recent years. The segmentation of different clothing components allows designers to better design new fashion items and allows consumers to better understand design concepts. Most of the current methods are based on deep convolutional neural networks (DCNN). However, most of the current instance segmentation neural networks are limited by the size of the receptive field and cannot capture the global dependence. For clothing images, the use of contextual information between different clothing and collocations can obtain fine-grained and higher clothing segmentation images. Previous studies have shown that attention-based methods can obtain non-local dependencies of the whole image and are mostly used for panoramic segmentation of aerial images. For instance segmentation of clothing images, we propose a new dual-branch attention module based on the Non-local attention mechanism, called Multiple Attention MaskRCNN (HAMaskRCNN). Specifically, for the attention module, we use two branches: position attention and channel attention. After feature fusion, the FPN module and the attention module are connected in parallel to form a multiple attention module. We use the Imaterialist-fashion (2019) dataset to conduct experiments and compare with the benchmark to prove the effectiveness of our HAMaskRCNN.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114932811","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00104
Sjsu Scholarworks, Unnathi Bhandary, Mike Wu, Samuel Chen
With the progression of the Internet and social media, people are given multiple platforms to share their thoughts and opinions about various subject matters freely. However, this freedom of speech is misused to direct hate towards individuals or group of people due to their race, religion, gender etc. The rise of hate speech has led to conflicts and cases of cyber bullying, causing many organizations to look for optimal solutions to solve this problem. Developments in the field of machine learning and deep learning have piqued the interest of researchers, leading them to research and implement solutions to solve the problem of hate speech. Currently, machine learning techniques are applied to textual data to detect hate speech. With the ample use of video sharing sites, there is a need to find a way to detect hate speech in videos. This research deals with classification of videos into normal or hateful categories based on the spoken content of the videos. The video dataset is built using a crawler to search and download videos based on offensive words that are specified as keywords. The audio is extracted from the videos and is converted into textual format using a Speech-to-Text converter to obtain a transcript of the videos. Experiments are conducted by training four models with three different feature sets extracted from the dataset. The models are evaluated by computing the specified evaluation metrics. The evaluated metrics indicate that Random Forrest Classifier model delivers the best results in classifying videos.
{"title":"Detection of Hate Speech in Videos Using Machine Learning","authors":"Sjsu Scholarworks, Unnathi Bhandary, Mike Wu, Samuel Chen","doi":"10.1109/CSCI51800.2020.00104","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00104","url":null,"abstract":"With the progression of the Internet and social media, people are given multiple platforms to share their thoughts and opinions about various subject matters freely. However, this freedom of speech is misused to direct hate towards individuals or group of people due to their race, religion, gender etc. The rise of hate speech has led to conflicts and cases of cyber bullying, causing many organizations to look for optimal solutions to solve this problem. Developments in the field of machine learning and deep learning have piqued the interest of researchers, leading them to research and implement solutions to solve the problem of hate speech. Currently, machine learning techniques are applied to textual data to detect hate speech. With the ample use of video sharing sites, there is a need to find a way to detect hate speech in videos. This research deals with classification of videos into normal or hateful categories based on the spoken content of the videos. The video dataset is built using a crawler to search and download videos based on offensive words that are specified as keywords. The audio is extracted from the videos and is converted into textual format using a Speech-to-Text converter to obtain a transcript of the videos. Experiments are conducted by training four models with three different feature sets extracted from the dataset. The models are evaluated by computing the specified evaluation metrics. The evaluated metrics indicate that Random Forrest Classifier model delivers the best results in classifying videos.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115980333","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00220
A. Martínez-Rebollar, Hugo Estrada-Esquivel, L. López-García, Leon Torres-Restrepo, J. Ortiz-Hernandez
The Internet of Things is promoting the generation of smart buildings. These buildings have as main requirement the navigation of interiors. However, GPS technology, which is used to carry out positioning, cannot be used within a building because satellite signals do not travel well through solid materials. In this paper, we present an indoor navigation proposal, which uses beacons technology and smartphones. Our software application obtains information from the context and generates the best route to reach the destination within an intelligent building. The Dijkstra algorithm was used to process all the information. Hence, our proposal aims to combine different technologies, and adapt developed algorithms to indoor navigation. The results obtained are encouraging and show that it is possible to obtain good results using this type of technology.
