Engineering Commons^™

Domain Adaptation In Unmanned Aerial Vehicles Landing Using Reinforcement Learning, Pedro Lucas Franca Albuquerque

Department of Computer Science and Engineering: Dissertations, Theses, and Student Research

Landing an unmanned aerial vehicle (UAV) on a moving platform is a challenging task that often requires exact models of the UAV dynamics, platform characteristics, and environmental conditions. In this thesis, we present and investigate three different machine learning approaches with varying levels of domain knowledge: dynamics randomization, universal policy with system identification, and reinforcement learning with no parameter variation. We first train the policies in simulation, then perform experiments both in simulation, making variations of the system dynamics with wind and friction coefficient, then perform experiments in a real robot system with wind variation. We initially expected that providing …

Identifying Regional Trends In Avatar Customization, Peter Mawhorter, Sercan Sengun, Haewoon Kwak, D. Fox Harrell

Research Collection School Of Computing and Information Systems

Since virtual identities such as social media profiles and avatars have become a common venue for self-expression, it has become important to consider the ways in which existing systems embed the values of their designers. In order to design virtual identity systems that reflect the needs and preferences of diverse users, understanding how the virtual identity construction differs between groups is important. This paper presents a new methodology that leverages deep learning and differential clustering for comparative analysis of profile images, with a case study of almost 100 000 avatars from a large online community using a popular avatar creation …

Tuning Networks For Prosocial Behavior: From Senseless Swarms To Smart Mobs [Commentary], Sun Sun Lim, Roland Bouffanais

Research Collection College of Integrative Studies

Social media have been seen to accelerate the spread of negative content such as disinformation and hate speech, often unleashing a reckless herd mentality within networks, further aggravated by malicious entities using bots for amplification. So far, the response to this emerging global crisis has centered around social media platform companies making reactive moves that appear to have greater symbolic value than practical utility. Proposes a solution to favor prosocial behavior via social networks.

Of Promoting Networking And Protecting Privacy: Effects Of Defaults And Regulatory Focus On Social Media Users’ Preference Settings, Hichang Cho, Sungjong Roh, Byungho Park

Research Collection Lee Kong Chian School Of Business

Privacyresearch has debated whether privacy decision-making is determined by users'stable preferences (i.e., individual traits), privacy calculus (i.e.,cost-benefit analysis), or “responses on the spot” that vary across contexts.This study focuses on two factors—default setting as a contextual factor andregulatory focus as an individual difference factor—and examines the degree towhich these factors affect social media users' decisionmaking when usingprivacy preference settings in a fictitious social networking site. Theresults, based on two experimental studies (study 1, n = 414; study 2, n =213), show that default settings significantly affect users' privacypreferences, such that users choose the defaults or alternatives proximal tothem. Study 2 …

Image Classification Using Fuzzy Fca, Niruktha Roy Gotoor

Department of Computer Science and Engineering: Dissertations, Theses, and Student Research

Formal concept analysis (FCA) is a mathematical theory based on lattice and order theory used for data analysis and knowledge representation. It has been used in various domains such as data mining, machine learning, semantic web, Sciences, for the purpose of data analysis and Ontology over the last few decades. Various extensions of FCA are being researched to expand it's scope over more departments. In this thesis,we review the theory of Formal Concept Analysis (FCA) and its extension Fuzzy FCA. Many studies to use FCA in data mining and text learning have been pursued. We extend these studies to include …

Pixel-Level Deep Multi-Dimensional Embeddings For Homogeneous Multiple Object Tracking, Mateusz Mittek

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

The goal of Multiple Object Tracking (MOT) is to locate multiple objects and keep track of their individual identities and trajectories given a sequence of (video) frames. A popular approach to MOT is tracking by detection consisting of two processing components: detection (identification of objects of interest in individual frames) and data association (connecting data from multiple frames). This work addresses the detection component by introducing a method based on semantic instance segmentation, i.e., assigning labels to all visible pixels such that they are unique among different instances. Modern tracking methods often built around Convolutional Neural Networks (CNNs) and additional, …

