Matching Items (93)
Filtering by
- Creators: Barrett, The Honors College
- Creators: Computer Science and Engineering Program
- Member of: Barrett, The Honors College Thesis/Creative Project Collection
- Status: Published
Description
Advances in computational processing have made big data analysis in fields like music information retrieval (MIR) possible. Through MIR techniques researchers have been able to study information on a song, its musical parameters, the metadata generated by the song's listeners, and contextual data regarding the artists and listeners (Schedl, 2014). MIR research techniques have been applied within the field of music and emotions research to help analyze the correlative properties between the music information and the emotional output. By pairing methods within music and emotions research with the analysis of the musical features extracted through MIR, researchers have developed predictive models for emotions within a musical piece. This research has increased our understanding of the correlative properties of certain musical features like pitch, timbre, rhythm, dynamics, mel frequency cepstral coefficients (MFCC's), and others, to the emotions evoked by music (Lartillot 2008; Schedl 2014) This understanding of the correlative properties has enabled researchers to generate predictive models of emotion within music based on listeners' emotional response to it. However, robust models that account for a user's individualized emotional experience and the semantic nuances of emotional categorization have eluded the research community (London, 2001). To address these two main issues, more advanced analytical methods have been employed. In this article we will look at two of these more advanced analytical methods, machine learning algorithms and deep learning techniques, and discuss the effect that they have had on music and emotions research (Murthy, 2018). Current trends within MIR research, the application of support vector machines and neural networks, will also be assessed to explain how these methods help to address the two main issues within music and emotion research. Finally, future research within the field of machine and deep learning will be postulated to show how individuate models may be developed from a user or a pool of user's listening libraries. Also how developments of semi-supervised classification models that assess categorization by cluster instead of by nominal data, may be helpful in addressing the nuances of emotional categorization.
ContributorsMcgeehon, Timothy Makoto (Author) / Middleton, James (Thesis director) / Knowles, Kristina (Committee member) / Mechanical and Aerospace Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2018-12
Description
Classical planning is a field of Artificial Intelligence concerned with allowing autonomous agents to make reasonable decisions in complex environments. This work investigates
the application of deep learning and planning techniques, with the aim of constructing generalized plans capable of solving multiple problem instances. We construct a Deep Neural Network that, given an abstract problem state, predicts both (i) the best action to be taken from that state and (ii) the generalized “role” of the object being manipulated. The neural network was tested on two classical planning domains: the blocks world domain and the logistic domain. Results indicate that neural networks are capable of making such
predictions with high accuracy, indicating a promising new framework for approaching generalized planning problems.
the application of deep learning and planning techniques, with the aim of constructing generalized plans capable of solving multiple problem instances. We construct a Deep Neural Network that, given an abstract problem state, predicts both (i) the best action to be taken from that state and (ii) the generalized “role” of the object being manipulated. The neural network was tested on two classical planning domains: the blocks world domain and the logistic domain. Results indicate that neural networks are capable of making such
predictions with high accuracy, indicating a promising new framework for approaching generalized planning problems.
ContributorsNakhleh, Julia Blair (Author) / Srivastava, Siddharth (Thesis director) / Fainekos, Georgios (Committee member) / Computer Science and Engineering Program (Contributor) / School of International Letters and Cultures (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
Rapid advancements in Artificial Intelligence (AI), Machine Learning, and Deep Learning technologies are widening the playing field for automated decision assistants in healthcare. The field of radiology offers a unique platform for this technology due to its repetitive work structure, ability to leverage large data sets, and high position for clinical and social impact. Several technologies in cancer screening, such as Computer Aided Detection (CAD), have broken the barrier of research into reality through successful outcomes with patient data (Morton, Whaley, Brandt, & Amrami, 2006; Patel et al, 2018). Technologies, such as the IBM Medical Sieve, are growing excitement with the potential for increased impact through the addition of medical record information ("Medical Sieve Radiology Grand Challenge", 2018). As the capabilities of automation increase and become a part of expert-decision-making jobs, however, the careful consideration of its integration into human systems is often overlooked. This paper aims to identify how healthcare professionals and system engineers implementing and interacting with automated decision-making aids in Radiology should take bureaucratic, legal, professional, and political accountability concerns into consideration. This Accountability Framework is modeled after Romzek and Dubnick’s (1987) public administration framework and expanded on through an analysis of literature on accountability definitions and examples in military, healthcare, and research sectors. A cohesive understanding of this framework and the human concerns it raises helps drive the questions that, if fully addressed, create the potential for a successful integration and adoption of AI in radiology and ultimately the care environment.
