Computer Sciences Commons^™

Directional Co-Clustering, Aghiles Salah, Mohamed Nadif

Research Collection School Of Computing and Information Systems

Co-clustering addresses the problem of simultaneous clustering of both dimensions of a data matrix. When dealing with high dimensional sparse data, co-clustering turns out to be more beneficial than one-sided clustering even if one is interested in clustering along one dimension only. Aside from being high dimensional and sparse, some datasets, such as document-term matrices, exhibit directional characteristics, and the L2 normalization of such data, so that it lies on the surface of a unit hypersphere, is useful. Popular co-clustering assumptions such as Gaussian or Multinomial are inadequate for this type of data. In this paper, we extend the scope …

Rotation Invariant Convolutions For 3d Point Clouds Deep Learning, Zhiyuan Zhang, Binh-Son Hua, David W. Rosen, Sai-Kit Yeung

Research Collection School Of Computing and Information Systems

Recent progresses in 3D deep learning has shown that it is possible to design special convolution operators to consume point cloud data. However, a typical drawback is that rotation invariance is often not guaranteed, resulting in networks that generalizes poorly to arbitrary rotations. In this paper, we introduce a novel convolution operator for point clouds that achieves rotation invariance. Our core idea is to use low-level rotation invariant geometric features such as distances and angles to design a convolution operator for point cloud learning. The well-known point ordering problem is also addressed by a binning approach seamlessly built into the …

Automatic Generation Of Non-Intrusive Updates For Third-Party Libraries In Android Applications, Yue Duan, Lian Gao, Jie Hu, Heng Yin

Research Collection School Of Computing and Information Systems

Third-Party libraries, which are ubiquitous in Android apps,have exposed great security threats to end users as they rarelyget timely updates from the app developers, leaving manysecurity vulnerabilities unpatched. This issue is due to thefact that manually updating libraries can be technically nontrivialand time-consuming for app developers. In this paper,we propose a technique that performs automatic generationof non-intrusive updates for third-party libraries in Androidapps. Given an Android app with an outdated library and anewer version of the library, we automatically update the oldlibrary in a way that is guaranteed to be fully backward compatibleand imposes zero impact to the library’s interactionswith …

Puncturable Proxy Re-Encryption Supporting To Group Messaging Service, Tran Viet Xuan Phuong, Willy Susilo, Jongkil Kim, Guomin Yang, Dongxi Liu

Research Collection School Of Computing and Information Systems

This work envisions a new encryption primitive for many-to-many paradigms such as group messaging systems. Previously, puncturable encryption (PE) was introduced to provide forward security for asynchronous messaging services. However, existing PE schemes were proposed only for one-to-one communication, and causes a significant overhead for a group messaging system. In fact, the group communication over PE can only be achieved by encrypting a message multiple times for each receiver by the sender’s device, which is usually suitable to restricted resources such as mobile phones or sensor devices. Our new suggested scheme enables to re-encrypt ciphertexts of puncturable encryption by a …

A Lattice-Based Linkable Ring Signature Supporting Stealth Addresses, Zhen Liu, Khoa Nguyen, Guomin Yang, Huaxiong Wang, Duncan S. Wong

Research Collection School Of Computing and Information Systems

First proposed in CryptoNote, a collection of popular privacy-centric cryptocurrencies have employed Linkable Ring Signature and a corresponding Key Derivation Mechanism (KeyDerM) for keeping the payer and payee of a transaction anonymous and unlinkable. The KeyDerM is used for generating a fresh signing key and the corresponding public key, referred to as a stealth address, for the transaction payee. The stealth address will then be used in the linkable ring signature next time when the payee spends the coin. However, in all existing works, including Monero, the privacy model only considers the two cryptographic primitives separately. In addition, to be …

A Common Approach For Consumer And Provider Fairness In Recommendations, Dimitris Sacharidis, Kyriakos Mouratidis, Dimitrios Kleftogiannis

Research Collection School Of Computing and Information Systems

We present a common approach for handling consumer and provider fairness in recommendations. Our solution requires defining two key components, a classification of items and a target distribution, which together define the case of perfect fairness. This formulation allows distinct fairness concepts to be specified in a common framework. We further propose a novel reranking algorithm that optimizes for a desired trade-off between utility and fairness of a recommendation list.

Predictors Of Ransomware From Binary Analysis, Aaron M. Otis

Master's Theses

Ransomware, a type of malware that extorts payment from a victim by encrypting her data, is a growing threat that is becoming more sophisticated with each generation. Attackers have shifted from targeting individuals to entire organizations, raising extortions from hundreds of dollars to hundreds of thousands of dollars. In this work, we analyze a variety of ransomware and benign software binaries in order to identify indicators that may be used to detect ransomware. We find that several combinations of strings, cryptographic constants, and a large number loops are key indicators useful for detecting ransomware.

