Gloria Bordogna, Luca Frigerio, A. Cuzzocrea, G. Psaila
Nowadays a great deal of attention is devoted to the issue of supporting big data analytics over big mobile social data. These data are generated by modern emerging social systems like Twitter, Facebook, Instagram, and so forth. Mining big mobile social data has been of great interest, as analyzing such data is critical for a wide spectrum of big data applications (e.g., smart cities). Among several proposals, clustering is a well-known solution for extracting interesting and actionable knowledge from massive amounts of big mobile (geo-located) social data. Inspired by this main thesis, this paper proposes an effective and efficient similarity-matrix-based algorithm for clustering big mobile social data, called TourMiner, which is specifically targeted to clustering trips extracted from tweets, in order to mine most popular tours. The main characteristic of TourMiner consists in applying clustering over a well-suited similarity matrix computed on top of trips. A comprehensive experimental assessment and analysis over Twitter data finally comfirms the benefits coming from our proposal.
{"title":"An Effective and Efficient Similarity-Matrix-Based Algorithm for Clustering Big Mobile Social Data","authors":"Gloria Bordogna, Luca Frigerio, A. Cuzzocrea, G. Psaila","doi":"10.1109/ICMLA.2016.0091","DOIUrl":"https://doi.org/10.1109/ICMLA.2016.0091","url":null,"abstract":"Nowadays a great deal of attention is devoted to the issue of supporting big data analytics over big mobile social data. These data are generated by modern emerging social systems like Twitter, Facebook, Instagram, and so forth. Mining big mobile social data has been of great interest, as analyzing such data is critical for a wide spectrum of big data applications (e.g., smart cities). Among several proposals, clustering is a well-known solution for extracting interesting and actionable knowledge from massive amounts of big mobile (geo-located) social data. Inspired by this main thesis, this paper proposes an effective and efficient similarity-matrix-based algorithm for clustering big mobile social data, called TourMiner, which is specifically targeted to clustering trips extracted from tweets, in order to mine most popular tours. The main characteristic of TourMiner consists in applying clustering over a well-suited similarity matrix computed on top of trips. A comprehensive experimental assessment and analysis over Twitter data finally comfirms the benefits coming from our proposal.","PeriodicalId":356182,"journal":{"name":"2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"233 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127669438","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}
W. Alghamdi, D. Stamate, K. Vang, D. Ståhl, M. Colizzi, G. Tripoli, D. Quattrone, O. Ajnakina, R. Murray, M. Forti
Over the last two decades, a significant body of research has established a link between cannabis use and psychotic outcomes. In this study, we aim to propose a novel symbiotic machine learning and statistical approach to pattern detection and to developing predictive models for the onset of first-episode psychosis. The data used has been gathered from real cases in cooperation with a medical research institution, and comprises a wide set of variables including demographic, drug-related, as well as several variables specifically related to the cannabis use. Our approach is built upon several machine learning techniques whose predictive models have been optimised in a computationally intensive framework. The ability of these models to predict first-episode psychosis has been extensively tested through large scale Monte Carlo simulations. Our results show that Boosted Classification Trees outperform other models in this context, and have significant predictive ability despite a large number of missing values in the data. Furthermore, we extended our approach by further investigating how different patterns of cannabis use relate to new cases of psychosis, via association analysis and bayesian techniques.
