Computer Sciences Commons^™

Leveraging Auxiliary Tasks For Document-Level Cross-Domain Sentiment Classification, Jianfei Yu, Jing Jiang

Research Collection School Of Computing and Information Systems

In this paper, we study domain adaptationwith a state-of-the-art hierarchicalneural network for document-level sentimentclassification. We first design a newauxiliary task based on sentiment scoresof domain-independent words. We thenpropose two neural network architecturesto respectively induce document embeddingsand sentence embeddings that workwell for different domains. When thesedocument and sentence embeddings areused for sentiment classification, we findthat with both pseudo and external sentimentlexicons, our proposed methods canperform similarly to or better than severalhighly competitive domain adaptationmethods on a benchmark dataset of productreviews.

Efficient Gate System Operations For A Multipurpose Port Using Simulation Optimization, Ketki Kulkarni, Trong Khiem Tran, Hai Wang, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

Port capacity is determined by three major infrastructural resources namely, berths, yards and gates. Theadvertised capacity is constrained by the least of the capacities of the three resources. While a lot ofattention has been paid to optimizing berth and yard capacities, not much attention has been given toanalyzing the gate capacity. The gates are a key node between the land-side and sea-side operations in anocean-to-cities value chain. The gate system under consideration, located at an important port in an Asiancity, is a multi-class parallel queuing system with non-homogeneous Poisson arrivals. It is hard to obtaina closed form analytic approach for …

Btci: A New Framework For Identifying Congestion Cascades Using Bus Trajectory Data, Meng-Fen Chiang, Ee Peng Lim, Wang-Chien Lee, Agus Trisnajaya Kwee

Research Collection School Of Computing and Information Systems

The knowledge of traffic health status is essential to the general public and urban traffic management. To identify congestion cascades, an important phenomenon of traffic health, we propose a Bus Trajectory based Congestion Identification (BTCI) framework that explores the anomalous traffic health status and structure properties of congestion cascades using bus trajectory data. BTCI consists of two main steps, congested segment extraction and congestion cascades identification. The former constructs path speed models from historical vehicle transitions and design a non-parametric Kernel Density Estimation (KDE) function to derive a measure of congestion score. The latter aggregates congested segments (i.e., those with …

Inferring Social Media Users’ Demographics From Profile Pictures: A Face++ Analysis On Twitter Users, Soon-Gyo Jung, Jisun An, Haewoon Kwak, Joni Salminen, Bernard J. Jansen

Research Collection School Of Computing and Information Systems

In this research, we evaluate the applicability of using facial recognition of social media account profile pictures to infer the demographic attributes of gender, race, and age of the account owners leveraging a commercial and well-known image service, specifically Face++. Our goal is to determine the feasibility of this approach for actual system implementation. Using a dataset of approximately 10,000 Twitter profile pictures, we use Face++ to classify this set of images for gender, race, and age. We determine that about 30% of these profile pictures contain identifiable images of people using the current state-of-the-art automated means. We then employ …

Approaches, Techniques, And Tools For Identifying Important Code Changes To Help Code Reviewers, Maneesh M. Mohanavilasam

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Software development is a collaborative process where many developers come together and work on a project. To make things easy and manageable, software is developed on a version control system. A version control system is a centralized system which stores code and adds code from all other developers as an increment to the code base in the repository. Since multiple people work on the same code repository together, it is important to make sure that their contributions do not conflict with each other. It is important to maintain the quality and integrity of the repository. This is where the code …

Web Application For Graduate Course Recommendation System, Sayali Dhumal

Electronic Theses, Projects, and Dissertations

The main aim of the course advising system is to build a course recommendation path for students to help them plan courses to successfully graduate on time. The recommendation path displays the list of courses a student can take in each quarter from the first quarter after admission until the graduation quarter. The courses are filtered as per the student’s interest obtained from a questionnaire asked to the student.

The business logic involves building the recommendation algorithm. Also, the application is functionality-tested end-to-end by using nightwatch.js which is built on top of node.js. Test cases are written for every module …

Web Application For Graduate Course Advising System, Sanjay Karrolla

Electronic Theses, Projects, and Dissertations

The main aim of the course recommendation system is to build a course recommendation path for students to help them plan courses to successfully graduate on time. The Model-View-Controller (MVC) architecture is used to isolate the user interface (UI) design from the business logic. The front-end of the application develops the UI using AngularJS. The front-end design is done by gathering the functionality system requirements -- input controls, navigational components, informational components and containers and usability testing. The back-end of the application involves setting up the database and server-side routing. Server-side routing is done using Express JS.