{"title":"Generating Indoor Navigation Routes Using Beacons","authors":"A. Martínez-Rebollar, Hugo Estrada-Esquivel, L. López-García, Leon Torres-Restrepo, J. Ortiz-Hernandez","doi":"10.1109/CSCI51800.2020.00220","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00220","url":null,"abstract":"The Internet of Things is promoting the generation of smart buildings. These buildings have as main requirement the navigation of interiors. However, GPS technology, which is used to carry out positioning, cannot be used within a building because satellite signals do not travel well through solid materials. In this paper, we present an indoor navigation proposal, which uses beacons technology and smartphones. Our software application obtains information from the context and generates the best route to reach the destination within an intelligent building. The Dijkstra algorithm was used to process all the information. Hence, our proposal aims to combine different technologies, and adapt developed algorithms to indoor navigation. The results obtained are encouraging and show that it is possible to obtain good results using this type of technology.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114789522","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00146
Chen-Yeou Yu, Carl K. Chang, Wensheng Zhang
The growth of telecommunication fraud has caused tremendous loss to end users. In particular, new technologies such as robocalling systems have been a new resource of harassment. Traditional approaches in detecting such activities simply rely on the construction of blacklisting number systems. However, criminals can easily masquerade their phone numbers simply by changing their numbers through VoIP (Voice over IP) or use virtual mobile numbers (VMN) with relatively low pricing, laxed ID checks and high-level API automation. In this paper, we present a novel situation-enabled approach to blacklist unwanted phone numbers while keeping high detection rate through distributed crowd sourcing. The system consists of two parts. First, we collect a user’s daily schedule in time series as situational data and use the data to train Long Short Term Memory (LSTM) deep learning model. This model is used to predict the user’s situation in the future. Then, we implement a semi-automatic tagging application to tag each incoming call by reading the call history against the predicted situation. An incoming phone number can be automatically tagged as malicious if it is in a wrong situation or could be benign otherwise. A user is also allowed to manually change the tagging afterwards if it is necessary. Second, a distributed crowdsourcing is used to aggregate highly ranked calling numbers from different devices in the same area. When a higher-level blacklist has been built, it can be used to update local ones by propagating back to end user devices with edge local blacklist and edge foreign blacklist. A simple evaluation has been made against real incoming calls on Android phones. The results show that our system design can attain decent detection rates.