{"title":"A Prediction Modelling and Pattern Detection Approach for the First-Episode Psychosis Associated to Cannabis Use","authors":"W. Alghamdi, D. Stamate, K. Vang, D. Ståhl, M. Colizzi, G. Tripoli, D. Quattrone, O. Ajnakina, R. Murray, M. Forti","doi":"10.1109/ICMLA.2016.0148","DOIUrl":"https://doi.org/10.1109/ICMLA.2016.0148","url":null,"abstract":"Over the last two decades, a significant body of research has established a link between cannabis use and psychotic outcomes. In this study, we aim to propose a novel symbiotic machine learning and statistical approach to pattern detection and to developing predictive models for the onset of first-episode psychosis. The data used has been gathered from real cases in cooperation with a medical research institution, and comprises a wide set of variables including demographic, drug-related, as well as several variables specifically related to the cannabis use. Our approach is built upon several machine learning techniques whose predictive models have been optimised in a computationally intensive framework. The ability of these models to predict first-episode psychosis has been extensively tested through large scale Monte Carlo simulations. Our results show that Boosted Classification Trees outperform other models in this context, and have significant predictive ability despite a large number of missing values in the data. Furthermore, we extended our approach by further investigating how different patterns of cannabis use relate to new cases of psychosis, via association analysis and bayesian techniques.","PeriodicalId":356182,"journal":{"name":"2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122797076","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}
An important task for training a robot (virtual or real) is to estimate state. State includes the state of the robot and its environment. Images from digital cameras are commonly used to monitor the robot due to the rich information, and low-cost hardware. Neural networks excel at catagorizing images, and should prove powerful to estimate the state of the robot from these images. There are many problems that occur when attempting to estimate state with neural networks, including high resolution of images, training time, vanishing gradient, and more. This paper presents several practical techniques for facilitating state estimation from images with neural networks.
{"title":"Practical Techniques for Using Neural Networks to Estimate State from Images","authors":"Stephen C. Ashmore, Michael S. Gashler","doi":"10.1109/ICMLA.2016.0164","DOIUrl":"https://doi.org/10.1109/ICMLA.2016.0164","url":null,"abstract":"An important task for training a robot (virtual or real) is to estimate state. State includes the state of the robot and its environment. Images from digital cameras are commonly used to monitor the robot due to the rich information, and low-cost hardware. Neural networks excel at catagorizing images, and should prove powerful to estimate the state of the robot from these images. There are many problems that occur when attempting to estimate state with neural networks, including high resolution of images, training time, vanishing gradient, and more. This paper presents several practical techniques for facilitating state estimation from images with neural networks.","PeriodicalId":356182,"journal":{"name":"2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124883727","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}
Evolving networks, which are composed of objects and relationships that change over time, are prevalent in many real-world domains and have become an significant research topic in recent years. Most of the previous link prediction studies neglect the evolution of the network over time and mainly focus on the predicting the future links based on a static features of nodes and links. However, real-world networks have complex dynamic structures and non-linear varying topological features, which means that both nodes and links of the networks may appear or disappear. These dynamicity of the networks make link prediction a more challenging task. To overcome these difficulties, link prediction in such networks must model nonlinear temporal evolution of the topological features and link occurrences information of the network structure simultaneously. In this article, we propose a novel link prediction method based on NARX Neural Network for evolving networks. Our model first calculates similarity scores based on quasi-local measures for each pair of nodes in different snapshots of the network and create time series for each pair. Then, NARX network is effectively applied to prediction of the future node similarity scores by using past node similarities and node connectivities. The proposed method is tested on DBLP coauthorship networks. It is shown that combining time information with node similarities and node connectivities improves the link prediction performance to a large extent.