Graph-Based Latent Embedding, Annotation And Representation Learning In Neural Networks For Semi-Supervised And Unsupervised Settings, Ismail Ozsel Kilinc

USF Tampa Graduate Theses and Dissertations

Machine learning has been immensely successful in supervised learning with outstanding examples in major industrial applications such as voice and image recognition. Following these developments, the most recent research has now begun to focus primarily on algorithms which can exploit very large sets of unlabeled examples to reduce the amount of manually labeled data required for existing models to perform well. In this dissertation, we propose graph-based latent embedding/annotation/representation learning techniques in neural networks tailored for semi-supervised and unsupervised learning problems. Specifically, we propose a novel regularization technique called Graph-based Activity Regularization (GAR) and a novel output layer modification called …

Using Data To Improve Services For Infants With Hearing Loss: Linking Newborn Hearing Screening Records With Early Intervention Records, Maria Gonzalez, Lori Iarossi, Yan Wu, Ying Huang, Kirsten Siegenthaler

Journal of Early Hearing Detection and Intervention

The purpose of this study was to match records of infants with permanent hearing loss from the New York Early Hearing Detection and Intervention Information System (NYEHDI-IS) to records of infants with permanent hearing loss receiving early intervention services from the New York State Early Intervention Program (NYSEIP) to identify areas in the state where hearing screening, diagnostic evaluations and referrals to the NYSEIP were not being made or documented in a timely manner. Data from 2014-2016 NYEHDI-IS and NYEIS information systems were matched using The Link King. There were 274 infants documented in NYEIS Information System as receiving early …

A Churn For The Better: Localizing Censorship Using Network-Level Path Churn And Network Tomography, Shinyoung Cho, Abbas Razaghpanah, Rishab Nithyanand, Phillipa Gill

Computer Science: Faculty Publications

Recent years have seen the Internet become a key vehicle for citizens around the globe to express political opinions and organize protests. This fact has not gone unnoticed, with countries around the world repurposing network management tools (e.g., URL iltering products) and protocols (e.g., BGP, DNS) for censorship. Previous work has focused on identifying how censorship is performed. However, there is no major studies to identify, at a global scale, the networks responsible for performing censorship. Also, repurposing network products for censorship can have unintended international impact, which we refer to as lcensorship leakagež. While there have been anecdotal reports …

Feasibility Of Using Virtual Reality In Requirements Elicitation Process, Aman Bhimani

Master of Science in Software Engineering Theses

Contemporary Virtual Reality (VR) technologies offer an increasing number of functionalities including head-mounted displays (HMD), haptic and sound feedback, as well as motion tracking. This gives us the opportunity to leverage the immersive power offered by these technologies in the context of requirements elicitation, especially to surface those requirements that cannot be expressed via traditional techniques such as interviews and focus groups. The goal of this thesis is to survey uses of VR in requirements engineering, and to describe a method of elicitation using VR as a tool.

To validate the methodology, a research plan is developed with a strong …

The Graph Database: Jack Of All Trades Or Just Not Sql?, George F. Hurlburt, Maria R. Lee, George K. Thiruvathukal

Computer Science: Faculty Publications and Other Works

This special issue of IT Professional focuses on the graph database. The graph database, a relatively new phenomenon, is well suited to the burgeoning information era in which we are increasingly becoming immersed. Here, the guest editors briefly explain how a graph database works, its relation to the relational database management system (RDBMS), and its quantitative and qualitative pros and cons, including how graph databases can be harnessed in a hybrid environment. They also survey the excellent articles submitted for this special issue.

Implementation Guidelines For Green Data Centres, Abdallah Tubaishat, Sara Sumaidaa

All Works

Increased reliance on technology and online transactions has increased the heat generated in data centres, due to greater access, storage, aggregation and analysis of such data. This paper discusses the downsides of traditional data centres and highlights the importance of applying green IT practices. It explores the benefits and proposes guidelines on shifting to green data centres. The recommendations to go green are to: a) reduce energy by applying green IT practices, b) eliminate redundancies in server systems and cooling modules, c) turn on power management tools on servers and terminals when applicable, d) utilise newer technologies of power consumption …

Examining The Electrical Excitation, Calcium Signaling, And Mechanical Contraction Cycle In A Heart Cell, Kristen Deetz, Nygel Foster, Darius Leftwich, Chad Meyer, Shalin Patel, Carlos Barajas, Matthias K. Gobbert, Zana Coulibaly

Spora: A Journal of Biomathematics

As the leading cause of death in the United States, heart disease has become a principal concern in modern society. Cardiac arrhythmias can be caused by a dysregulation of calcium dynamics in cardiomyocytes. Calcium dysregulation, however, is not yet fully understood and is not easily predicted; this provides motivation for the subsequent research. Excitation-contraction coupling (ECC) is the process through which cardiomyocytes undergo contraction from an action potential. Calcium induced calcium release (CICR) is the mechanism through which electrical excitation is coupled with mechanical contraction through calcium signaling. The study of the interplay between electrical excitation, calcium signaling, and mechanical …