Ion Software-Defined Radio Metadata Standard Final Report, Sanjeev Gunawardena, Alexander Rugamer, Muhammad Subhan Hameed, Markel Arizabaleta, Thomas Pany, Javier Arribas

Faculty Publications

The ION GNSS SDR Metadata Standard describes the formatting and other essential PNT-related parameters of sampled data streams and files. This allows processors to seamlessly consume such data without the need to input these parameters manually. The technical development phase of the initial version of the standard has now been deemed complete and is currently undergoing the last remaining procedural steps towards adoption as a formal standard by the Institute of Navigation. This paper reports on the activities of the working group since September 2018 and summarizes the final products of the standard. It also reports on examples of early …

Interlocking Structure Design And Assembly, Yinan Zhang

Dartmouth College Ph.D Dissertations

Many objects in our life are not manufactured as whole rigid pieces. Instead, smaller components are made to be later assembled into larger structures. Chairs are assembled from wooden pieces, cabins are made of logs, and buildings are constructed from bricks. These components are commonly designed by many iterations of human thinking. In this report, we will look at a few problems related to interlocking components design and assembly. Given an atomic object, how can we design a package that holds the object firmly without a gap in-between? How many pieces should the package be partitioned into? How can we …

Optimal Sampling Paths For Autonomous Vehicles In Uncertain Ocean Flows, Andrew J. De Stefan

Dissertations

Despite an extensive history of oceanic observation, researchers have only begun to build a complete picture of oceanic currents. Sparsity of instrumentation has created the need to maximize the information extracted from every source of data in building this picture. Within the last few decades, autonomous vehicles, or AVs, have been employed as tools to aid in this research initiative. Unmanned and self-propelled, AVs are capable of spending weeks, if not months, exploring and monitoring the oceans. However, the quality of data acquired by these vehicles is highly dependent on the paths along which they collect their observational data. The …

Applied Deep Learning In Intelligent Transportation Systems And Embedding Exploration, Xiaoyuan Liang

Dissertations

Deep learning techniques have achieved tremendous success in many real applications in recent years and show their great potential in many areas including transportation. Even though transportation becomes increasingly indispensable in people’s daily life, its related problems, such as traffic congestion and energy waste, have not been completely solved, yet some problems have become even more critical. This dissertation focuses on solving the following fundamental problems: (1) passenger demand prediction, (2) transportation mode detection, (3) traffic light control, in the transportation field using deep learning. The dissertation also extends the application of deep learning to an embedding system for visualization …

Deep Machine Learning Techniques For The Detection And Classification Of Sperm Whale Bioacoustics, Peter C. Bermant, Michael M. Bronstein, Robert J. Wood, Shane Gero, David F. Gruber

Publications and Research

We implemented Machine Learning (ML) techniques to advance the study of sperm whale (Physeter macrocephalus) bioacoustics. This entailed employing Convolutional Neural Networks (CNNs) to construct an echolocation click detector designed to classify spectrograms generated from sperm whale acoustic data according to the presence or absence of a click. The click detector achieved 99.5% accuracy in classifying 650 spectrograms. The successful application of CNNs to clicks reveals the potential of future studies to train CNN-based architectures to extract finer-scale details from cetacean spectrograms. Long short-term memory and gated recurrent unit recurrent neural networks were trained to perform classification tasks, including (1) …

Editorial: Machine Learning In Biomolecular Simulations, Gennady M. Verkhivker, Vojtech Spiwok, Francesco Luigi Gervasio

Mathematics, Physics, and Computer Science Faculty Articles and Research

"Interest in machine learning is growing in all fields of science, industry, and business. This interest was not primarily initiated by new theoretical findings. Interestingly, the theoretical basis of the majority of machine learning techniques, such as artificial neural networks, decision trees, or kernel methods, have been known for a relatively long time. Instead, there are other effects that triggered the recent boom of machine learning."

Going Big: A Large-Scale Study On What Big Data Developers Ask, Mehdi Bagherzadeh, Raffi Khatchadourian

Publications and Research

Software developers are increasingly required to write big data code. However, they find big data software development challenging. To help these developers it is necessary to understand big data topics that they are interested in and the difficulty of finding answers for questions in these topics. In this work, we conduct a large-scale study on Stackoverflow to understand the interest and difficulties of big data developers. To conduct the study, we develop a set of big data tags to extract big data posts from Stackoverflow; use topic modeling to group these posts into big data topics; group similar topics into …

Study On The Development Of Mass Based On Safety, Qiyu Yu

Maritime Safety & Environment Management Dissertations (Dalian)

No abstract provided.