Finding Needles In A Haystack: Leveraging Co-Change Dependencies To Recommend Refactorings, Marcos César De Oliveira, Davi Freitas, Rodrigo Bonifacio, Gustavo Pinto, David Lo

Research Collection School Of Computing and Information Systems

A fine-grained co-change dependency arises when two fine-grained source-code entities, e.g., a method,change frequently together. This kind of dependency is relevant when considering remodularization efforts (e.g., to keep methods that change together in the same class). However, existing approaches forrecommending refactorings that change software decomposition (such as a move method) do not explorethe use of fine-grained co-change dependencies. In this paper we present a novel approach for recommending move method and move field refactorings, which removes co-change dependencies and evolutionary smells, a particular type of dependency that arise when fine-grained entities that belong to different classes frequently change together. First …

Do My Students Understand? Automated Identification Of Doubts From Informal Reflections, Siaw Ling Lo, Kar Way Tan, Eng Lieh Ouh

Research Collection School Of Computing and Information Systems

Traditionallyteaching is usually one directional where the instructor imparts knowledge andthere is minimal interaction between learners and instructor. With the focus onlearner-centred pedagogy, it can be a challenge to provide timely and relevantguidance to individual learners according to their levels of understanding. Oneof the options available is to collect reflections from learners after eachlesson to extract relevant and high-value feedback so that doubts or questionscan be addressed in a timely manner. In this paper, we derived an approach toautomate the identification of doubts from the informal reflections through featuresanalysis and machine learning. Using reflections as a feedback mechanism andaligning it …

Digitalization In Practice: The Fifth Discipline Advantage, Siu Loon Hoe

Research Collection School Of Computing and Information Systems

Purpose The purpose of this paper is to provide advice to organizations on how to become successful in the digital age. The paper revisits Peter Senge's (1990) notion of the learning organization and discusses the relevance of systems thinking and the other four disciplines, namely, personal mastery, mental models, shared vision and team learning in the context of the current digitalization megatrend. Design/methodology/approach This paper is based on content analysis of essays from international organizations, strategy experts and management scholars, and insights gained from the author's consulting experience. A comparative case study from the health and social sector is also …

Robot Simulation Analysis, Jacob Miller, Jeremy Evert

Student Research

• Simulate virtual robot for test and analysis

• Analyze SLAM solutions using ROS

• Assemble a functional Turtlebot

• Emphasize projects related to current research trajectories for NASA, and general robotics applications

Reputation-Aware Trajectory-Based Data Mining In The Internet Of Things (Iot), Samia Tasnim

FIU Electronic Theses and Dissertations

Internet of Things (IoT) is a critically important technology for the acquisition of spatiotemporally dense data in diverse applications, ranging from environmental monitoring to surveillance systems. Such data helps us improve our transportation systems, monitor our air quality and the spread of diseases, respond to natural disasters, and a bevy of other applications. However, IoT sensor data is error-prone due to a number of reasons: sensors may be deployed in hazardous environments, may deplete their energy resources, have mechanical faults, or maybe become the targets of malicious attacks by adversaries. While previous research has attempted to improve the quality of …

Trajectory Privacy Preservation And Lightweight Blockchain Techniques For Mobility-Centric Iot, Abdur Bin Shahid

FIU Electronic Theses and Dissertations

Various research efforts have been undertaken to solve the problem of trajectory privacy preservation in the Internet of Things (IoT) of resource-constrained mobile devices. Most attempts at resolving the problem have focused on the centralized model of IoT, which either impose high delay or fail against a privacy-invading attack with long-term trajectory observation. These proposed solutions also fail to guarantee location privacy for trajectories with both geo-tagged and non-geo-tagged data, since they are designed for geo-tagged trajectories only. While a few blockchain-based techniques have been suggested for preserving trajectory privacy in decentralized model of IoT, they require large storage capacity …

Malware Analysis For Evaluating The Integrity Of Mission Critical Devices, Robert Heras

FIU Electronic Theses and Dissertations

The rapid evolution of technology in our society has brought great advantages, but at the same time it has increased cybersecurity threats. At the forefront of these threats is the proliferation of malware from traditional computing platforms to the rapidly expanding Internet-of-things. Our research focuses on the development of a malware detection system that strives for early detection as a means of mitigating the effects of the malware's execution.