{"title":"An Edge Computing Based Situation Enabled Crowdsourcing Blacklisting System for Efficient Identification of Scammer Phone Numbers","authors":"Chen-Yeou Yu, Carl K. Chang, Wensheng Zhang","doi":"10.1109/CSCI51800.2020.00146","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00146","url":null,"abstract":"The growth of telecommunication fraud has caused tremendous loss to end users. In particular, new technologies such as robocalling systems have been a new resource of harassment. Traditional approaches in detecting such activities simply rely on the construction of blacklisting number systems. However, criminals can easily masquerade their phone numbers simply by changing their numbers through VoIP (Voice over IP) or use virtual mobile numbers (VMN) with relatively low pricing, laxed ID checks and high-level API automation. In this paper, we present a novel situation-enabled approach to blacklist unwanted phone numbers while keeping high detection rate through distributed crowd sourcing. The system consists of two parts. First, we collect a user’s daily schedule in time series as situational data and use the data to train Long Short Term Memory (LSTM) deep learning model. This model is used to predict the user’s situation in the future. Then, we implement a semi-automatic tagging application to tag each incoming call by reading the call history against the predicted situation. An incoming phone number can be automatically tagged as malicious if it is in a wrong situation or could be benign otherwise. A user is also allowed to manually change the tagging afterwards if it is necessary. Second, a distributed crowdsourcing is used to aggregate highly ranked calling numbers from different devices in the same area. When a higher-level blacklist has been built, it can be used to update local ones by propagating back to end user devices with edge local blacklist and edge foreign blacklist. A simple evaluation has been made against real incoming calls on Android phones. The results show that our system design can attain decent detection rates.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116793455","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00053
L. Ramamoorthi, Gabrielle Peko, D. Sundaram
In today’s digital world mobile phones have a significant impact on our day-to-day lives including the use of internet chat applications designed for smartphone users. Generally, these mobile messaging apps claim they protect the user’s information using encryption techniques. Yet, information security attacks that exploit the apps’ vulnerabilities are increasingly common. These vulnerabilities are the main gateway for hackers to access information. Considering the four most popular messaging apps, a taxonomy of attack targets of messaging applications is introduced that consists of three broad categories of attacks. Each of these categories is discussed and analyzed in order to propose several combinations of technological and procedural solutions to mitigate the vulnerabilities. Further, it is envisioned that these solutions provide the foundation for building prevention and protection mechanisms against such attacks.
{"title":"Information Security Attacks on Mobile Messaging Applications: Procedural and Technological Responses","authors":"L. Ramamoorthi, Gabrielle Peko, D. Sundaram","doi":"10.1109/CSCI51800.2020.00053","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00053","url":null,"abstract":"In today’s digital world mobile phones have a significant impact on our day-to-day lives including the use of internet chat applications designed for smartphone users. Generally, these mobile messaging apps claim they protect the user’s information using encryption techniques. Yet, information security attacks that exploit the apps’ vulnerabilities are increasingly common. These vulnerabilities are the main gateway for hackers to access information. Considering the four most popular messaging apps, a taxonomy of attack targets of messaging applications is introduced that consists of three broad categories of attacks. Each of these categories is discussed and analyzed in order to propose several combinations of technological and procedural solutions to mitigate the vulnerabilities. Further, it is envisioned that these solutions provide the foundation for building prevention and protection mechanisms against such attacks.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116370442","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00102
Daniel A. Sanaguano, José Lucio-Naranjo, R. Tenenbaum
This work aims to give an overview of Artificial Neural Networks (ANN) approaches applied for BIRs generation published in the literature and to expose gaps in the academic research. The literature review shows that several successful studies are using ANNs approaches for BIRs generation with a reduction in the computational effort by up to 90% with respect to the Traditional Method. Nevertheless, these approaches are bounded by a fixed pair of a sound-source and binaural-receptor, meaning that they do not take into account dynamic variations in the position of the receptor. In this sense, this work also introduces a conceptual model for a real-time BIRs generator that considers a moving binaural-receptor using a set of Artificial Neural Networks.
{"title":"A Conceptual Model for real-time Binaural-Room Impulse Responses generation using ANNs in Virtual Environments: State of the Art","authors":"Daniel A. Sanaguano, José Lucio-Naranjo, R. Tenenbaum","doi":"10.1109/CSCI51800.2020.00102","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00102","url":null,"abstract":"This work aims to give an overview of Artificial Neural Networks (ANN) approaches applied for BIRs generation published in the literature and to expose gaps in the academic research. The literature review shows that several successful studies are using ANNs approaches for BIRs generation with a reduction in the computational effort by up to 90% with respect to the Traditional Method. Nevertheless, these approaches are bounded by a fixed pair of a sound-source and binaural-receptor, meaning that they do not take into account dynamic variations in the position of the receptor. In this sense, this work also introduces a conceptual model for a real-time BIRs generator that considers a moving binaural-receptor using a set of Artificial Neural Networks.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"132 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114862330","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：