Cooperation In Maritime Search And Rescue Between Democratic People’S Republic Of Korea, The People’S Republic Of China And The Russian Federation, Kwangmyong Ri

Maritime Safety & Environment Management Dissertations (Dalian)

No abstract provided.

A Machine Learning Model For Clustering Securities, Vanessa Torres, Travis Deason, Michael Landrum, Nibhrat Lohria

SMU Data Science Review

In this paper, we evaluate the self-declared industry classifications and industry relationships between companies listed on either the Nasdaq or the New York Stock Exchange (NYSE) markets. Large corporations typically operate in multiple industries simultaneously; however, for investment purposes they are classified as belonging to a single industry. This simple classification obscures the actual industries within which a company operates, and, therefore, the investment risks of that company.
By using Natural Language Processing (NLP) techniques on Security and Exchange Commission (SEC) filings, we obtained self-defined industry classifications per company. Using clustering techniques such as Hierarchical Agglomerative and k-means clustering we …

Is Ai Intelligent, Really?, Bruce D. Baker

SPU Works

The question of intelligence opens up a bouquet of interrelated questions:

Suppose that some future AGI systems (on-screen or robots) equaled human performance. Would they have real intelligence, real understanding, real creativity? Would they have selves, moral standing, free choice? Would they be conscious? And without consciousness, could they have any of those other properties?[1]

The only way out of the morass is to recognize that truth claims do not stand on their own, aloof and cut off from the sea of meaning which grants epistemic access. In other words, truth presumes access to: (1) a way of knowing, …

Machine Learning In Support Of Electric Distribution Asset Failure Prediction, Robert D. Flamenbaum, Thomas Pompo, Christopher Havenstein, Jade Thiemsuwan

SMU Data Science Review

In this paper, we present novel approaches to predicting as- set failure in the electric distribution system. Failures in overhead power lines and their associated equipment in particular, pose significant finan- cial and environmental threats to electric utilities. Electric device failure furthermore poses a burden on customers and can pose serious risk to life and livelihood. Working with asset data acquired from an electric utility in Southern California, and incorporating environmental and geospatial data from around the region, we applied a Random Forest methodology to predict which overhead distribution lines are most vulnerable to fail- ure. Our results provide evidence …

Document Images And Machine Learning: A Collaboratory Between The Library Of Congress And The Image Analysis For Archival Discovery (Aida) Lab At The University Of Nebraska, Lincoln, Ne, Yi Liu, Chulwoo Pack, Leen-Kiat Soh, Elizabeth Lorang

School of Computing: Conference and Workshop Papers

This presentation summarized and presented preliminary results from the first weeks of work conducted by the Aida research team in response to Library of Congress funding notice ID 030ADV19Q0274, “The Library of Congress – Pre-processing Pilot.” It includes overviews of projects on historic document segmentation, document classification, document quality assessment, figure and graph extraction from historic documents, text-line extraction from figures, subject and objective quality assesments, and digitization type differentiation.

Learnfca: A Fuzzy Fca And Probability Based Approach For Learning And Classification, Suraj Ketan Samal

School of Computing: 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. Over the past several years, many of its extensions have been proposed and applied in several domains including data mining, machine learning, knowledge management, semantic web, software development, chemistry ,biology, medicine, data analytics, biology and ontology engineering.

This thesis reviews the state-of-the-art of theory of Formal Concept Analysis(FCA) and its various extensions that have been developed and well-studied in the past several years. We discuss their historical roots, reproduce the original definitions and derivations with illustrative examples. Further, we provide …

A Two-Step Classification Platform To Classify Construction Sounds To Enhance Real-Time Construction Zone Monitoring, Jayati Halder Jui

LSU Master's Theses

In recent years, sounds generated from equipment used at a construction site have been found to be useful in identifying construction activities and associated equipment usage providing insightful information regarding project status and hazard issues. This study proposes a Two-Step Neural Classifier (TSNC) for accurate classification of real world construction sounds. Experimental results from real world construction site audio data showed an average classification accuracy of 98% for TSNC compared to 91% average accuracy for an ordinary (one-step) Artificial Neural Network (ANN). The TSNC model is a hierarchical model that exploits the idea of creating disjoint subgroups of construction sounds …

Machine Learning Predicts Aperiodic Laboratory Earthquakes, Olha Tanyuk, Daniel Davieau, Charles South, Daniel W. Engels

SMU Data Science Review

In this paper we find a pattern of aperiodic seismic signals that precede earthquakes at any time in a laboratory earthquake’s cycle using a small window of time. We use a data set that comes from a classic laboratory experiment having several stick-slip displacements (earthquakes), a type of experiment which has been studied as a simulation of seismologic faults for decades. This data exhibits similar behavior to natural earthquakes, so the same approach may work in predicting the timing of them. Here we show that by applying random forest machine learning technique to the acoustic signal emitted by a laboratory …

