Physical Sciences and Mathematics Commons^™

Cnn-Lstm Vs Ann: Option Pricing Theory, Edward Chang

Undergraduate Student Research Internships Conference

The modern derivatives market has been steadily growing since the development of the first accurate option pricing model by Fischer Black, Robert Merton, and Myron Scholes. Since then, there have been many different approaches to more accurately price options like the binomial option pricing model and approaches using technology such as machine learning. There are many different research papers on option pricing with artificial neural networks (“ANN”) but not many with other neural network types. We contribute to the existing literature by developing a convolutional neural network – long short-term memory (“CNN-LSTM”) model to price options and compare it to …

Reporting Standards For Machine Learning Research In Type 2 Diabetes, Grace Kang

Undergraduate Student Research Internships Conference

In this project, three people scored 90 papers on machine learning predictive models for type 2 diabetes to assess their adherence to TRIPOD, MI-CLAIM, and DOME reporting guidelines.

A Qualitative Look Into Repair Practices, Jumana Labib

Undergraduate Student Research Internships Conference

This research poster is based on a working research paper which moves beyond the traditional scope of repair and examines the Right to Repair movement from a smaller, more personal lens by detailing the 6 categorical impediments as dubbed by Dr. Alissa Centivany (design, law, economic/business strategy, material asymmetry, informational asymmetry, and social impediments) have continuously inhibited repair and affected repair practices, which has consequently had larger implications (environmental, economic, social, etc.) on ourselves, our objects, and our world. The poster builds upon my research from last year (see "The Right to Repair: (Re)building a better future"), this time pulling …

Model Transformations Between Sequence Diagram And Activity Diagram With Qvto, Yutong Xia

Undergraduate Student Research Internships Conference

Complex software systems are specified by various models denoting the behavior of the system components, the exchanges of messages and data among components, the intents of the system stakeholders, the flow of system processes, and the structure of the system as a collection of modules.

When such systems are maintained and evolved (e.g. by adding new functionality, fixing bugs, or porting to a new operating environment), one or more of these models are altered. This brings the system specification to an inconsistent state since some models reflect the new behavior while other models were not appropriately evolved.

This research presents …

Damage Assessment In Aging Structures Using Augmented Reality, Omar Zuhair Awadallah, Ayan Sadhu

Undergraduate Student Research Internships Conference

Structural Health Monitoring (SHM) is the assessment of bridges and observation of data regarding these bridges over time to monitor their evolution and detect the presence of any possible damages. However, existing methods to perform structural inspections in bridges are high in cost, time-consuming and risky. Inspectors use expensive equipment to reach a certain area of the bridge to inspect it, and at different heights, this can pose a risk to the inspector’s safety. This study aims to find cheaper, faster, and safer ways to perform structural inspections using augmented reality and artificial intelligence. The developed system uses a machine …

How Can Social Networks Impact Careers In Game Development?, Tongzhang Wang

Undergraduate Student Research Internships Conference

The game development industry is one characterized by young workers and fast worker turnover. The popularity of using twitter as a professional tool within the video games industry presents a potentially insightful view port into a professional's informal network. Investigating characteristics of the social networks of newly graduated students in the game development industry may reveal what factors contribute to fast worker turnover and how certain cohorts may face additional barriers.

Data Preprocessing For Machine Learning Modules, Rawan El Moghrabi

Undergraduate Student Research Internships Conference

Data preprocessing is an essential step when building machine learning solutions. It significantly impacts the success of machine learning modules and the output of these algorithms. Typically, data preprocessing is made-up of data sanitization, feature engineering, normalization, and transformation. This paper outlines the data preprocessing methodology implemented for a data-driven predictive maintenance solution. The above-mentioned project entails acquiring historical electrical data from industrial assets and creating a health index indicating each asset's remaining useful life. This solution is built using machine learning algorithms and requires several data processing steps to increase the solution's accuracy and efficiency. In this project, the …

A Kuramoto Model Approach To Predicting Chaotic Systems With Echo State Networks, Sophie Wu, Jackson Howe

Undergraduate Student Research Internships Conference

An Echo State Network (ESN) with an activation function based on the Kuramoto model (Kuramoto ESN) is implemented, which can successfully predict the logistic map for a non-trivial number of time steps. The reservoir in the prediction stage exhibits binary dynamics when a good prediction is made, but the oscillators in the reservoir display a larger variability in states as the ESN’s prediction becomes worse. Analytical approaches to quantify how the Kuramoto ESN’s dynamics relate to its prediction are explored, as well as how the dynamics of the Kuramoto ESN relate to another widely studied physical model, the Ising model.