Modular Mechanistic Networks: On Bridging Mechanistic And Phenomenological Models With Deep Neural Networks In Natural Language Processing, Simon Dobnik, John D. Kelleher

Books/Book chapters

Natural language processing (NLP) can be done using either top-down (theory driven) and bottom-up (data driven) approaches, which we call mechanistic and phenomenological respectively. The approaches are frequently considered to stand in opposition to each other. Examining some recent approaches in deep learning we argue that deep neural networks incorporate both perspectives and, furthermore, that leveraging this aspect of deep learning may help in solving complex problems within language technology, such as modelling language and perception in the domain of spatial cognition.

Tracking You Through Dns Traffic: Linking User Sessions By Clustering With Dirichlet Mixture Model, Mingxuan Sun, Junjie Zhang, Guangyue Xu, Dae Wook Kim

Computer Science and Engineering Faculty Publications

The Domain Name System (DNS), which does not encrypt domain names such as "bank.us" and "dentalcare.com", commonly accurately reflects the specific network services. Therefore, DNS-based behavioral analysis is extremely attractive for many applications such as forensics investigation and online advertisement. Traditionally, a user can be trivially and uniquely identified by the device’s IP address if it is static (i.e., a desktop or a laptop). As more and more wireless and mobile devices are deeply ingrained in our lives and the dynamic IP address such as DHCP has been widely applied, it becomes almost impossible to use one IP address to …

What Is Not Where: The Challenge Of Integrating Spatial Representations Into Deep Learning Architectures, John D. Kelleher, Simon Dobnik

Books/Book chapters

This paper examines to what degree current deep learning architectures for image caption generation capture spatial lan- guage. On the basis of the evaluation of examples of generated captions from the literature we argue that systems capture what objects are in the image data but not where these objects are located: the cap- tions generated by these systems are the output of a language model conditioned on the output of an object detector that cannot capture fine-grained location information. Although language models provide useful knowledge for image captions, we argue that deep learning image captioning architectures should also model geometric …

A Spatial Collaboration: Building A Multi-Institution Geospatial Data Discovery Portal, Mara Blake, Karen Majewicz, Ryan Mattke, Kathleen W. Weessies

Collaborative Librarianship

As academic education and research increasingly take advantage of geospatial data and methodologies, we see a corresponding exponential growth in the number of available geospatial resources in the form of GIS datasets and scanned historical maps. However, users can experience difficulty finding these resources due to the unconnected multitude of platforms and clearinghouses that host them. Additionally, the resources are not always well described with web semantic metadata that facilitates discovery. In response to this challenge, The Big Ten Academic Alliance Geospatial Data Project began in 2015 to provide discoverability, facilitate access, and connect scholars to geospatial resources. Our project …

Building The Ascend Siu Web Application, Lee Cooper, Cody Lingle, Ren Jing, Hallie Martin, Matt Gross, Nancy Martin

ASA Multidisciplinary Research Symposium

This poster represents a year-long web application development project completed by undergraduate students in the Information Systems Technologies department at Southern Illinois University. The web application was developed for the Ascend registered student organization using Agile Scrum software engineering methods.

Reengineering A 2-Tier Database Application With Software Architecture, Hong G. Jung

ASA Multidisciplinary Research Symposium

The purpose of this research is to demonstrate how to reengineer a legacy Database Application using to a target system with MVC and 3-layered architecture. A Coffee Inventory Management database application is used for legacy application. The benefits of the reengineering are discussed.

Software Reengineering: Reverse Engineering With Using 4+1 Architectural Views And Forward Engineering With Mvc Architecture, Shane Mueller

ASA Multidisciplinary Research Symposium

As software complexity is increasing exponentially in our modern era, software architecture becomes increasingly important. The separation of concerns through architecture allows each programmer the opportunity to limit their need of understanding to only the portion of code for which they are responsible for, thus saving large amounts of time.

Wearete: A Scalable Wearable E-Textile Triboelectric Energy Harvesting System For Human Motion Scavenging, Xian Li, Ye Sun

Michigan Tech Publications, Part 1

In this paper, we report the design, experimental validation and application of a scalable, wearable e-textile triboelectric energy harvesting (WearETE) system for scavenging energy from activities of daily living. The WearETE system features ultra-low-cost material and manufacturing methods, high accessibility, and high feasibility for powering wearable sensors and electronics. The foam and e-textile are used as the two active tribomaterials for energy harvester design with the consideration of flexibility and wearability. A calibration platform is also developed to quantify the input mechanical power and power efficiency. The performance of the WearETE system for human motion scavenging is validated and calibrated …

