Physical Sciences and Mathematics Commons^™

Synthesizing Aspect-Driven Recommendation Explanations From Reviews, Trung-Hoang Le, Hady W. Lauw

Research Collection School Of Computing and Information Systems

Explanations help to make sense of recommendations, increasing the likelihood of adoption. However, existing approaches to explainable recommendations tend to rely on rigid, standardized templates, customized only via fill-in-the-blank aspect sentiments. For more flexible, literate, and varied explanations covering various aspects of interest, we synthesize an explanation by selecting snippets from reviews, while optimizing for representativeness and coherence. To fit target users' aspect preferences, we contextualize the opinions based on a compatible explainable recommendation model. Experiments on datasets of several product categories showcase the efficacies of our method as compared to baselines based on templates, review summarization, selection, and text …

Health-Aware Food Planner: A Personalized Recipe Generation Approach Based On Gpt-2, Bushra Aljbawi

Theses and Dissertations (Comprehensive)

"What to eat today?" With the flourish of Internet, more and more people nowadays are inclined to find an answer to this most problematic question online. The recent explosion of food networks; however, produces large volumes of recipes, making it even harder to make an informed decision. This yields the need for advanced decision-making algorithms and efficient recommendation systems. Conventional recommender systems are not feasible anymore as food is a complicated feature that presents unique challenges and is less studied. For example, it can be one of the main reasons for obesity and many other chronic diseases. Food recommender system …

How Machine Learning And Probability Concepts Can Improve Nba Player Evaluation, Harrison Miller

CMC Senior Theses

In this paper I will be breaking down a scholarly article, written by Sameer K. Deshpande and Shane T. Jensen, that proposed a new method to evaluate NBA players. The NBA is the highest level professional basketball league in America and stands for the National Basketball Association. They proposed to build a model that would result in how NBA players impact their teams chances of winning a game, using machine learning and probability concepts. I preface that by diving into these concepts and their mathematical backgrounds. These concepts include building a linear model using ordinary least squares method, the bias …

Multimodal Fusion Strategies For Outcome Prediction In Stroke, Esra Zihni, John D. Kelleher, Vince I. Madai, Ahmed Khalil, Ivana Galinovic, Jochen Fiebach, Michelle Livne, Dietmar Frey

Conference papers

Data driven methods are increasingly being adopted in the medical domain for clinical predictive modeling. Prediction of stroke outcome using machine learning could provide a decision support system for physicians to assist them in patient-oriented diagnosis and treatment. While patient-specific clinical parameters play an important role in outcome prediction, a multimodal fusion approach that integrates neuroimaging with clinical data has the potential to improve accuracy. This paper addresses two research questions: (a) does multimodal fusion aid in the prediction of stroke outcome, and (b) what fusion strategy is more suitable for the task at hand. The baselines for our experimental …

Modelling Interleaved Activities Using Language Models, Eoin Rogers, Robert J. Ross, John D. Kelleher

Conference papers

We propose a new approach to activity discovery, based on the neural language modelling of streaming sensor events. Our approach proceeds in multiple stages: we build binary links between activities using probability distributions generated by a neural language model trained on the dataset, and combine the binary links to produce complex activities. We then use the activities as sensor events, allowing us to build complex hierarchies of activities. We put an emphasis on dealing with interleaving, which represents a major challenge for many existing activity discovery systems. The system is tested on a realistic dataset, demonstrating it as a promising …

Mutual Information Decay Curves And Hyper-Parameter Grid Search Design For Recurrent Neural Architectures, Abhijit Mahalunkar, John Kelleher

Conference papers

We present an approach to design the grid searches for hyper-parameter optimization for recurrent neural architectures. The basis for this approach is the use of mutual information to analyze long distance dependencies (LDDs) within a dataset. We also report a set of experiments that demonstrate how using this approach, we obtain state-of-the-art results for DilatedRNNs across a range of benchmark datasets.