Longitudinal Analysis With Modes Of Operation For Aes, Dana Geislinger, Cory Thigpen, Daniel W. Engels

SMU Data Science Review

In this paper, we present an empirical evaluation of the randomness of the ciphertext blocks generated by the Advanced Encryption Standard (AES) cipher in Counter (CTR) mode and in Cipher Block Chaining (CBC) mode. Vulnerabilities have been found in the AES cipher that may lead to a reduction in the randomness of the generated ciphertext blocks that can result in a practical attack on the cipher. We evaluate the randomness of the AES ciphertext using the standard key length and NIST randomness tests. We evaluate the randomness through a longitudinal analysis on 200 billion ciphertext blocks using logistic regression and …

Pristine Sentence Translation: A New Approach To A Timeless Problem, Meenu Ahluwalia, Brian Coari, Ben Brock

SMU Data Science Review

Abstract.

Pristine Sentence Translation (PST) is a new approach to language translation based upon sentence-level granularity. Traditional translation approaches, including those utilizing advanced machine learning or neural network-based approaches, translate on a word-by-word or phrase-by-phrase basis; thereby, potentially missing the context or meaning of the complete sentence. Instead of these piecewise translations, PST utilizes deep learning and predictive modeling techniques to translate complete sentences from their source language into their target language. With these approaches we were able to translate sentences that closely conveyed the meaning of the original sentences. Our results demonstrated that PST’s method of translating an entire …

Texture-Based Deep Neural Network For Histopathology Cancer Whole Slide Image (Wsi) Classification, Nelson Zange Tsaku

Master of Science in Computer Science Theses

Automatic histopathological Whole Slide Image (WSI) analysis for cancer classification has been highlighted along with the advancements in microscopic imaging techniques. However, manual examination and diagnosis with WSIs is time-consuming and tiresome. Recently, deep convolutional neural networks have succeeded in histopathological image analysis. In this paper, we propose a novel cancer texture-based deep neural network (CAT-Net) that learns scalable texture features from histopathological WSIs. The innovation of CAT-Net is twofold: (1) capturing invariant spatial patterns by dilated convolutional layers and (2) Reducing model complexity while improving performance. Moreover, CAT-Net can provide discriminative texture patterns formed on cancerous regions of histopathological …

Proton Exchange Membrane Fuel Cell Stack Design Optimization Using An Improved Jaya Algorithm, Uday K. Chakraborty

Educator Preparation & Leadership Faculty Works

Fuel cell stack configuration optimization is known to be a problem that, in addition to presenting engineering challenges, is computationally hard. This paper presents an improved computational heuristic for solving the problem. The problem addressed in this paper is one of constrained optimization, where the goal is to seek optimal (or near-optimal) values of (i) the number of proton exchange membrane fuel cells (PEMFCs) to be connected in series to form a group, (ii) the number of such groups to be connected in parallel, and (iii) the cell area, such that the PEMFC assembly delivers the rated voltage at the …

Bootbandit: A Macos Bootloader Attack, Armen Boursalian, Mark Stamp

Faculty Publications, Computer Science

Historically, the boot phase on personal computers left systems in a relatively vulnerable state. Because traditional antivirus software runs within the operating system, the boot environment is difficult to protect from malware. Examples of attacks against bootloaders include so‐called “evil maid” attacks, in which an intruder physically obtains a boot disk to install malicious software for obtaining the password used to encrypt a disk. The password then must be stored and retrieved again through physical access. In this paper, we discuss an attack that borrows concepts from the evil maid. We assume exploitation can be used to infect a bootloader …

Emergent Behavior Development And Control In Multi-Agent Systems, David W. King

Theses and Dissertations

Emergence in natural systems is the development of complex behaviors that result from the aggregation of simple agent-to-agent and agent-to-environment interactions. Emergence research intersects with many disciplines such as physics, biology, and ecology and provides a theoretical framework for investigating how order appears to spontaneously arise in complex adaptive systems. In biological systems, emergent behaviors allow simple agents to collectively accomplish multiple tasks in highly dynamic environments; ensuring system survival. These systems all display similar properties: self-organized hierarchies, robustness, adaptability, and decentralized task execution. However, current algorithmic approaches merely present theoretical models without showing how these models actually create hierarchical, …

A Note On Distance-Based Entropy Of Dendrimers, Modjtaba Ghorbani, Matthias Dehmer, Samaneh Zangi, Abbe Mowshowitz, Frank Emmert-Streib

Publications and Research

This paper introduces a variant of entropy measures based on vertex eccentricity and applies it to all graphs representing the isomers of octane. Taking into account the vertex degree as well (degree-ecc-entropy), we find a good correlation with the acentric factor of octane isomers. In particular, we compute the degree-ecc-entropy for three classes of dendrimer graphs.