ContributorsGilmore, Emily Anne (Author) / Chiou, Erin (Thesis director) / Wu, Teresa (Committee member) / Industrial, Systems & Operations Engineering Prgm (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
This thesis examines the applications of the Internet of Things and Artificial Intelligence within small-to-medium sized retail businesses. These technologies have become a common aspect of a modern business environment, yet there remains a level of unfamiliarity with these concepts for business owners to fully utilize these tools. The complexity behind IoT and AI has been simplified to provide benefits for a brick and mortar business store in regards to security, logistics, profit optimization, operations, and analytics. While these technologies can contribute to a business’s success, they potentially come with a high and unattainable financial cost. In order to investigate which aspects of businesses can benefit the most from these technologies, interviews with small-to-medium business owners were conducted and paired with an analysis of published research. These interviews provided specific pain points and issues that could potentially be solved by these technologies. The analysis conducted in this thesis gives a detailed summary of this research and provides a business model for two small businesses to optimize their Internet of Things and Artificial Intelligence to solve these pain points, while staying in their financial budget.
ContributorsAldrich, Lauren (Co-author) / Bricker, Danielle (Co-author) / Sebold, Brent (Thesis director) / Vermeer, Brandon (Committee member) / Computer Science and Engineering Program (Contributor) / Department of Information Systems (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
This thesis surveys and analyzes applications of machine learning techniques to the fields of animation and computer graphics. Data-driven techniques utilizing machine learning have in recent years been successfully applied to many subfields of animation and computer graphics. These include, but are not limited to, fluid dynamics, kinematics, and character modeling. I argue that such applications offer significant advantages which will be pivotal in advancing the fields of animation and computer graphics. Further, I argue these advantages are especially relevant in real-time implementations when working with finite computational resources.
ContributorsSaba, Raphael Lucas (Author) / Foy, Joseph (Thesis director) / Olson, Loren (Committee member) / School of Mathematical and Statistical Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
iLieDown - Improved Display Orientation For Handheld Devices Using Convolutional Neural Networks.pdf
Description
91% of smartphone and tablet users experience a problem with their device screen being oriented the wrong way during use [11]. In [11], the authors proposed iRotate, a previous solution which uses computer vision to solve the orientation problem. We propose iLieDown, an improved method of automatically rotating smartphones, tablets, and other device displays. This paper introduces a new algorithm to correctly orient the display relative to the user’s face using a convolutional neural network (CNN). The CNN model is trained to predict the rotation of faces in various environments through data augmentation, uses a confidence threshold, and analyzes multiple images to be accurate and robust. iLieDown is battery and CPU efficient, causes no noticeable lag to the user during use, and is 6x more accurate than iRotate.
ContributorsTallman, Riley Paul (Author) / Yang, Yezhou (Thesis director) / Fang, Zhiyuan (Committee member) / Computer Science and Engineering Program (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2019-12
Description
Surgical site infections do not need to be a common complication in the healthcare field. They can be avoided through the use of surgical site infection prevention bundles. More specifically, the bundles can be personalized to each patient to offer further infection prevention when the patient presents with a higher comorbidity risk. Hospitals could reduce their surgical site infection rates through the use of artificial intelligence combing electronic health records and calculating the Charlson Comorbidity Index (CCI) and American Society of Anesthesiologists (ASA) scores to ultimately form an automatic operating room checklist. Low-risk patients will have a standard primary checklist of interventions. Higher risk patients have additional secondary and tertiary interventions added to their primary checklists.