A Machine Learning Approach To Estimate The Annihilation Photon Interactions Inside The Scintillator Of A Pet Scanner, Sai Akhil Bharthavarapu

Graduate Theses, Dissertations, and Problem Reports

Biochemical processes are chemical processes that occur in living organisms. They can be studied with nuclear medicine through the help of radioactive tracers. Based on the radioisotope used, the photons that are emitted from the body tissue are either detected by single-photon emission computed tomography (SPECT) or by positron emission tomography (PET) scanners. SPECT uses gamma rays as tracer but gives a weaker contrast and spatial resolution compared to a PET scanner which uses positrons as tracer. PET scans show the metabolic changes occurring at the cellular level in an organ or a tissue. This detection is important because diseases …

Speech Mode Classification Using The Fusion Of Cnns And Lstm Networks, Pratyusha Chowdary Vakkantula

Graduate Theses, Dissertations, and Problem Reports

Speech mode classification is an area that has not been as widely explored in the field of sound classification as others such as environmental sounds, music genre, and speaker identification. But what is speech mode? While mode is defined as the way or the manner in which something occurs or is expressed or done, speech mode is defined as the style in which the speech is delivered by a person.

There are some reports on speech mode classification using conventional methods, such as whispering and talking using a normal phonetic sound. However, to the best of our knowledge, deep learning-based …

A Study On Real-Time Database Technology And Its Applications, Geethmi Nimantha Dissanayake

Masters Theses

No abstract provided.

Data Governance And The Emerging University, Michael J. Madison

Book Chapters

Knowledge and information governance questions are tractable primarily in institutional terms, rather than in terms of abstractions such as knowledge itself or individual or social interests. This chapter offers the modern research university as an example. Practices of data-intensive research by university-based researchers, sometimes reduced to the popular phrase “Big Data,” pose governance challenges for the university. The chapter situates those challenges in the traditional understanding of the university as an institution for understanding forms and flows of knowledge. At a broad level, the chapter argues that the new salience of data exposes emerging shifts in the social, cultural, and …

Experiments On The Neural Network Approach To The Handwritten Digit Classification Problem, William Meissner

Electronic Theses and Dissertations

When the MNIST dataset was introduced in 1998, training a network was a multiple week problem in order to receive results far less accurate than an average CPU can produce within a couple of hours today. While this indicates that training a network on such a dataset is not the complicated problem it may have been twenty years ago, the MNIST dataset makes a good tool for study and testing with beginner and medium complexity neural networks. This paper follows along with the work presented in the online textbook “Neural Networks and Deep Learning” by Michael Nielson and an updated …

Continuous Deployment Transitions At Scale, Laurie Williams, Kent Beck, Jeffrey Creasey, Andrew Glover, James Holman, Jez Humble, David Mclaughlin, John Thomas Micco, Brendan Murphy, Jason A. Cox, Vishnu Pendyala, Steven Place, Zachary T. Pritchard, Chuck Rossi, Tony Savor, Michael Stumm, Chris Parnin

Faculty Research, Scholarly, and Creative Activity

Predictable, rapid, and data-driven feature rollout; lightning-fast; and automated fix deployment are some of the benefits most large software organizations worldwide are striving for. In the process, they are transitioning toward the use of continuous deployment practices. Continuous deployment enables companies to make hundreds or thousands of software changes to live computing infrastructure every day while maintaining service to millions of customers. Such ultra-fast changes create a new reality in software development. Over the past four years, the Continuous Deployment Summit, hosted at Facebook, Netflix, Google, and Twitter has been held. Representatives from companies like Cisco, Facebook, Google, IBM, Microsoft, …

Evolution Of Integration, Build, Test, And Release Engineering Into Devops And To Devsecops, Vishnu Pendyala

Faculty Research, Scholarly, and Creative Activity

Software engineering operations in large organizations are primarily comprised of integrating code from multiple branches, building, testing the build, and releasing it. Agile and related methodologies accelerated the software development activities. Realizing the importance of the development and operations teams working closely with each other, the set of practices that automated the engineering processes of software development evolved into DevOps, signifying the close collaboration of both development and operations teams. With the advent of cloud computing and the opening up of firewalls, the security aspects of software started moving into the applications leading to DevSecOps. This chapter traces the journey …

Sediment Dynamics In The Magdalena River Basin, Colombia: Implications For Understanding Tropical River Processes And Hydropower Development, Luke H. Fisher

Graduate Student Theses, Dissertations, & Professional Papers

The Magdalena River Basin of Colombia has a globally relevant sediment flux, however, studies of the sediment regime in the basin are limited in scope. This knowledge gap limits application of understanding of sediment dynamics to hydropower decision making. To close this gap, we implemented a sediment budget framework to quantify the impacts of hydropower development in a 118,000 km² portion of the Magdalena River basin. We informed this framework with analysis of background erosion rates derived from ¹⁰Be cosmogenic nuclides and modern sediment fluxes derived from monitoring and optical remote sensing. We standardized these data to spatially …