Pub Date : 2020-10-28DOI: 10.1109/UEMCON51285.2020.9298079
Lanier A Watkins, Yue Yu, Sifan Li, W. H. Robinson, A. Rubin
In bring-your-own-device (BYOD) and guest wireless networks, the use of mobile devices within industry, government, and academic enterprise networks represents a difficult security challenge for system administrators. Devices not owned by the enterprise can pose additional risk. Our prior research demonstrated a dynamic anomaly detection method that used side-channel analysis of ping responses to infer whether devices were compromised. Initial results showed promise for a limited dataset. Our extension of this prior work now uses deep learning, twice as many features, and analyzes ten times more malware. Additional experiments demonstrate that our deep learning model generalizes to the detection of unseen threats across multiple families of malware.
{"title":"Using Deep Learning to Identify Security Risks of Personal Mobile Devices in Enterprise Networks","authors":"Lanier A Watkins, Yue Yu, Sifan Li, W. H. Robinson, A. Rubin","doi":"10.1109/UEMCON51285.2020.9298079","DOIUrl":"https://doi.org/10.1109/UEMCON51285.2020.9298079","url":null,"abstract":"In bring-your-own-device (BYOD) and guest wireless networks, the use of mobile devices within industry, government, and academic enterprise networks represents a difficult security challenge for system administrators. Devices not owned by the enterprise can pose additional risk. Our prior research demonstrated a dynamic anomaly detection method that used side-channel analysis of ping responses to infer whether devices were compromised. Initial results showed promise for a limited dataset. Our extension of this prior work now uses deep learning, twice as many features, and analyzes ten times more malware. Additional experiments demonstrate that our deep learning model generalizes to the detection of unseen threats across multiple families of malware.","PeriodicalId":433609,"journal":{"name":"2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130157448","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 : 2020-10-28DOI: 10.1109/UEMCON51285.2020.9298103
Amina Shrestha, Rhishav Mahaju, S. Kuruppu
Growing complexity of the computational algorithms in control systems invites prolonged calculation periods when executed in real-time. Feedback measurement signal path is a crucial signal path in closed-loop controlled systems, especially when requiring high bandwidth control. With high processing time, sampling of current signal tends to be delayed by multiple Pulse Width Modulation (PWM) periods depending on the interrupt priorities. This paper studies the effect of current feedback sampling delay effect on the overall stability of field oriented controlled (FOC) Permanent Magnet Synchronous Motor (PMSM) drive system. The delay effect study is done by first analyzing theoretical equation of system transfer function with the feedback sampling delay incorporated, followed by simulation results and finally by performing experiments on actual hardware. The impact of increasing sampling delay on the current feedback signal is presented with other controller design parameters.
{"title":"Feedback Path Delay Effect on Stability of Controlled PMSMs","authors":"Amina Shrestha, Rhishav Mahaju, S. Kuruppu","doi":"10.1109/UEMCON51285.2020.9298103","DOIUrl":"https://doi.org/10.1109/UEMCON51285.2020.9298103","url":null,"abstract":"Growing complexity of the computational algorithms in control systems invites prolonged calculation periods when executed in real-time. Feedback measurement signal path is a crucial signal path in closed-loop controlled systems, especially when requiring high bandwidth control. With high processing time, sampling of current signal tends to be delayed by multiple Pulse Width Modulation (PWM) periods depending on the interrupt priorities. This paper studies the effect of current feedback sampling delay effect on the overall stability of field oriented controlled (FOC) Permanent Magnet Synchronous Motor (PMSM) drive system. The delay effect study is done by first analyzing theoretical equation of system transfer function with the feedback sampling delay incorporated, followed by simulation results and finally by performing experiments on actual hardware. The impact of increasing sampling delay on the current feedback signal is presented with other controller design parameters.","PeriodicalId":433609,"journal":{"name":"2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129126631","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 : 2020-10-28DOI: 10.1109/UEMCON51285.2020.9298115
A. Hassebo, Mohamed A. Ali
Smart connected LED streetlights are emerging as an important infrastructure that can support basic lighting control services as well as a wide range of current and future smart city applications and services. Each smart streetlight is turned into multi-sensor-equipped smart node, a sensor ‘hub’ node, capable of capturing and transmitting/receiving real-time data (digitally controllable nodes). A smart LED has sensors embedded into and connectivity to the cloud. This paper assesses the feasibility and quantifies the performance of commercial point-to point (P2P) 4G LTE cellular networks when used to provide robust connectivity between a massive number of smart streetlight hub nodes and the cloud. Each smart streetlight hub node is assumed to be running simultaneously few basic lighting control services as well as smart city services and applications, including mission-critical with strict latency and reliability requirements, with particular emphasis on HD IP video surveillance cameras.