Simulating Salience: Developing A Model Of Choice In The Visual Coordination Game, Adib Sedig

Undergraduate Student Research Internships Conference

This project is primarily inspired by three papers: Colin Camerer and Xiaomin Li’s (2019 working paper)—Using Visual Salience in Empirical Game Theory, Ryan Oprea’s (2020)—What Makes a Rule Complex?, and Caplin et. al.’s (2011)—Search and Satisficing. Over the summer, I worked towards constructing a model of choice for the visual coordination game that can model player behavior more accurately than traditional game theoretic predictions. It attempts to do so by incorporating a degree of bias towards salience into a cellular automaton search algorithm and utilizing it alongside a sequential search mechanism of satisficing. This …

Simulating 129-Xe Hyperpolarization, Jacob F. Abiad

Undergraduate Student Research Internships Conference

Hyperpolarized 129-Xe is an important resource in many fields of medical physics and MRI research. The physics of the efficient production of hyperpolarized 129-Xe is therefore equally worth investigation. The main process of hyperpolarizing 129-Xe is Spin Exchange Optical Pumping (SEOP) and is dependent on several physical factors that can be difficult to constantly change in a lab setting. Physical modelling of 129-Xe hyperpolarization allows for the more efficient testing of hyperpolarization physics in a wide array of experimental setups to better determine the optimal values for hyperpolarization. This research project attempted to create a working model for 129-Xe hyperpolarization …

Automated Extraction Of Key Words And Abstract, Oluwadarasimi Temitope Ogunshote Mr.

Undergraduate Student Research Internships Conference

An application/program that automates the extraction of key words and abstracts from documents.

Searching For New Relations Among The Eilenberg-Zilber Maps, Owen T. Abma

Undergraduate Student Research Internships Conference

The goal of this project was to write a computer program that would aid in the search for relations among the Eilenberg-Zilber maps, which relate to simplicial objects in algebraic topology. This presentation outlines the process of writing this program, the challenges faced along the way, and the final results of the project.

Contemporary Mathematical Approaches To Computability Theory, Luis Guilherme Mazzali De Almeida

Undergraduate Student Research Internships Conference

In this paper, I present an introduction to computability theory and adopt contemporary mathematical definitions of computable numbers and computable functions to prove important theorems in computability theory. I start by exploring the history of computability theory, as well as Turing Machines, undecidability, partial recursive functions, computable numbers, and computable real functions. I then prove important theorems in computability theory, such that the computable numbers form a field and that the computable real functions are continuous.

Monitoring At-Home Care Patients Through A Scalar Polar Plot Visualization Of Motion Sensor Data, Michael Mcgavin

Undergraduate Student Research Internships Conference

In Canada, approximately 18 percent (6.6 million) of the total population are age 65 or older, and 88 percent of people over age 65 want to stay in their residence for as long as possible. This older demographic is a group that is dependent on proactive and preventative healthcare. Using motion sensor data collected from a local company providing home-care services to this demographic, a data visualization was constructed to assist users in observing patient behavior and improving their quality of life while maintaining their independence. However, since the collected data is time-based, it results in a dataset that is …

Evaluating Machine Learning Model Stability For Software Bug Prediction, Joud El-Shawa

Undergraduate Student Research Internships Conference

Large software systems are implemented using many different programming languages and scripts, and consequently the dependencies between their components are very complex. It is therefore difficult to extract and understand these dependencies by solely analyzing the source code, so that failure risks can be detected accurately. On the other hand, it is a common practice for software engineers to keep track of process related metrics such as the number of times a component was maintained, with which other components it has been co-committed, whether the maintenance activity was a bug-fixing activity, and how many lines of source code have been …

Numerical Simulation Of Adaptive Metabolic Response To Anti-Angiogenic Treatment In Renal Cell Carcinoma, Saranya Varakunan

Undergraduate Student Research Internships Conference

Renal cell carcinoma, a malignant kidney cancer, is often treated using anti-angiogenic drugs to prevent the growth of blood vessels within the tumour. Although tumours initially respond to this treatment, they eventually develop resistance. This resistance is hypothesized to be caused by a switch to a symbiotic metabolism that allows cells to survive even with a low blood supply.

This project seeks to computationally model the transport of oxygen, lactate, and glucose within a tumour in order to examine how cancer metabolism adapts to changes in blood vessels.