{"title":"Temporal Link Prediction Using Time Series of Quasi-Local Node Similarity Measures","authors":"Alper Ozcan, Ş. Öğüdücü","doi":"10.1109/ICMLA.2016.0068","DOIUrl":"https://doi.org/10.1109/ICMLA.2016.0068","url":null,"abstract":"Evolving networks, which are composed of objects and relationships that change over time, are prevalent in many real-world domains and have become an significant research topic in recent years. Most of the previous link prediction studies neglect the evolution of the network over time and mainly focus on the predicting the future links based on a static features of nodes and links. However, real-world networks have complex dynamic structures and non-linear varying topological features, which means that both nodes and links of the networks may appear or disappear. These dynamicity of the networks make link prediction a more challenging task. To overcome these difficulties, link prediction in such networks must model nonlinear temporal evolution of the topological features and link occurrences information of the network structure simultaneously. In this article, we propose a novel link prediction method based on NARX Neural Network for evolving networks. Our model first calculates similarity scores based on quasi-local measures for each pair of nodes in different snapshots of the network and create time series for each pair. Then, NARX network is effectively applied to prediction of the future node similarity scores by using past node similarities and node connectivities. The proposed method is tested on DBLP coauthorship networks. It is shown that combining time information with node similarities and node connectivities improves the link prediction performance to a large extent.","PeriodicalId":356182,"journal":{"name":"2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123417401","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}
Seismic data acquisition in remote locations involves sampling using regular grids of receivers in a field. Extracting the maximum possible information from fewer measurements is cost-effective and often necessary due to malfunctions or terrain limitations. Compressive Sensing (CS) is an emerging framework that allows reconstruction of sparse signals from fewer measurements than conventional sampling rates. In seismic CS, the utilization of sparse solvers has proven to be successful, however, algorithms lack predictive uncertainties. We apply the Relevance Vector Machine (RVM) to seismic CS and propose a novel utilization of multi-scale dictionaries of basis functions that capture different variations in the data. Furthermore, we propose the use of a new predictive uncertainty measure using the information from the neighbours of each estimation to produce accurate uncertainty maps. We apply the RVM to different seismic signals and obtain state-of-the-art reconstruction accuracy. Using the RVM and its predictive uncertainty map, it is possible to quantify risk associated with seismic data acquisition and at the same time guide future survey design.
{"title":"Relevance Vector Machines with Uncertainty Measure for Seismic Bayesian Compressive Sensing and Survey Design","authors":"G. Pilikos, Anita C. Faul","doi":"10.1109/ICMLA.2016.0166","DOIUrl":"https://doi.org/10.1109/ICMLA.2016.0166","url":null,"abstract":"Seismic data acquisition in remote locations involves sampling using regular grids of receivers in a field. Extracting the maximum possible information from fewer measurements is cost-effective and often necessary due to malfunctions or terrain limitations. Compressive Sensing (CS) is an emerging framework that allows reconstruction of sparse signals from fewer measurements than conventional sampling rates. In seismic CS, the utilization of sparse solvers has proven to be successful, however, algorithms lack predictive uncertainties. We apply the Relevance Vector Machine (RVM) to seismic CS and propose a novel utilization of multi-scale dictionaries of basis functions that capture different variations in the data. Furthermore, we propose the use of a new predictive uncertainty measure using the information from the neighbours of each estimation to produce accurate uncertainty maps. We apply the RVM to different seismic signals and obtain state-of-the-art reconstruction accuracy. Using the RVM and its predictive uncertainty map, it is possible to quantify risk associated with seismic data acquisition and at the same time guide future survey design.","PeriodicalId":356182,"journal":{"name":"2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123763660","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}
In the past twenty years, progress in intrusion detection has been steady but slow. The biggest challenge is to detect new attacks in real time. In this work, a deep learning approach for anomaly detection using a Restricted Boltzmann Machine (RBM) and a deep belief network are implemented. Our method uses a one-hidden layer RBM to perform unsupervised feature reduction. The resultant weights from this RBM are passed to another RBM producing a deep belief network. The pre-trained weights are passed into a fine tuning layer consisting of a Logistic Regression (LR) classifier with multi-class soft-max. We have implemented the deep learning architecture in C++ in Microsoft Visual Studio 2013 and we use the DARPA KDDCUP'99 dataset to evaluate its performance. Our architecture outperforms previous deep learning methods implemented by Li and Salama in both detection speed and accuracy. We achieve a detection rate of 97.9% on the total 10% KDDCUP'99 test dataset. By improving the training process of the simulation, we are also able to produce a low false negative rate of 2.47%. Although the deficiencies in the KDDCUP'99 dataset are well understood, it still presents machine learning approaches for predicting attacks with a reasonable challenge. Our future work will include applying our machine learning strategy to larger and more challenging datasets, which include larger classes of attacks.
