Browsing by Type "conference papers and proceedings postprints"
Now showing 1 - 20 of 39
Results Per Page
Sort Options
Item Anomaly Detection Models for Smart Home Security(IEEE, 2019-08-29) Ramapatruni, Sowmya; Narayanan, Sandeep Nair; Mittal, Sudip; Joshi, Anupam; Joshi, KarunaRecent years have seen significant growth in the adoption of smart homes devices. These devices provide convenience, security, and energy efficiency to users. For example, smart security cameras can detect unauthorized movements, and smoke sensors can detect potential fire accidents. However, many recent examples have shown that they open up a new cyber threat surface. There have been several recent examples of smart devices being hacked for privacy violations and also misused so as to perform DDoS attacks. In this paper, we explore the application of big data and machine learning to identify anomalous activities that can occur in a smart home environment. A Hidden Markov Model (HMM) is trained on network level sensor data, created from a test bed with multiple sensors and smart devices. The generated HMM model is shown to achieve an accuracy of 97% in identifying potential anomalies that indicate attacks. We present our approach to build this model and compare with other techniques available in the literature.Item Bypassing Detection of URL-based Phishing Attacks Using Generative Adversarial Deep Neural Networks(Association for Computing Machinery, 2020-03-18) AlEroud, Ahmed; Karabatis, GeorgeThe URL components of web addresses are frequently used in creating phishing detection techniques. Typically, machine learning techniques are widely used to identify anomalous patterns in URLs as signs of possible phishing. However, adversaries may have enough knowledge and motivation to bypass URL classification algorithms by creating examples that evade classification algorithms. This paper proposes an approach that generates URL-based phishing examples using Generative Adversarial Networks. The created examples can fool Blackbox phishing detectors even when those detectors are created using sophisticated approaches such as those relying on intra-URL similarities. These created instances are used to deceive Blackbox machine learning-based phishing detection models. We tested our approach using actual phishing datasets. The results show that GAN networks are very effective in creating adversarial phishing examples that can fool both simple and sophisticated machine learning phishing detection models.Item Catching the Cuckoo: Verifying TPM Proximity Using a Quote Timing Side-Channel(Springer, Berlin, Heidelberg, 2011-06-22) Fink, Russell A.; Sherman, Alan T.; Mitchell, Alexander O.; Challener, David C.We present a Trusted Platform Module (TPM) application protocol that detects a certain man in the middle attack where an adversary captures and replaces a legitimate computing platform with an imposter that forwards platform authentication challenges to the captive over a high speed data link. This revised Cuckoo attack allows the imposter to satisfy a user's query of platform integrity, tricking the user into divulging sensitive information to the imposter. Our protocol uses an ordinary smart card to verify the platform boot integrity through TPM quote requests, and to verify TPM proximity by measuring TPM tickstamp times required to answer the quotes. Quotes not answered in an expected amount of time may indicate the presence of an imposter's data link, revealing the Cuckoo attack. We describe a timing model for the Cuckoo attack, and summarize experimental results that demonstrate the feasibility of using timing to detect the Cuckoo attack over practical levels of adversary link speeds.Item Cluster Quality Analysis Using Silhouette Score(IEEE, 2020-11-20) Shahapure, Ketan Rajshekhar; Nicholas, CharlesClustering is an important phase in data mining. Selecting the number of clusters in a clustering algorithm, e.g. choosing the best value of k in the various k-means algorithms [1], can be difficult. We studied the use of silhouette scores and scatter plots to suggest, and then validate, the number of clusters we specified in running the k-means clustering algorithm on two publicly available data sets. Scikit-learn's [4] silhouette score method, which is a measure of the quality of a cluster, was used to find the mean silhouette co-efficient of all the samples for different number of clusters. The highest silhouette score indicates the optimal number of clusters. We present several instances of utilizing the silhouette score to determine the best value of k for those data sets.Item Coarse spaces over the ages(2010) Mandel, Jan; Sousedık, BedrichThe objective of this paper is to explain the principles of the design of a coarse space in a simplified way and by pictures. The focus is on ideas rather than on a more historically complete presentation. That can be found, e.g., in Widlund [2008]. Also, space limitation does not allow us to even the mention many important methods and