Through a combination of literature, expert opinion, and various seminars at the APIC (Association for Professionals in Infection Control and Epidemiology), I determined an evidence based primary list of SSI prevention strategies that should be standard amongst all patients. I also gained information on interventions that should be included when patients have higher CCI and ASA scores. My presentation will demonstrate the need for standardization of surgical site infection prevention strategies, the ease that would come from using an artificial intelligence robot to derive the exact intervention checklist best suited for the patient and a cost analysis to demonstrate the current spending and potential savings from using such technology.
Through a combination of literature, expert opinion, and various seminars at the APIC (Association for Professionals in Infection Control and Epidemiology), I determined an evidence based primary list of SSI prevention strategies that should be standard amongst all patients. I also gained information on interventions that should be included when patients have higher CCI and ASA scores. My presentation will demonstrate the need for standardization of surgical site infection prevention strategies, the ease that would come from using an artificial intelligence robot to derive the exact intervention checklist best suited for the patient and a cost analysis to demonstrate the current spending and potential savings from using such technology.
ContributorsDelp, Meredith Diann (Author) / Dirksen, Shannon (Thesis director) / Lalley, Cathy (Committee member) / Edson College of Nursing and Health Innovation (Contributor) / Barrett, The Honors College (Contributor)
Created2019-12
Description
Background: Natural Language Processing models have been trained to locate questions and answers in forum settings before but on topics such as cancer and diabetes. Also, studies have used filtering methods to understand themes in forum settings regarding opioid use. However, studies have not been conducted regarding training an NLP model to locate the questions people addicted to opioids are asking their peers and the answers they are receiving in forums. There are a variety of annotation tools available to help aid the data collection to train NLP models. For academic purposes, brat is the best tool for this purpose. This study will inform clinical practice by indicating what the inner thoughts of their patients who are addicted to opioids are so that they will be able to have more meaningful conversations during appointments that the patient may be too afraid to start.
Methods: The standard NLP process was used for this study in which a gold standard was reached through matched paired annotations of the forum text in brat and a neural network was trained on the content. Following the annotation process, adjudication occurred to increase the inter-annotator agreement. Categories were developed by local physicians to describe the questions and three pilots were run to test the best way to categorize the questions.
Results: The inter-annotator agreement, calculated via F-score, before adjudication for a 0.7 threshold was 0.378 for the annotation activity. After adjudication at a threshold of 0.7, the inter-annotator agreement increased to 0.560. Pilots 1, 2, and 3 of the categorization activity had an inter-annotator agreement of 0.375, 0.5, and 0.966 respectively.
Discussion: The inter-annotator agreement of the annotation activity may have been low initially since the annotators were students who may have not been as invested in the project as necessary to accurately annotate the text. Also, as everyone interprets the text slightly differently, it is possible that that contributed to the differences in the matched pairs’ annotations. The F-score variation for the categorization activity partially had to do with different delivery systems of the instructions and partially with the area of study of the participants. The first pilot did not mandate the use of the original context located in brat and the instructions were provided in the form of a downloadable document. The participants were computer science graduate students. The second pilot also had the instructions delivered via a document, but it was strongly suggested that the context be used to gain an understanding of the questions’ meanings. The participants were also computer science graduate students who upon a discussion of their results after the pilot expressed that they did not have a good understanding of the medical jargon in the posts. The final pilot used a combination of students with and without medical background, required to use the context, and included verbal instructions in combination with the written ones. The combination of these factors increased the F-score significantly. For a full-scale experiment, students with a medical background should be used to categorize the questions.
Methods: The standard NLP process was used for this study in which a gold standard was reached through matched paired annotations of the forum text in brat and a neural network was trained on the content. Following the annotation process, adjudication occurred to increase the inter-annotator agreement. Categories were developed by local physicians to describe the questions and three pilots were run to test the best way to categorize the questions.
Results: The inter-annotator agreement, calculated via F-score, before adjudication for a 0.7 threshold was 0.378 for the annotation activity. After adjudication at a threshold of 0.7, the inter-annotator agreement increased to 0.560. Pilots 1, 2, and 3 of the categorization activity had an inter-annotator agreement of 0.375, 0.5, and 0.966 respectively.