在过去的二十年里，入侵检测技术的发展是稳定而缓慢的。最大的挑战是实时检测新的攻击。在这项工作中，实现了一种基于受限玻尔兹曼机(RBM)和深度信念网络的深度学习异常检测方法。我们的方法使用一个单隐藏层RBM来执行无监督特征约简。从这个RBM得到的结果权重被传递到另一个RBM，产生一个深度的信念网络。将预训练的权重传递到一个微调层，该微调层由一个具有多类soft-max的逻辑回归(LR)分类器组成。我们在Microsoft Visual Studio 2013中使用c++实现了深度学习架构，并使用了DARPA KDDCUP'99数据集来评估其性能。我们的架构在检测速度和准确性方面都优于Li和Salama之前实现的深度学习方法。我们在总共10%的KDDCUP'99测试数据集上实现了97.9%的检测率。通过改进仿真的训练过程，我们也能够产生2.47%的低假阴性率。虽然KDDCUP'99数据集的不足之处已经得到了很好的理解，但它仍然提供了机器学习方法来预测具有合理挑战的攻击。我们未来的工作将包括将我们的机器学习策略应用于更大、更具挑战性的数据集，其中包括更大类别的攻击。
{"title":"Toward an Online Anomaly Intrusion Detection System Based on Deep Learning","authors":"Khaled Alrawashdeh, C. Purdy","doi":"10.1109/ICMLA.2016.0040","DOIUrl":"https://doi.org/10.1109/ICMLA.2016.0040","url":null,"abstract":"In the past twenty years, progress in intrusion detection has been steady but slow. The biggest challenge is to detect new attacks in real time. In this work, a deep learning approach for anomaly detection using a Restricted Boltzmann Machine (RBM) and a deep belief network are implemented. Our method uses a one-hidden layer RBM to perform unsupervised feature reduction. The resultant weights from this RBM are passed to another RBM producing a deep belief network. The pre-trained weights are passed into a fine tuning layer consisting of a Logistic Regression (LR) classifier with multi-class soft-max. We have implemented the deep learning architecture in C++ in Microsoft Visual Studio 2013 and we use the DARPA KDDCUP'99 dataset to evaluate its performance. Our architecture outperforms previous deep learning methods implemented by Li and Salama in both detection speed and accuracy. We achieve a detection rate of 97.9% on the total 10% KDDCUP'99 test dataset. By improving the training process of the simulation, we are also able to produce a low false negative rate of 2.47%. Although the deficiencies in the KDDCUP'99 dataset are well understood, it still presents machine learning approaches for predicting attacks with a reasonable challenge. Our future work will include applying our machine learning strategy to larger and more challenging datasets, which include larger classes of attacks.","PeriodicalId":356182,"journal":{"name":"2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125512186","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}
License Plate Recognition System (LPRS) plays a vital role in smart city initiatives such as traffic control, smart parking, toll management and security. In this article, a cloud-based LPRS is addressed in the context of efficiency where accuracy and speed of processing plays a critical role towards its success. Signature-based features technique as a deep convolutional neural network in a cloud platform is proposed for plate localization, character detection and segmentation. Extracting significant features makes the LPRS to adequately recognize the license plate in a challenging situation such as i) congested traffic with multiple plates in the image ii) plate orientation towards brightness, iii) extra information on the plate, iv) distortion due to wear and tear and v) distortion about captured images in bad weather like as hazy images. Furthermore, the deep learning algorithm computed using bare-metal cloud servers with kernels optimized for NVIDIA GPUs, which speed up the training phase of the CNN LPDS algorithm. The experiments and results show the superiority of the performance in both recall and precision and accuracy in comparison with traditional LP detecting systems.