The proposed scheme consists of a dual-stage detector providing malware detection for compromised devices in order to mitigate the devices malicious behavior. Furthermore, the framework analyzes task structure features as well …

Security Of The Internet Of Things: Vulnerabilities, Attacks And Countermeasures, Ismail Butun, Houbing Song, Patrik Osterberg

Publications

Wireless Sensor Networks (WSNs) constitute one of the most promising third-millennium technologies and have wide range of applications in our surrounding environment. The reason behind the vast adoption of WSNs in various applications is that they have tremendously appealing features, e.g., low production cost, low installation cost, unattended network operation, autonomous and longtime operation. WSNs have started to merge with the Internet of Things (IoT) through the introduction of Internet access capability in sensor nodes and sensing ability in Internet-connected devices. Thereby, the IoT is providing access to huge amount of data, collected by the WSNs, over the Internet. Hence, …

Water Pipeline Leakage Detection Based On Machine Learning And Wireless Sensor Networks, Yang Liu, Xuehui Ma, Yong Tie, Yinghui Zhang, Jing Gao

Department of Electrical and Computer Engineering: Faculty Publications

The detection of water pipeline leakage is important to ensure that water supply networks can operate safely and conserve water resources. To address the lack of intelligent and the low efficiency of conventional leakage detection methods, this paper designs a leakage detection method based on machine learning and wireless sensor networks (WSNs). The system employs wireless sensors installed on pipelines to collect data and utilizes the 4G network to perform remote data transmission. A leakage triggered networking method is proposed to reduce the wireless sensor network’s energy consumption and prolong the system life cycle effectively. To enhance the precision and …

Machine Current Sensor Fdi Strategy In Pmsms, Haibo Li, Yi Qian, Sohrab Asgarpoor, Hamid Sharif

Department of Electrical and Computer Engineering: Faculty Publications

This work proposes a machine current sensor fault detection and isolation (FDI) strategy in permanent magnet synchronous machines (PMSMs) resilient to multiple faults. The fault detection is performed by comparing the measured and estimated DC link currents. The fault isolation is achieved according to machine phase signal estimation and the corresponding residual examination. Single sensor fault, multiple sensor faults and non-sensor fault are covered by the proposed FDI method. The proposed sensor FDI method is not influenced by machine imbalance, feasible for FDI of both single and multiple machine current sensor faults, and capable of distinguishing between machine current sensor …

Algorithms And Circuits For Analog-Digital Hybrid Multibeam Arrays, Paboda Viduneth A. Beruwawela Pathiranage

FIU Electronic Theses and Dissertations

Fifth generation (5G) and beyond wireless communication systems will rely heavily on larger antenna arrays combined with beamforming to mitigate the high free-space path-loss that prevails in millimeter-wave (mmW) and above frequencies. Sharp beams that can support wide bandwidths are desired both at the transmitter and the receiver to leverage the glut of bandwidth available at these frequency bands. Further, multiple simultaneous sharp beams are imperative for such systems to exploit mmW/sub-THz wireless channels using multiple reflected paths simultaneously. Therefore, multibeam antenna arrays that can support wider bandwidths are a key enabler for 5G and beyond systems.

In general, N- …

Bibliometric Survey Of Privacy Of Social Media Network Data Publishing, Rupali Gangarde Ass. Prof., Amit Sharma Dr., Ambika Pawar Dr.

Library Philosophy and Practice (e-journal)

We are witness to see exponential growth of the social media network since the year 2002. Leading social media networking sites used by people are Twitter, Snapchats, Facebook, Google, and Instagram, etc. The latest global digital report (Chaffey and Ellis-Chadwick 2019) states that there exist more than 800 million current online social media users, and the number is still exploding day by day. Users share their day to day activities such as their photos and locations etc. on social media platforms. This information gets consumed by third party users, like marketing companies, researchers, and government firms. Depending upon the purpose, …

The Art Of Selecting Phd Students: Combination Of Bibliometric And Ahp Approach, Preeti Mulay Dr., Rahul Raghvendra Joshi Prof., Sophia Gaikwad Dr.