papers that should be rightfully included. The coarse space facilitates a global exchange of information in multigrid and domain decomposition methods for elliptic problems. This exchange is necessary, because the solution is non-local: its value at any point depends on the right-hand-side at any other point. Both multigrid and domain decomposition combine a global correction in coarse space with local corrections, called smoothing in multigrid and subdomain solves in domain decomposition. In multigrid the coarse space is large (typically, the mesh ratio is 2 or 3 at most) and the local solvers are not very powerful (usually, relaxation). In domain decomposition, the coarse space is small (just one or a few degrees of freedom per subdomain), and the local solvers are powerful (direct solvers on subdomain). But the mathematics is more or less the same.Item Conditional-UNet: A Condition-aware Deep Model for Coherent Human Activity Recognition From Wearables(IEEE) Zhang, Liming; Zhang, Wenbin; Japkowicz, NathalieRecognizing human activities from multi-channel time series data collected from wearable sensors has become an important practical application of machine learning. A serious challenge comes from the presence of coherent activities or body movements, such as movements of the head while walking or sitting, since signals representing these movements are mixed and interfere with each other. Basic multi-label classification typically assumes independence within the multiple activities. This is oversimplified and reduces modeling power even when using stateof-the-art deep learning methods. In this paper, we investigate this new problem, which we name “Coherent Human Activity Recognition (Co-HAR)”, that keeps complete conditional dependency between the multiple labels. Additionally, we treat CoHAR as a dense labelling problem that classifies each sample on a time step with multiple coherent labels to provide high-fidelity and duration-sensitive support to high-precision applications. To explicitly model conditional dependency, a novel conditionaware deep architecture “Conditional-UNet” is developed to allow for multiple dense labeling for Co-HAR. We also contribute a first-of-its-kind Co-HAR dataset for head gesture recognition associated with a user’s activity, walking or sitting, to the research community. Extensive experiments on this dataset show that our model outperforms state-of-the-art deep learning methods and achieves up to 92% accuracy on context-based head gesture classification.Item Data Compression for Optimization of a Molecular Dynamics System: Preserving Basins of Attraction(Springer, Cham, 2019-06-08) Retzlaff, Michael; Munson, Todd; Di, Zichao (Wendy)Understanding the evolution of atomistic systems is essential in various fields such as materials science, biology, and chemistry. The gold standard for these calculations is molecular dynamics, which simulates the dynamical interaction between pairs of molecules. The main challenge of such simulation is the numerical complexity, given a vast number of atoms over a long time scale. Furthermore, such systems often contain exponentially many optimal states, and the simulation tends to get trapped in local configurations. Recent developments leverage the existing temporal evolution of the system to improve the stability and scalability of the method; however, they suffer from large data storage requirements. To efficiently compress the data while retaining the basins of attraction, we have developed a framework to determine the acceptable level of compression for an optimization method by application of a Kantorovich-type theorem, using binary digit rounding as our compression technique. Choosing the Lennard-Jones potential function as a model problem, we present a method for determining the local Lipschitz constant of the Hessian with low computational cost, thus allowing the use of our technique in real-time computation.Item Deep Discriminative Learning for Autism Spectrum Disorder Classification(Springer, Cham, 2020-09-14) Zhang, Mingli; Zhao, Xin; Zhang, Wenbin; Chaddad, Ahmad; Evans, Alan; Poline, Jean BaptisteAutism spectrum disorder (ASD) is a complex neurodevelopmental disorder characterized by deficiencies in social, communication and repetitive behaviors. We propose imaging-based ASD biomarkers to find the neural patterns related ASD as the primary goal of identifying ASD. The secondary goal is to investigate the impact of imaging-patterns for ASD. In this paper, we model and explore the identification of ASD by learning a representation of the T1 MRI and fMRI by fusioning a discriminative learning (DL) approach and deep convolutional neural network. Specifically, a class-wise analysis dictionary to generate non-negative low-rank encoding coefficients with the multi-model data, and an orthogonal synthesis dictionary to reconstruct the data. Then, we map the reconstructed data with the original multi-modal data as input of the deep learning model. Finally, the learned