{"title":"A Next-Generation Secure Cloud-Based Deep Learning License Plate Recognition for Smart Cities","authors":"Rohith Polishetty, M. Roopaei, P. Rad","doi":"10.1109/ICMLA.2016.0054","DOIUrl":"https://doi.org/10.1109/ICMLA.2016.0054","url":null,"abstract":"License Plate Recognition System (LPRS) plays a vital role in smart city initiatives such as traffic control, smart parking, toll management and security. In this article, a cloud-based LPRS is addressed in the context of efficiency where accuracy and speed of processing plays a critical role towards its success. Signature-based features technique as a deep convolutional neural network in a cloud platform is proposed for plate localization, character detection and segmentation. Extracting significant features makes the LPRS to adequately recognize the license plate in a challenging situation such as i) congested traffic with multiple plates in the image ii) plate orientation towards brightness, iii) extra information on the plate, iv) distortion due to wear and tear and v) distortion about captured images in bad weather like as hazy images. Furthermore, the deep learning algorithm computed using bare-metal cloud servers with kernels optimized for NVIDIA GPUs, which speed up the training phase of the CNN LPDS algorithm. The experiments and results show the superiority of the performance in both recall and precision and accuracy in comparison with traditional LP detecting systems.","PeriodicalId":356182,"journal":{"name":"2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126609607","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}
Bilgehan Arslan, Ezgi Yorulmaz, Burcin Akca, Ş. Sağiroğlu
Biometric systems may be used to create a remote access model on devices, ensure personal data protection, personalize and facilitate the access security. Biometric systems are generally used to increase the security level in addition to the previous authentication methods and they seen as a good solution. Biometry occupies an important place between the areas of daily life of the machine learning. In this study;the techniques, methods, technologies used in biometric systems are researched, machine learning techniques used biometric aplications are investigated for the security perspective, the advantages and disadvantages that these tecniques provide are given. The studies in the literature between 2010-2016 years, used algorithms, technologies, metrics, usage areas, the machine learning techniques used for different biometric systems such as face, palm prints, iris, voice, fingerprint recognition are researched and the studies made are evaluated. The level of security provided by the use of biometric systems by developed using machine learning and disadvantages that arise in the use of these systems are stated in detail in the study. Also, impact on people of biometric methods in terms of ease of use, security and usages areas are examined.
{"title":"Security Perspective of Biometric Recognition and Machine Learning Techniques","authors":"Bilgehan Arslan, Ezgi Yorulmaz, Burcin Akca, Ş. Sağiroğlu","doi":"10.1109/ICMLA.2016.0087","DOIUrl":"https://doi.org/10.1109/ICMLA.2016.0087","url":null,"abstract":"Biometric systems may be used to create a remote access model on devices, ensure personal data protection, personalize and facilitate the access security. Biometric systems are generally used to increase the security level in addition to the previous authentication methods and they seen as a good solution. Biometry occupies an important place between the areas of daily life of the machine learning. In this study;the techniques, methods, technologies used in biometric systems are researched, machine learning techniques used biometric aplications are investigated for the security perspective, the advantages and disadvantages that these tecniques provide are given. The studies in the literature between 2010-2016 years, used algorithms, technologies, metrics, usage areas, the machine learning techniques used for different biometric systems such as face, palm prints, iris, voice, fingerprint recognition are researched and the studies made are evaluated. The level of security provided by the use of biometric systems by developed using machine learning and disadvantages that arise in the use of these systems are stated in detail in the study. Also, impact on people of biometric methods in terms of ease of use, security and usages areas are examined.","PeriodicalId":356182,"journal":{"name":"2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"9 8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128972083","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}
Amnay Amimeur, Nhathai Phan, D. Dou, David Kil, B. Piniewski
Human behavior prediction is critical to studying how healthy behavior can spread through a social network. In this work we present a novel user representation based human behavior prediction model, the User Representation-based Socialized Gaussian Process model (UrSGP). First, we present the Deep Interaction Representation Learning (Deep Interaction) model for learning latent representations of interaction social networks in which each user is characterized by a set of attributes. In particular, we consider social interaction factors and user attribute factors to build a bimodal, fixed representation of each user in the network. Our model aims to capture the evolution of social interactions and user attributes and learn the hidden correlations between them. We then use our latent features for human behavior prediction via the UrSGP model. An empirical experiment conducted on a real health social network demonstrates that our model outperforms baseline approaches for human behavior prediction.