Context-Based Human Activity Recognition Using Multimodal Wearable Sensors, Pratool Bharti

USF Tampa Graduate Theses and Dissertations

In the past decade, Human Activity Recognition (HAR) has been an important part of the regular day to day life of many people. Activity recognition has wide applications in the field of health care, remote monitoring of elders, sports, biometric authentication, e-commerce and more. Each HAR application needs a unique approach to provide solutions driven by the context of the problem. In this dissertation, we are primarily discussing two application of HAR in different contexts. First, we design a novel approach for in-home, fine-grained activity recognition using multimodal wearable sensors on multiple body positions, along with very small Bluetooth beacons …

Efficient Pebble Game Algorithms Engineered For Protein Rigidity Applications, Mojtaba Nouri Bygi, Ileana Streinu

Computer Science: Faculty Publications

Pebble game rigidity analysis is an efficient method for extracting rigidity and flexibility information of biomolecules without performing costly molecular dynamics simulations. The standard algorithm works on a multi-graph associated to a mechanical model constructed from an arbitrary atom-bond network. Motivated by large scale protein flexibility and simulated unfolding applications, we have developed a faster and more robust variation tailored to the specificities of bio-polymers. We describe this new phased pebble game as implemented in the new version of our software Kinari-2.

Efficient Pebble Game Algorithms Engineered For Protein Rigidity Applications, Mojtaba Nouri Bygi, Ileana Streinu

Computer Science: Faculty Publications

Pebble game rigidity analysis is an efficient method for extracting rigidity and flexibility information of biomolecules without performing costly molecular dynamics simulations. The standard algorithm works on a multi-graph associated to a mechanical model constructed from an arbitrary atom-bond network. Motivated by large scale protein flexibility and simulated unfolding applications, we have developed a faster and more robust variation tailored to the specificities of bio-polymers. We describe this new phased pebble game as implemented in the new version of our software Kinari-2.

An Advanced Private Social Activity Invitation Framework With Friendship Protection, Weitian Tong, Lei Chen, Scott Buglass, Weinan Gao

Information Technology: Faculty Publications

Due to the popularity of social networks and human-carried/human-affiliated devices with sensing abilities, like smartphones and smart wearable devices, a novel application was necessitated recently to organize group activities by learning historical data gathered from smart devices and choosing invitees carefully based on their personal interests. We proposed a private and efficient social activity invitation framework. Our main contributions are (1) defining a novel friendship to reduce the communication/update cost within the social network and enhance the privacy guarantee at the same time; () designing a strong privacy-preserving algorithm for graph publication, which addresses an open concern proposed recently; () …

Switching From A Semi-Computerized To An Online Employment Application System: A Case Study, Deanna House

Information Systems and Quantitative Analysis Faculty Publications

This case explores the switch from semi-computerized to an online employment application system. This case documents the struggles experienced with user expectations related to requirements and how customization of a third-party product derailed the project’s success. The project was eventually implemented, but not without a significant development effort to customize. The end project was over budget, over time, and did not have all of the functionality that the users were expecting. Key factors leading to the project failure were: lack of user involvement and participation throughout the project including documentation of requirements for the target system, lack of a dedicated …

A Brief Overview Of Intelligent Mobility Management For Future Wireless Mobile Networks, Ilsun You, Yuh-Shyan Chen, Sherali Zeadally, Fei Song

Information Science Faculty Publications

No abstract provided.

Bikemate: Bike Riding Behavior Monitoring With Smartphones, Weixi Gu, Zimu Zhou, Yuxun Zhou, Han Zou, Yunxin Liu, Costas J. Spanos, Lin Zhang

Research Collection School Of Computing and Information Systems

Detecting dangerous riding behaviors is of great importance to improve bicycling safety. Existing bike safety precautionary measures rely on dedicated infrastructures that incur high installation costs. In this work, we propose BikeMate, a ubiquitous bicycling behavior monitoring system with smartphones. BikeMate invokes smartphone sensors to infer dangerous riding behaviors including lane weaving, standing pedalling and wrong-way riding. For easy adoption, BikeMate leverages transfer learning to reduce the overhead of training models for different users, and applies crowdsourcing to infer legal riding directions without prior knowledge. Experiments with 12 participants show that BikeMate achieves an overall accuracy of 86.8% for lane …

The Birds Of A Feather Research Challenge, Todd W. Neller

Computer Science Faculty Publications

Neller presented a set of research challenges for undergraduates that allow an excellent formative experience of research, writing, peer review, and potential presentation and publication through a top-tier conference. The focus problem is the analysis of a newly-designed solitaire card game, Birds of a Feather, so potentials for discovery abound. Open access talk slides, research code, solvability data sets, research tutorial videos, and more are also available at http://cs.gettysburg.edu/~tneller/puzzles/boaf .