priors from both model are returned to the fusion framework to perform classification. The effectiveness of the proposed approach was tested on a world-wide cross-site (34) database of 1127 subjects, experiments show competitive results of the proposed approach. Furthermore, we were able to capture the status of brain neural patterns with the known input of the same modality.Item A Deep Learning Model for Detecting Dust in Earth's Atmosphere from Satellite Remote Sensing Data(IEEE, 2020-11-06) Hou, Ping; Guo, Pei; Wu, Peng; Wang, Jianwu; Gangopadhyay, Aryya; Zhang, ZhiboIn this paper we develop a deep learning model to distinguish dust from cloud and surface using satellite remote sensing image data. The occurrence of dust storms is increasing along with global climate change, especially in the arid and semi-arid regions. Originated from the soil, dust acts as a type of aerosol that causes significant impacts on the environment and human health. The dust and cloud data labels used in this paper are from CALIPSO (Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation) satellite. The radiometric channels and geometric parameters from VIIRS (Visible Infrared Imaging Radiometer Suite) satellite sensor serve as features for our model. We trained and tested our deep learning model using 10,000 samples in March 2012. The developed model has five hidden layers and 512 neurons in each layer. The classification accuracy on the test set is 71.1%. In addition, we performed a shuffling procedure to identify the importance of features, which is calculated as the increase in the prediction error after we permute the feature's values. We also developed a method based on genetic algorithm to find the best subset of features for dust detection. The results show that the genetic algorithm can select a subset of features that have comparable performance as that of a model with all features. The shuffling procedure and the genetic algorithm both identify geometric information as important features for detecting mineral dust. The chosen subset will improve computational efficiency for dust detection and improve physical based methods.Item Design and Deployment of a Flash Flood Monitoring IoT: Challenges and Opportunities(IEEE, 2020-11-06) Basnyat, Bipendra; Singh, Neha; Roy, Nirmalya; Gangopadhyay, AryyaSuccessful implementation of the Internet of Thing (IoT) is precursory to a thriving smart city. However, the technical, physical, and environmental conditions can often pose challenges in their successful deployments. The deployment is further complicated if the time and location of implementation are amidst a natural disaster. In this work, we use flash flood detection as a natural hazard testbed and describe various IoT deployment, our progression, and first-hand experience from those implementations. We compare and contrast three IoTs and their performance in real-time execution. Next, we discuss systems architecture and their end-to-end design and present lessons learned from these heterogeneous deployments. Additionally, we evaluate and outline our observations, challenges, and opportunities for further improvement. We also formulate standard evaluation metrics for their scoring and document our deployment journey.Item Detecting Common Insulation Problems in Built Environments using Thermal Images(IEEE, 2019-08-01) Khan, Naima; Pathak, Nilavra; Roy, NirmalyaProper thermal insulation yields optimum energy expenses in buildings by maintaining necessary heat gain or loss through the built envelope. However, improper thermal insulation causes significant energy wastage along with infusing various damages on indoor and outdoor walls of the buildings, for example, damp areas, black stains, cracks, paint bubbles etc. Therefore, it is important to inspect the temperature variations in different areas of the built environments in regular basis. We propose a method for identifying temperature variance in building thermal images based on Symbolic Aggregated Approximation (SAX). Our process helps detect the temperature variation over different image segments and infers the fault prone segments of leakages. We have collected about 50 thermal images associated with different types of wall specific insulation problems in indoor built environment and were able to identify the affected area with approximately 75% accuracy using our proposed method based on temperature variation detection approach.Item Differential current-mode clock distribution(IEEE, 2015-10-01) Islam, Riadul; Fahmy, Hany; Lin, Ping-Yao; Guthaus, Matthew R.In this paper, we present a differential current-mode pulsed flip-flop (DCMPFF) for low-power clock distribution using a representative 45nm CMOS technology. Experimental results show that the DCMPFF has 47% faster clock-to-output (CLK-Q) delay than a traditional voltage-mode (VM) pulsed flip-flop. When the DCMPFF is integrated with a differential current-mode clock distribution, the differential technique saves 62% and 17% power compared to a