{"title":"Interaction Network Representations for Human Behavior Prediction","authors":"Amnay Amimeur, Nhathai Phan, D. Dou, David Kil, B. Piniewski","doi":"10.1109/ICMLA.2016.0023","DOIUrl":"https://doi.org/10.1109/ICMLA.2016.0023","url":null,"abstract":"Human behavior prediction is critical to studying how healthy behavior can spread through a social network. In this work we present a novel user representation based human behavior prediction model, the User Representation-based Socialized Gaussian Process model (UrSGP). First, we present the Deep Interaction Representation Learning (Deep Interaction) model for learning latent representations of interaction social networks in which each user is characterized by a set of attributes. In particular, we consider social interaction factors and user attribute factors to build a bimodal, fixed representation of each user in the network. Our model aims to capture the evolution of social interactions and user attributes and learn the hidden correlations between them. We then use our latent features for human behavior prediction via the UrSGP model. An empirical experiment conducted on a real health social network demonstrates that our model outperforms baseline approaches for human behavior prediction.","PeriodicalId":356182,"journal":{"name":"2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129066222","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}
Iulian Carpatorea, Sławomir Nowaczyk, Thorsteinn S. Rögnvaldsson, M. Elmer, J. Lodin
Fuel used by heavy duty trucks is a major cost for logistics companies, and therefore improvements in this area are highly desired. Many of the factors that influence fuel consumption, such as the road type, vehicle configuration or external environment, are difficult to influence. One of the most under-explored ways to lower the costs is training and incentivizing drivers. However, today it is difficult to measure driver performance in a comprehensive way outside of controlled, experimental setting. This paper proposes a machine learning methodology for quantifying and qualifying driver performance, with respect to fuel consumption, that is suitable for naturalistic driving situations. The approach is a knowledge-based feature extraction technique, constructing a normalizing fuel consumption value denoted Fuel under Predefined Conditions (FPC), which captures the effect of factors that are relevant but are not measured directly. The FPC, together with information available from truck sensors, is then compared against the actual fuel used on a given road segment, quantifying the effects associated with driver behavior or other variables of interest. We show that raw fuel consumption is a biased measure of driver performance, being heavily influenced by other factors such as high load or adversary weather conditions, and that using FPC leads to more accurate results. In this paper we also show evaluation the proposed method using large-scale, real-world, naturalistic database of heavy-duty vehicle operation.
{"title":"Learning of Aggregate Features for Comparing Drivers Based on Naturalistic Data","authors":"Iulian Carpatorea, Sławomir Nowaczyk, Thorsteinn S. Rögnvaldsson, M. Elmer, J. Lodin","doi":"10.1109/ICMLA.2016.0194","DOIUrl":"https://doi.org/10.1109/ICMLA.2016.0194","url":null,"abstract":"Fuel used by heavy duty trucks is a major cost for logistics companies, and therefore improvements in this area are highly desired. Many of the factors that influence fuel consumption, such as the road type, vehicle configuration or external environment, are difficult to influence. One of the most under-explored ways to lower the costs is training and incentivizing drivers. However, today it is difficult to measure driver performance in a comprehensive way outside of controlled, experimental setting. This paper proposes a machine learning methodology for quantifying and qualifying driver performance, with respect to fuel consumption, that is suitable for naturalistic driving situations. The approach is a knowledge-based feature extraction technique, constructing a normalizing fuel consumption value denoted Fuel under Predefined Conditions (FPC), which captures the effect of factors that are relevant but are not measured directly. The FPC, together with information available from truck sensors, is then compared against the actual fuel used on a given road segment, quantifying the effects associated with driver behavior or other variables of interest. We show that raw fuel consumption is a biased measure of driver performance, being heavily influenced by other factors such as high load or adversary weather conditions, and that using FPC leads to more accurate results. In this paper we also show evaluation the proposed method using large-scale, real-world, naturalistic database of heavy-duty vehicle operation.","PeriodicalId":356182,"journal":{"name":"2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"105 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124071892","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：