Discussion: The inter-annotator agreement of the annotation activity may have been low initially since the annotators were students who may have not been as invested in the project as necessary to accurately annotate the text. Also, as everyone interprets the text slightly differently, it is possible that that contributed to the differences in the matched pairs’ annotations. The F-score variation for the categorization activity partially had to do with different delivery systems of the instructions and partially with the area of study of the participants. The first pilot did not mandate the use of the original context located in brat and the instructions were provided in the form of a downloadable document. The participants were computer science graduate students. The second pilot also had the instructions delivered via a document, but it was strongly suggested that the context be used to gain an understanding of the questions’ meanings. The participants were also computer science graduate students who upon a discussion of their results after the pilot expressed that they did not have a good understanding of the medical jargon in the posts. The final pilot used a combination of students with and without medical background, required to use the context, and included verbal instructions in combination with the written ones. The combination of these factors increased the F-score significantly. For a full-scale experiment, students with a medical background should be used to categorize the questions.
ContributorsPawlik, Katie (Author) / Devarakonda, Murthy (Thesis director) / Murcko, Anita (Committee member) / Green, Ellen (Committee member) / College of Health Solutions (Contributor) / Barrett, The Honors College (Contributor)
Created2019-12
Description
The focus of this research paper is understanding the impacts of human factors on the technology innovations in automobiles and the direction our society is headed. There will be an assessment of our current state and the possible solutions to combat the issue of creating technology advancements for automobiles that cater towards the human factors. There will be an introduction on the history of the first automobile invented to provide an understanding of the what the first automobile consisted of and will continue discussing the technological innovations that were implemented due to human factors. Diving into the types of technological innovations such as the ignition system, car radio, the power steering system, and self-driving, it will show the progression of the technological advancements that was implemented in relation to the human factors that was prominent among society. From there, it is important to understand what human factors and the concept of human factor engineering are. It will provide a better understanding of why humans have created technology in relation to the human factors. Then, there will be an introduction of the mobile phone industry history/timeline as a comparison to show the impacts of how human factors have had on the development of the technology in mobile phones and how heavily it catered towards human factors. There will be a discussion of the 3 key human factors that have been catered towards the development and implementation of technology in automobiles. They are selecting the path that requires the least cognitive effort, overestimating the performance of technology, and reducing the attention due to an automated system being put into place. Lastly, is understanding that if we create or implement technology such as self-driving, it should not solely be for comfort and ease of use, but for the overall efficient use of transportation in the future. This way humans would not rely heavily too much on the technology and limit the effect that human factors have on us.
ContributorsParham, Gi-onli (Author) / Keane, Katy (Thesis director) / Collins, Gregory (Committee member) / Department of Supply Chain Management (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05
Description
This thesis serves as a baseline for the potential for prediction through machine learning (ML) in baseball. Hopefully, it also will serve as motivation for future work to expand and reach the potential of sabermetrics, advanced Statcast data and machine learning. The problem this thesis attempts to solve is predicting the outcome of a pitch. Given proper pitch data and situational data, is it possible to predict the result or outcome of a pitch? The result or outcome refers to the specific outcome of a pitch, beyond ball or strike, but if the hitter puts the ball in play for a double, this thesis shows how I attempted to predict that type of outcome. Before diving into my methods, I take a deep look into sabermetrics, advanced statistics and the history of the two in Major League Baseball. After this, I describe my implemented machine learning experiment. First, I found a dataset that is suitable for training a pitch prediction model, I then analyzed the features and used some feature engineering to select a set of 16 features, and finally, I trained and tested a pair of ML models on the data. I used a decision tree classifier and random forest classifier to test the data. I attempted to us a long short-term memory to improve my score, but came up short. Each classifier performed at around 60% accuracy. I also experimented using a neural network approach with a long short-term memory (LSTM) model, but this approach requires more feature engineering to beat the simpler classifiers. In this thesis, I show examples of five hitters that I test the models on and the accuracy for each hitter. This work shows promise that advanced classification models (likely requiring more feature engineering) can provide even better prediction outcomes, perhaps with 70% accuracy or higher! There is much potential for future work and to improve on this thesis, mainly through the proper construction of a neural network, more in-depth feature analysis/selection/extraction, and data visualization.
ContributorsGoodman, Avi (Author) / Bryan, Chris (Thesis director) / Hsiao, Sharon (Committee member) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05