conventional VM and a previous current-mode (CM) clock network, respectively.Item A Digital Dashboard for Supporting Online Student Teamwork(Association for Computing Machinery, 2019-11) Ahuja, Rohan; Khan, Daniyal; Symonette, Danilo; desJardins, Marie; Stacey, Simon; Engel, DonTeamwork skills are crucial to college students, both at university and afterwards. However, few tools exist to monitor student teamwork and to help students develop teamwork skills. We present a tool which collects the interactions of students who are using online platforms to complete a sustained task as a team; conducts a range of analyses of these data; and then presents information about team and team member behaviors in real time on a digital dashboard. This dashboard provides instructors with a user-friendly picture of team and team-member dynamics, which can also be made available, as appropriate, to both teams and team members. While some behaviors have been shown to be (or are self-evidently) beneficial or harmful to team performance, these data and analyses also make possible exploration of whether less obvious behaviors affect team outcomes and performance.Item FEAT: A Fairness-Enhancing and Concept-Adapting Decision Tree Classifier(Springer, Cham, 2020-10-15) Zhang, Wenbin; Bifet, AlbertFairness-aware learning is increasingly important in socially-sensitive applications for the sake of achieving optimal and non-discriminative decision-making. Most of the proposed fairness-aware learning algorithms process the data in offline settings and assume that the data is generated by a single concept without drift. Unfortunately, in many real-world applications, data is generated in a streaming fashion and can only be scanned once. In addition, the underlying generation process might also change over time. In this paper, we propose and illustrate an efficient algorithm for mining fair decision trees from discriminatory and continuously evolving data streams. This algorithm, called FEAT (Fairness-Enhancing and concept-Adapting Tree), is based on using the change detector to learn adaptively from non-stationary data streams, that also accounts for fairness. We study FEAT’s properties and demonstrate its utility through experiments on a set of discriminated and time-changing data streams.Item Flexible and Adaptive Fairness-aware Learning in Non-stationary Data Streams(IEEE) Zhang, Wenbin; Zhang, Mingli; Zhang, Ji; Liu, Zhen; Chen, Zhiyuan; Wang, Jianwu; Raff, Edward; Messina, EnzaArtificial intelligence (AI)-based decision-making systems are employed nowadays in an ever growing number of online as well as offline services–some of great importance. Depending on sophisticated learning algorithms and available data, these systems are increasingly becoming automated and data-driven. However, these systems can impact individuals and communities with ethical or legal consequences. Numerous approaches have therefore been proposed to develop decision making systems that are discrimination-conscious by-design. However, these methods assume the underlying data distribution is stationary without drift, which is counterfactual in many real world applications. In addition, their focus has been largely on minimizing discrimination while maximizing prediction performance without necessary flexibility in customizing the tradeoff according to different applications. To this end, we propose a learning algorithm for fair classification that also adapts to evolving data streams and further allows for a flexible control on the degree of accuracy and fairness. The positive results on a set of discriminated and non-stationary data streams demonstrate the effectiveness and flexibility of this approach.Item Flood Detection Framework Fusing The Physical Sensing & Social Sensing(IEEE, 2020-09) Singh, Neha; Basnyat, Bipendra; Roy, Nirmalya; Gangopadhyay, AryyaWe investigate the practical challenge of localized flood detection in real smart city environment using the fusion of physical sensor and social sensing models to depict a reliable and accurate flood monitoring and detection framework. Our proposed framework efficiently utilize the physical and social sensing models to provide the flood-related updates to the city officials. We deployed our flood monitoring system in Ellicott City, Maryland, USA and connect it to the social sensing module to perform the flood-related sensor and social data integration and analysis. Our ground-based sensor network model record and performs the predictive data analytic by forecasting the rise in water level (RMSE=0.2) that demonstrates the severity of upcoming flash floods whereas, our social sensing model helps collect and track the flood-related feeds from Twitter. We employ a pre-trained model and inductive transfer learning based approach to classify the flood-related tweets with 90% accuracy in the use of unseen target flood events. Finally our flood detection framework categorizes the flood relevant localized contextual details into more meaningful classes