Library Philosophy and Practice (e-journal)

For the PhD guide or the advisor selecting the accurate PhD scholar is the most elephantine task. It actually requires an art for the perfect selection; as the length, breadth, depth and volume of PhD work is spread across the years and this relationship between the scholar and the guide should start and flourish positively for the immense experience throughout the PhD process. Hence it was essential to understand bibliometric details including how many researchers have already published their contributions in the form of papers and patents, in the Scopus database. In addition to the bibliometric details, in this study, …

Personality Prediction Through Curriculam Vitae Analysis Involving Password Encryption And Prediction Analysis, Gagandeep Kaur, Shruti Maheshwari

Library Philosophy and Practice (e-journal)

A recruitment process requires an eligibility check, an aptitude evaluation and a psychometric analysis of prospective candidates. The work puts forward an application where the system allows employers to post new job offerings and registered candidates can apply. The application estimates applicant’s emotional aptitude through a psychometric analysis based on a test whereas the professional standard is verified via a technical aptitude test. OCEAN Model is used to assess emotional quotient and predict the personality traits. Machine learning techniques such as Logistic Regression are used for modelling the personality predictor. The details of the candidates are kept secure by using …

Aspect And Opinion Aware Abstractive Review Summarization With Reinforced Hard Typed Decoder, Yufei Tian, Jianfei Yu, Jing Jiang

Research Collection School Of Computing and Information Systems

In this paper, we study abstractive review summarization. Observing that review summaries often consist of aspect words, opinion words and context words, we propose a two-stage reinforcement learning approach, which first predicts the output word type from the three types, and then leverages the predicted word type to generate the final word distribution. Experimental results on two Amazon product review datasets demonstrate that our method can consistently outperform several strong baseline approaches based on ROUGE scores.

Data Analytics And Machine Learning To Enhance The Operational Visibility And Situation Awareness Of Smart Grid High Penetration Photovoltaic Systems, Aditya Sundararajan

FIU Electronic Theses and Dissertations

Electric utilities have limited operational visibility and situation awareness over grid-tied distributed photovoltaic systems (PV). This will pose a risk to grid stability when the PV penetration into a given feeder exceeds 60% of its peak or minimum daytime load. Third-party service providers offer only real-time monitoring but not accurate insights into system performance and prediction of productions. PV systems also increase the attack surface of distribution networks since they are not under the direct supervision and control of the utility security analysts.

Six key objectives were successfully achieved to enhance PV operational visibility and situation awareness: (1) conceptual cybersecurity …

Automated Dynamic Detection Of Self-Hiding Behaviors, Luke Baird

Student Works

Certain Android applications, such as but not limited to malware, conceal their presence from the user, exhibiting a self-hiding behavior. Consequently, these apps put the user’s security and privacy at risk by performing tasks without the user’s awareness. Static analysis has been used to analyze apps for self-hiding behavior, but this approach is prone to false positives and suffers from code obfuscation. This research proposes a set of three tools utilizing a dynamic analysis method of detecting self-hiding behavior of an app in the home, installed, and running application lists on an Android emulator. Our approach proves both highly accurate …

A Co-Optimal Coverage Path Planning Method For Aerial Scanning Of Complex Structures, Zhexiong Shang, Justin Bradley, Zhigang Shen

Department of Construction Engineering and Management: Faculty Publications

The utilization of unmanned aerial vehicles (UAVs) in survey and inspection of civil infrastructure has been growing rapidly. However, computationally efficient solvers that find optimal flight paths while ensuring high-quality data acquisition of the complete 3D structure remains a difficult problem. Existing solvers typically prioritize efficient flight paths, or coverage, or reducing computational complexity of the algorithm – but these objectives are not co-optimized holistically. In this work we introduce a co-optimal coverage path planning (CCPP) method that simultaneously co-optimizes the UAV path, the quality of the captured images, and reducing computational complexity of the solver all while adhering to …