{"title":"Robust Cellular connectivity-Based Smart LED Street Lighting System: A Platform For Innovative Mission Critical Smart City IoT Applications","authors":"A. Hassebo, Mohamed A. Ali","doi":"10.1109/UEMCON51285.2020.9298115","DOIUrl":"https://doi.org/10.1109/UEMCON51285.2020.9298115","url":null,"abstract":"Smart connected LED streetlights are emerging as an important infrastructure that can support basic lighting control services as well as a wide range of current and future smart city applications and services. Each smart streetlight is turned into multi-sensor-equipped smart node, a sensor ‘hub’ node, capable of capturing and transmitting/receiving real-time data (digitally controllable nodes). A smart LED has sensors embedded into and connectivity to the cloud. This paper assesses the feasibility and quantifies the performance of commercial point-to point (P2P) 4G LTE cellular networks when used to provide robust connectivity between a massive number of smart streetlight hub nodes and the cloud. Each smart streetlight hub node is assumed to be running simultaneously few basic lighting control services as well as smart city services and applications, including mission-critical with strict latency and reliability requirements, with particular emphasis on HD IP video surveillance cameras.","PeriodicalId":433609,"journal":{"name":"2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129305613","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 : 2020-10-28DOI: 10.1109/UEMCON51285.2020.9298033
T. Kumar, Pradyumn Sharma, N. Prakash
Parkinson's Disease (PD) is a chronic degenerative disease that mainly affects the nervous system and motor controls in human beings. Early symptoms such as muscle stiffness, tremors, impaired balance and difficulty with walking are considerably less noticeable. Blood tests and Scans also do not provide sufficient evidence for early diagnosis. Hence it is very difficult for doctors to diagnose the onset of Parkinson's Disease. However, smearing of speech gives an early warning and can be effectively used for the prediction of PD. This paper, the voice recording samples of Parkinson’s disease affected and healthy patients have been used for PD prediction. Thirteen predictive models using various Machine Learning techniques have been formulated using the University of California, Irvine (UCI) dataset. A comparative study of these predictive models has been carried out on the UCI dataset consisting of biomedical voice recording samples of healthy and Parkinson’s Disease affected peoples. These predictive models have been trained and tested for their accuracy and efficiency. The performance analysis of the best five models has been presented in this paper, for accurate prediction of Parkinson's Disease at an early stage. The processing speed of these models has also been analysed, to assess their suitability for light weight mobile applications in the ubiquitous computing environment.
{"title":"Comparison of Machine learning models for Parkinson’s Disease prediction","authors":"T. Kumar, Pradyumn Sharma, N. Prakash","doi":"10.1109/UEMCON51285.2020.9298033","DOIUrl":"https://doi.org/10.1109/UEMCON51285.2020.9298033","url":null,"abstract":"Parkinson's Disease (PD) is a chronic degenerative disease that mainly affects the nervous system and motor controls in human beings. Early symptoms such as muscle stiffness, tremors, impaired balance and difficulty with walking are considerably less noticeable. Blood tests and Scans also do not provide sufficient evidence for early diagnosis. Hence it is very difficult for doctors to diagnose the onset of Parkinson's Disease. However, smearing of speech gives an early warning and can be effectively used for the prediction of PD. This paper, the voice recording samples of Parkinson’s disease affected and healthy patients have been used for PD prediction. Thirteen predictive models using various Machine Learning techniques have been formulated using the University of California, Irvine (UCI) dataset. A comparative study of these predictive models has been carried out on the UCI dataset consisting of biomedical voice recording samples of healthy and Parkinson’s Disease affected peoples. These predictive models have been trained and tested for their accuracy and efficiency. The performance analysis of the best five models has been presented in this paper, for accurate prediction of Parkinson's Disease at an early stage. The processing speed of these models has also been analysed, to assess their suitability for light weight mobile applications in the ubiquitous computing environment.","PeriodicalId":433609,"journal":{"name":"2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122803089","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 : 2020-10-28DOI: 10.1109/UEMCON51285.2020.9298141
B. Abegaz, Naxi Shah
The safety and security of self-driving vehicles is challenging considering large types of sensors and computational units developed by multiple manufacturers. In this paper, a method of identifying the optimal combination of sensors, methods and algorithms are presented for the safety of lane keeping and cruise control functionality of self-driving vehicles. Moreover, an adaptive model predictive control approach is presented that incorporates optimal number of sensors to improve the performance of such vehicles. Results indicate that the presented approach could improve the safety of lane keeping and cruise control functionality as compared to other approaches. This work could pave the way for the future smart and safe self-driving transportation systems.
