Other Computer Engineering Commons^™

Bert For Question Answering On Bioasq, Eric R. Fu, Rikel Djoko, Maysam Mansor, Robert Slater

SMU Data Science Review

Machine reading comprehension and question answering are topics of considerable focus in the field of Natural Language Processing (NLP). In recent years, language models like Bidirectional Encoder Representations from Transformers (BERT) [3] have been very successful in language related tasks like question answering. The difficulty of the question answering task lies in developing accurate representations of language and being able to produce answers for questions. In this study, the focus is to investigate how to train and fine tune a BERT model to improve its performance on BioASQ, a challenge on large scale biomedical question answering. Our most accurate BERT …

Qlime-A Quadratic Local Interpretable Model-Agnostic Explanation Approach, Steven Bramhall, Hayley Horn, Michael Tieu, Nibhrat Lohia

SMU Data Science Review

In this paper, we introduce a proof of concept that addresses the assumption and limitation of linear local boundaries by Local Interpretable Model-Agnostic Explanations (LIME), a popular technique used to add interpretability and explainability to black box models. LIME is a versatile explainer capable of handling different types of data and models. At the local level, LIME creates a linear relationship for a given prediction through generated sample points to present feature importance. We redefine the linear relationships presented by LIME as quadratic relationships and expand its flexibility in non-linear cases and improve the accuracy of feature interpretations. We coin …

Identifying Customer Churn In After-Market Operations Using Machine Learning Algorithms, Vitaly Briker, Richard Farrow, William Trevino, Brent Allen

SMU Data Science Review

This paper presents a comparative study on machine learning methods as they are applied to product associations, future purchase predictions, and predictions of customer churn in aftermarket operations. Association rules are used help to identify patterns across products and find correlations in customer purchase behaviour. Studying customer behaviour as it pertains to Recency, Frequency, and Monetary Value (RFM) helps inform customer segmentation and identifies customers with propensity to churn. Lastly, Flowserve’s customer purchase history enables the establishment of churn thresholds for each customer group and assists in constructing a model to predict future churners. The aim of this model is …

Aws Ec2 Instance Spot Price Forecasting Using Lstm Networks, Jeffrey Lancon, Yejur Kunwar, David Stroud, Monnie Mcgee, Robert Slater

SMU Data Science Review

Cloud computing is a network of remote computing resources hosted on the Internet that allow users to utilize cloud resources on demand. As such, it represents a paradigm shift in the way businesses and industries think about digital infrastructure. With the shift from IT resources being a capital expenditure to a managed service, companies must rethink how they approach utilizing and optimizing these resources in order to maximize productivity and minimize costs. With proper resource management, cloud resources can be instrumental in reducing computing expenses.

Cloud resources are perishable commodities; therefore, cloud service providers have developed strategies to maximize utilization …

Self-Driving Cars: Evaluation Of Deep Learning Techniques For Object Detection In Different Driving Conditions, Ramesh Simhambhatla, Kevin Okiah, Shravan Kuchkula, Robert Slater

SMU Data Science Review

Deep Learning has revolutionized Computer Vision, and it is the core technology behind capabilities of a self-driving car. Convolutional Neural Networks (CNNs) are at the heart of this deep learning revolution for improving the task of object detection. A number of successful object detection systems have been proposed in recent years that are based on CNNs. In this paper, an empirical evaluation of three recent meta-architectures: SSD (Single Shot multi-box Detector), R-CNN (Region-based CNN) and R-FCN (Region-based Fully Convolutional Networks) was conducted to measure how fast and accurate they are in identifying objects on the road, such as vehicles, pedestrians, …

Finding Truth In Fake News: Reverse Plagiarism And Other Models Of Classification, Matthew Przybyla, David Tran, Amber Whelpley, Daniel W. Engels

SMU Data Science Review

As the digital age creates new ways of spreading news, fake stories are propagated to widen audiences. A majority of people obtain both fake and truthful news without knowing which is which. There is not currently a reliable and efficient method to identify “fake news”. Several ways of detecting fake news have been produced, but the various algorithms have low accuracy of detection and the definition of what makes a news item ‘fake’ remains unclear. In this paper, we propose a new method of detecting on of fake news through comparison to other news items on the same topic, as …

Comparative Study Of Sentiment Analysis With Product Reviews Using Machine Learning And Lexicon-Based Approaches, Heidi Nguyen, Aravind Veluchamy, Mamadou Diop, Rashed Iqbal

SMU Data Science Review

In this paper, we present a comparative study of text sentiment classification models using term frequency inverse document frequency vectorization in both supervised machine learning and lexicon-based techniques. There have been multiple promising machine learning and lexicon-based techniques, but the relative goodness of each approach on specific types of problems is not well understood. In order to offer researchers comprehensive insights, we compare a total of six algorithms to each other. The three machine learning algorithms are: Logistic Regression (LR), Support Vector Machine (SVM), and Gradient Boosting. The three lexicon-based algorithms are: Valence Aware Dictionary and Sentiment Reasoner (VADER), Pattern, …

How Much Privacy Do We Have Today? A Study Of The Life Of Marc Mezvinsky, Miguel Mares, Salomon Gilles, Brian D. Gobran, Dan Engels

SMU Data Science Review

In this paper, we present a case study evaluating the level of information available about an individual through public, Internet-accessible sources. Privacy is a basic tenet of democratic society, but technological advances have made access to information and the identification of individuals much easier through Internet-accessible databases and information stores. To determine the potential level of privacy available to an individual in today’s interconnected world, we sought to develop a detailed history of Marc Mezvinsky, a semi-public figure, husband of Chelsea Clinton, and son of two former members of the United States House of Representatives. By utilizing only publicly and …

Comparative Study Of Deep Learning Models For Network Intrusion Detection, Brian Lee, Sandhya Amaresh, Clifford Green, Daniel Engels

SMU Data Science Review

In this paper, we present a comparative evaluation of deep learning approaches to network intrusion detection. A Network Intrusion Detection System (NIDS) is a critical component of every Internet connected system due to likely attacks from both external and internal sources. A NIDS is used to detect network born attacks such as Denial of Service (DoS) attacks, malware replication, and intruders that are operating within the system. Multiple deep learning approaches have been proposed for intrusion detection systems. We evaluate three models, a vanilla deep neural net (DNN), self-taught learning (STL) approach, and Recurrent Neural Network (RNN) based Long Short …