in order to help the emergency services and local authorities for effective decision making.Item Human motor cortical activity recorded with Micro-ECoG electrodes, during individual finger movements(IEEE, 2009-11-13) Wang, W.; Degenhart, A. D.; Collinger, J. L.; Vinjamuri, Ramana; Sudre, G. P.; Adelson, P. D.; Holder, D. L.; Leuthardt, E. C.; Moran, D. W.; Boninger, M. L.; Schwartz, A. B.; Crammond, D. J.; Tyler-Kabara, E. C.; Weber, D. J.In this study human motor cortical activity was recorded with a customized micro-ECoG grid during individual finger movements. The quality of the recorded neural signals was characterized in the frequency domain from three different perspectives: (1) coherence between neural signals recorded from different electrodes, (2) modulation of neural signals by finger movement, and (3) accuracy of finger movement decoding. It was found that, for the high frequency band (60-120 Hz), coherence between neighboring micro-ECoG electrodes was 0.3. In addition, the high frequency band showed significant modulation by finger movement both temporally and spatially, and a classification accuracy of 73% (chance level: 20%) was achieved for individual finger movement using neural signals recorded from the micro-ECoG grid. These results suggest that the micro-ECoG grid presented here offers sufficient spatial and temporal resolution for the development of minimally-invasive brain-computer interface applications.Item Integrating Text Embedding with Traditional NLP Features for Clinical Relation Extraction(IEEE) Hasan, Fatema; Roy, Arpita; Pan, ShimeiRecently, text embedding techniques such as Word2Vec and BERT have produced state-of-the-art results in a wide variety of NLP tasks. As a result, traditional NLP features frequently used in Information Extraction (IE) such as POS tags, dependency relations and semantic types have received less attention. In this paper, we investigate whether traditional NLP features can be combined with word and sentence embeddings to improve relation extraction. We have explored diverse feature sets and different neural network architectures and evaluated our models on a benchmark clinical text dataset. Our new models significantly outperformed all the baselines on the same dataset.Item Measuring Peer Mentoring Effectiveness in Computing Courses: A Case study in Data Analytics for Cybersecurity(IEEE, 2020-02-20) Faridee, Abu Zaher Md; Janeja, Vandana P.Computing courses often suffer from lack of diversity. In this paper we evaluate an intervention method of peer mentoring to help increase interest in data analytics in cybersecurity. We present a text mining approach to analyze student assignments while they undergo a peer mentoring exercise. In our prior work, we have shown that the peer mentoring approach is effective at improving the students' interest in cybersecurity careers and contributes to an overall better knowledge gain throughout the semester. This was also reflected by an improvement in grades with two years of anonymous survey results. Across the years we also observed that peer mentoring is particularly effective in diverse groups. In this paper, we perform text mining of the written assignments for analyzing the group behavior of the control and experiment sections of a class while also documenting the effectiveness of intervention methods such as peer mentoring. We employ a few text mining techniques, namely Text Frequency Analysis, Lexical Diversity, Readability Analysis, Word Cloud Visualization, Hyperlink usage and Objectivity Analysis on the text assignments submitted by the students and show that students who receive peer mentoring are able to express more complex ideas with fewer words and thereby receive higher grades by the end of the semester. Based on these results, we also discuss how our methodology would be applicable in increasing reachability and diversity in other specialized computing courses such as Big Data and distributed systems.Item Mobile agents can benefit from standards efforts on interagent communication(IEEE, 1998-07-01) Finin, T.; Labrou, Y.; Peng, YunOn the road for the future success of mobile agents, we believe that inter-agent communication is an issue that has not been adequately addressed by the mobile agents community. Supplementing mobile agents with the ability to interact with other mobile or static agents, or agentified information sources is a necessity in the vastly heterogeneous arena where mobile agents are called to compete. Thus, an agent communication language should be interpreted as a tool with the capacity to integrate disparate sources of information. In the first segment, we argue that mobile agents can benefit from current standards efforts on agent communication since the focus of such work is to address heterogeneity by defining a “common language” for communicating agents. In the second part, we discuss ongoing research on agent to agent communication and we present current standards efforts relevant to agent communication.