{"title":"Sensors based Lane Keeping and Cruise Control of Self Driving Vehicles","authors":"B. Abegaz, Naxi Shah","doi":"10.1109/UEMCON51285.2020.9298141","DOIUrl":"https://doi.org/10.1109/UEMCON51285.2020.9298141","url":null,"abstract":"The safety and security of self-driving vehicles is challenging considering large types of sensors and computational units developed by multiple manufacturers. In this paper, a method of identifying the optimal combination of sensors, methods and algorithms are presented for the safety of lane keeping and cruise control functionality of self-driving vehicles. Moreover, an adaptive model predictive control approach is presented that incorporates optimal number of sensors to improve the performance of such vehicles. Results indicate that the presented approach could improve the safety of lane keeping and cruise control functionality as compared to other approaches. This work could pave the way for the future smart and safe self-driving transportation systems.","PeriodicalId":433609,"journal":{"name":"2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"141 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122913525","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 : 2020-10-28DOI: 10.1109/UEMCON51285.2020.9298114
Wei Pang, Yufeng Li, Shengli Lu
With the development of convolutional neural network (CNN), the accuracy of face recognition has been greatly improved. But the huge amount of weights and calculations hinders its implementation in portable devices. Designing hardware accelerator is an effective solution to the problem. In this paper, a face recognition algorithm is designed based on deep separable convolution. The weights and activations are quantified to 8 bits, reducing the requirement of data access and bandwidth. In addition, a generic CNN accelerator based on systolic array is designed and validated on Xilinx Zynq-XC7Z035 FPGA. The face recognition algorithm achieved an accuracy of 94.4% in the LFW dataset. The performance and power efficiency of the accelerator are 52.9 GOPS and 9.71GOPS/W at 100MHz, respectively. And the accelerator can process 160×160 face image at 25FPS.
{"title":"8-bit Convolutional Neural Network Accelerator for Face Recognition","authors":"Wei Pang, Yufeng Li, Shengli Lu","doi":"10.1109/UEMCON51285.2020.9298114","DOIUrl":"https://doi.org/10.1109/UEMCON51285.2020.9298114","url":null,"abstract":"With the development of convolutional neural network (CNN), the accuracy of face recognition has been greatly improved. But the huge amount of weights and calculations hinders its implementation in portable devices. Designing hardware accelerator is an effective solution to the problem. In this paper, a face recognition algorithm is designed based on deep separable convolution. The weights and activations are quantified to 8 bits, reducing the requirement of data access and bandwidth. In addition, a generic CNN accelerator based on systolic array is designed and validated on Xilinx Zynq-XC7Z035 FPGA. The face recognition algorithm achieved an accuracy of 94.4% in the LFW dataset. The performance and power efficiency of the accelerator are 52.9 GOPS and 9.71GOPS/W at 100MHz, respectively. And the accelerator can process 160×160 face image at 25FPS.","PeriodicalId":433609,"journal":{"name":"2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123851756","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 : 2020-10-28DOI: 10.1109/UEMCON51285.2020.9298184
J. Piou
One of the key advantages of deep learning over traditional automatic target recognition (ATR) is its features can be directly selected by the network and do not necessary need, like the ATR, the designer input parametric constraints to carry out its operation. However, this advantage has its pitfall; because the network considers too many features the possibility to develop Cramer-Rao lower bounds that take into account size of an input image, its key scattering centers and its signal-to-noise ratio (SNR), and also the depth, width, weight and bias matrices from different layers of the network, is limited. In this paper, state space matrices that capture the features of an input image, the state vector and output observation matrix that allow computation of the noise covariance matrices together with the network parameters and weight matrices are used to develop Cramer-Rao bounds from an input image that is fed to a multiple-layer deep learning network. The proposed bounds are computed from a 5-layer deep learning network that is trained and tested on MSTAR data collected at HH polarization by a 0.596 GHz radar bandwidth at fifteen- and seventeen-degree depression angles, respectively.