A Comparison Of Contextual Bandit Approaches To Human-In-The-Loop Robot Task Completion With Infrequent Feedback, Matt Mcneill, Damian Lyons

Faculty Publications

Artificially intelligent assistive agents are playing an increased role in our work and homes. In contrast with currently predominant conversational agents, whose intelligence derives from dialogue trees and external modules, a fully autonomous domestic or workplace robot must carry out more complex reasoning. Such a robot must make good decisions as soon as possible, learn from experience, respond to feedback, and rely on feedback only as much as necessary. In this research, we narrow the focus of a hypothetical robot assistant to a room tidying task in a simulated domestic environment. Given an item, the robot chooses where to put …

Deep Learning (Partly) Demystified, Vladik Kreinovich, Olga Kosheleva

Departmental Technical Reports (CS)

Successes of deep learning are partly due to appropriate selection of activation function, pooling functions, etc. Most of these choices have been made based on empirical comparison and heuristic ideas. In this paper, we show that many of these choices -- and the surprising success of deep learning in the first place -- can be explained by reasonably simple and natural mathematics.

Computing Without Computing: Dna Version, Vladik Kreinovich, Julio C. Urenda

Departmental Technical Reports (CS)

The traditional DNA computing schemes are based on using or simulating DNA-related activity. This is similar to how quantum computers use quantum activities to perform computations. Interestingly, in quantum computing, there is another phenomenon known as computing without computing, when, somewhat surprisingly, the result of the computation appears without invoking the actual quantum processes. In this chapter, we show that similar phenomenon is possible for DNA computing: in addition to the more traditional way of using or simulating DNA activity, we can also use DNA inactivity to solve complex problems. We also show that while DNA computing without …

Why Deep Learning Is More Efficient Than Support Vector Machines, And How It Is Related To Sparsity Techniques In Signal Processing, Laxman Bokati, Olga Kosheleva, Vladik Kreinovich

Departmental Technical Reports (CS)

Several decades ago, traditional neural networks were the most efficient machine learning technique. Then it turned out that, in general, a different technique called support vector machines is more efficient. Reasonably recently, a new technique called deep learning has been shown to be the most efficient one. These are empirical observations, but how we explain them -- thus making the corresponding conclusions more reliable? In this paper, we provide a possible theoretical explanation for the above-described empirical comparisons. This explanation enables us to explain yet another empirical fact -- that sparsity techniques turned out to be very efficient in signal …

Global Research Trend On Cyber Security: A Scientometric Analysis, Somesh Rai, Kunwar Singh Dr, Akhilesh Kumar Varma

Library Philosophy and Practice (e-journal)

Scientometrics is a quantitative analysis of scholarly literature related to a particular subject or area (well defined by some limits, scope and coverage), which helps to understand different aspects about the scholarly literature’s growth in various dimensions of knowledge. Similarly, this study is a quantitative analysis of the Global research trends in cyber security. Some works related to scientometrics of ‘deception, counter-deception in cyberspace’ had been published in 2011, but we have focused on ‘cyber security’ as the topic of research. For analysis we have utilised the published data available in Scopus database, which is directly related to ‘cyber security’. …

Flexibility Of Remediation Methods For Winding Open Circuit Faults In A Multiphase Pm Machine Considering Iron Losses Minimization, Fan Wu, Ayman M. El-Refaie

Electrical and Computer Engineering Faculty Research and Publications

The flexibility of post-fault control in multiphase machine systems stems from their multiple degrees of freedom. A post-fault loss-minimization method is proposed and investigated in this paper, in which both the machine copper and iron losses are considered during the derivation of post-fault remediation methods. Therefore, machine efficiency during post-fault operation can be further improved compared to the conventional stator-ohmic-loss-minimization approach. In addition, the combination of three key factors/constraints that can influence the post-fault control strategy of a six-phase permanent magnet (PM) machine has been investigated. By comparing four selected remediation methods based on three constraints, the pros and cons …