{"title":"Computation of Posterior Cramer-Rao Bounds for Deep Learning Networks","authors":"J. Piou","doi":"10.1109/UEMCON51285.2020.9298184","DOIUrl":"https://doi.org/10.1109/UEMCON51285.2020.9298184","url":null,"abstract":"One of the key advantages of deep learning over traditional automatic target recognition (ATR) is its features can be directly selected by the network and do not necessary need, like the ATR, the designer input parametric constraints to carry out its operation. However, this advantage has its pitfall; because the network considers too many features the possibility to develop Cramer-Rao lower bounds that take into account size of an input image, its key scattering centers and its signal-to-noise ratio (SNR), and also the depth, width, weight and bias matrices from different layers of the network, is limited. In this paper, state space matrices that capture the features of an input image, the state vector and output observation matrix that allow computation of the noise covariance matrices together with the network parameters and weight matrices are used to develop Cramer-Rao bounds from an input image that is fed to a multiple-layer deep learning network. The proposed bounds are computed from a 5-layer deep learning network that is trained and tested on MSTAR data collected at HH polarization by a 0.596 GHz radar bandwidth at fifteen- and seventeen-degree depression angles, respectively.","PeriodicalId":433609,"journal":{"name":"2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124010052","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 : 2020-10-28DOI: 10.1109/UEMCON51285.2020.9298093
Kyle Spurlock, H. Elgazzar
Today, social media has grown in usage to the point where it is often deeply intertwined with life offline. People share their thoughts, passions, and lives online, and in many ways, these social networks can be considered abstractions of real-world society. The idea for this research is that by modeling on these social networks, these glimpses into people's lives through their words and posts are capable of showing their current health situation, and their susceptibility to outside influences affecting it. The goal of this research project is to design and implement unsupervised machine learning techniques to group together sub-networks of connected individuals in hopes that it may be beneficial to current disease surveillance systems. Using Python programming language and the tools available to it, data was collected from the social network platform Twitter and analyzed using three clustering and centrality measurements. The criterion to be included in the data found tweets containing symptomatic keywords, like those of which experienced by people afflicted with the novel coronavirus disease (COVID-19). It is our findings in this research that by simulating the real-world connections that people have with their surrounding cliques using the ones that they exist within the virtual world, new possibilities for viral control and disease prevention become available using easily sourced, and quickly gatherable information.
{"title":"Predicting COVID-19 Infection Groups using Social Networks and Machine Learning Algorithms","authors":"Kyle Spurlock, H. Elgazzar","doi":"10.1109/UEMCON51285.2020.9298093","DOIUrl":"https://doi.org/10.1109/UEMCON51285.2020.9298093","url":null,"abstract":"Today, social media has grown in usage to the point where it is often deeply intertwined with life offline. People share their thoughts, passions, and lives online, and in many ways, these social networks can be considered abstractions of real-world society. The idea for this research is that by modeling on these social networks, these glimpses into people's lives through their words and posts are capable of showing their current health situation, and their susceptibility to outside influences affecting it. The goal of this research project is to design and implement unsupervised machine learning techniques to group together sub-networks of connected individuals in hopes that it may be beneficial to current disease surveillance systems. Using Python programming language and the tools available to it, data was collected from the social network platform Twitter and analyzed using three clustering and centrality measurements. The criterion to be included in the data found tweets containing symptomatic keywords, like those of which experienced by people afflicted with the novel coronavirus disease (COVID-19). It is our findings in this research that by simulating the real-world connections that people have with their surrounding cliques using the ones that they exist within the virtual world, new possibilities for viral control and disease prevention become available using easily sourced, and quickly gatherable information.","PeriodicalId":433609,"journal":{"name":"2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"131 8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124249652","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 : 2020-10-28DOI: 10.1109/UEMCON51285.2020.9298176
Omar Alruwaili, I. Kostanic, A. Al-Sabbagh, Hamad Almohamedh
Major problem facing urban areas today is air pollution. Gas emissions from cars are considered the most important source of this kind of pollution. Pollutant gases emitted as parts of car exhaust consist of chemicals such as carbon monoxide (CO), nitrogen dioxide (NO2), ozone (O3), particulate matter (PM), and Sulphur dioxide (SO2). The environmental Protection Agency (EPA) guides to measure these chemicals by several methods to calculate the gases’ concentration. An Internet of Things (IoT) device is used to monitor air quality in real-time is also described in this paper. It uses a set of sensors that measure air quality at the street level. This paper determined the relationship between traffic volume and the Air Quality Index (AQI) as defined by EPA guidelines. Multiple Linear Regression (MLR) is used to create a mathematical model for the relationship between traffic volume and the Air Quality Index (AQI). This model has been tested on one of the streets in the city of Melbourne, Florida.
{"title":"IoT Based: Air Quality Index and Traffic Volume Correlation","authors":"Omar Alruwaili, I. Kostanic, A. Al-Sabbagh, Hamad Almohamedh","doi":"10.1109/UEMCON51285.2020.9298176","DOIUrl":"https://doi.org/10.1109/UEMCON51285.2020.9298176","url":null,"abstract":"Major problem facing urban areas today is air pollution. Gas emissions from cars are considered the most important source of this kind of pollution. Pollutant gases emitted as parts of car exhaust consist of chemicals such as carbon monoxide (CO), nitrogen dioxide (NO2), ozone (O3), particulate matter (PM), and Sulphur dioxide (SO2). The environmental Protection Agency (EPA) guides to measure these chemicals by several methods to calculate the gases’ concentration. An Internet of Things (IoT) device is used to monitor air quality in real-time is also described in this paper. It uses a set of sensors that measure air quality at the street level. This paper determined the relationship between traffic volume and the Air Quality Index (AQI) as defined by EPA guidelines. Multiple Linear Regression (MLR) is used to create a mathematical model for the relationship between traffic volume and the Air Quality Index (AQI). This model has been tested on one of the streets in the city of Melbourne, Florida.","PeriodicalId":433609,"journal":{"name":"2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117277627","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 : 2020-10-28DOI: 10.1109/UEMCON51285.2020.9298169
Aaron Hunter, Konstantin Boyarinov
In order to develop effective ubiquitous computing systems, we often need to predict an agent’s behaviour based on past data. One way to do this is to maintain a model of what the agent believes at any point in time, as well as a mechanism for changing the beliefs to incorporate new information. In the knowledge representation community, this process is captured through formal belief revision operators. In this paper, we assume that we are monitoring the behaviour of an agent that uses a belief revision operator to incorporate new information; but we do not know exactly which operator is being used. Given past data about the beliefs of the agent, we propose two approaches for predicting future changes in belief. In the first approach, we simply search for all revision operators consistent with the data. In the second approach, we use machine learning to predict if a certain formula will be believed based on past data. We describe work in progress on prototype software to experiment with both approaches, and discuss when each is appropriate. We argue that modelling the dynamic beliefs of an agent in this way can be a useful component of a software system tasked with predicting behaviour when new information is received.
{"title":"On the Development of Tools for Modelling Dynamic Beliefs Based on Past Data","authors":"Aaron Hunter, Konstantin Boyarinov","doi":"10.1109/UEMCON51285.2020.9298169","DOIUrl":"https://doi.org/10.1109/UEMCON51285.2020.9298169","url":null,"abstract":"In order to develop effective ubiquitous computing systems, we often need to predict an agent’s behaviour based on past data. One way to do this is to maintain a model of what the agent believes at any point in time, as well as a mechanism for changing the beliefs to incorporate new information. In the knowledge representation community, this process is captured through formal belief revision operators. In this paper, we assume that we are monitoring the behaviour of an agent that uses a belief revision operator to incorporate new information; but we do not know exactly which operator is being used. Given past data about the beliefs of the agent, we propose two approaches for predicting future changes in belief. In the first approach, we simply search for all revision operators consistent with the data. In the second approach, we use machine learning to predict if a certain formula will be believed based on past data. We describe work in progress on prototype software to experiment with both approaches, and discuss when each is appropriate. We argue that modelling the dynamic beliefs of an agent in this way can be a useful component of a software system tasked with predicting behaviour when new information is received.","PeriodicalId":433609,"journal":{"name":"2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123355125","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：