Physical Sciences and Mathematics Commons^™

Impact Of Movements On Facial Expression Recognition, Zhebin Yin

Honors Theses

The ability to recognize human emotions can be a useful skill for robots. Emotion recognition can help robots understand our responses to robot movements and actions. Human emotions can be recognized through facial expressions. Facial Expression Recognition (FER) is a well-established research area, how- ever, the majority of prior research is based on static datasets of images. With robots often the subject is moving, the robot is moving, or both. The purpose of this research is to determine the impact of movement on facial expression recognition. We apply a pre-existing model for FER, which performs around 70.86% on a given …

Tracking Xenophobic Terminology On Twitter Using Nlp, Harper Lyon

Honors Theses

Social media is a major driver of political thought, with platforms like Facebook, Twitter, and TikTok having a massive impact on how people think and vote. For this reason we should take seriously any large shifts in the language used to describe issues or groups on social media, as these are likely to either denote a change in political thought or even forecast the same. Of particular interest, given the international reach of social media, is the way that discussions around foreign relations and immigration play out. In the United States of America online spaces have become the default space …

Corrective Feedback Timing In Kanji Writing Instruction Apps, Phoenix Mulgrew

Honors Theses

The focus of this research paper is to determine the correct time to provide corrective feedback to people who are learning how to write Japanese kanji. To do this, we developed a system that is able to recognize Japanese kanji that is handwritten onto an iPad screen and check for errors such as wrong stroke order. Previous research has achieved success in developing similar systems, but this project is unique because the research question involves the timing of corrective feedback. In particular, we are looking at whether immediate or delayed corrective feedback results in better learning.

Iot Garden Frost Alarm, Andrew James

Honors Theses

Home gardeners are faced with yearly challenges due to spring frosts harming young plants. This is frequently mitigated by covering crops with frost blankets, but only on nights when a frost is predicted. In areas with less predictable climate, an unexpected frost can kill vulnerable plants, reducing the amount of food produced. A system is proposed and designed here to use internet of things (IoT) technology to enable a small weather station in the home garden to report current climate data and predict frosts, then alert the gardener in time for them to cover their plants.

The system as designed …

The Future Of Artificial Intelligence In The Healthcare Industry, Erika Bonnist

Honors Theses

Technology has played an immense role in the evolution of healthcare delivery for the United States and on an international scale. Today, perhaps no innovation offers more potential than artificial intelligence. Utilizing machine intelligence as opposed to human intelligence for the purposes of planning, offering solutions, and providing insights, AI has the ability to alter traditional dynamics between doctors, patients, and administrators; this reality is now producing both elation at artificial intelligence's medical promise and uncertainty regarding its capacity in current systems. Nevertheless, current trends reveal that interest in AI among healthcare stakeholders is continuously increasing, and with the current …

Effects Of Bounding Tree Complexity On Immersion In A Virtual Environment, Jonathan Caputo

Honors Theses

Bounding boxes are volumetric spaces in a 3D environment defined by a set of points that denote the area of collision for a given object. Bounding trees are known as a collection of bounding boxes. Simple bounding trees contain few bounding boxes and complex bounding trees contain many. Our research aims to find the relationship between the complexity of a bounding tree and a user's immersion in a 3D virtual environment. We created an algorithm that used the bounding trees to check for when an object collides with another. Using this algorithm, we ran a user study where users would …

Human Facial Emotion Recognition System In A Real-Time, Mobile Setting, Claire Williamson

Honors Theses

The purpose of this project was to implement a human facial emotion recognition system in a real-time, mobile setting. There are many aspects of daily life that can be improved with a system like this, like security, technology and safety.

There were three main design requirements for this project. The first was to get an accuracy rate of 70%, which must remain consistent for people with various distinguishing facial features. The second goal was to have one execution of the system take no longer than half of a second to keep it as close to real time as possible. Lastly, …

Preventing Browser Fingerprinting By Randomizing Canvas, Rianna Quiogue

Honors Theses

Whether users know it or not, their online behaviors are being tracked and stored by many of the websites they visit regularly through a technique called browser fingerprinting. Just like a person's physical fingerprint can identify them, users' browser fingerprints can identify them on the Internet. This thesis outlines the techniques used in browser fingerprinting and explains how although it can be used for good, it can also be a major threat to people's online privacy and security. Since browser fingerprinting has gained popularity among many websites and advertising companies, researchers have been developing ways to counteract its effectiveness by …

Exploring The Behavior Repertoire Of A Wireless Vibrationally Actuated Tensegrity Robot, Zongliang Ji

Honors Theses

Soft robotics is an emerging field of research due to its potential to explore and operate in unstructured, rugged, and dynamic environments. However, the properties that make soft robots compelling also make them difficult to robustly control. Here at Union, we developed the world’s first wireless soft tensegrity robot. The goal of my thesis is to explore effective and efficient methods to explore the diverse behavior our tensegrity robot. We will achieve that by applying state-of-art machine learning technique and a novelty search algorithm.

Iterated Local Search Algorithms For Bike Route Generation, Aidan Pieper

Honors Theses

Planning routes for recreational cyclists is challenging because they prefer longer more scenic routes, not the shortest one. This problem can be modeled as an instance of the Arc Orienteering Problem (AOP), a known NP-Hard optimization problem. Because no known algorithms exist to solve this optimization problem efficiently, we solve the AOP using heuristic algorithms which trade accuracy for speed. We implement and evaluate two different Iterated Local Search (ILS) heuristic algorithms using an open source routing engine called GraphHopper and the OpenStreetMap data set. We propose ILS variants which our experimental results show can produce better routes at the …

2018 Ieee Signal Processing Cup: Forensic Camera Model Identification Challenge, Michael Geiger

Honors Theses

The goal of this Senior Capstone Project was to lead Union College’s first ever Signal Processing Cup Team to compete in IEEE’s 2018 Signal Processing Cup Competition. This year’s competition was a forensic camera model identification challenge and was divided into two separate stages of competition: Open Competition and Final Competition. Participation in the Open Competition was open to any teams of undergraduate students, but the Final Competition was only open to the three finalists from Open Competition and is scheduled to be held at ICASSP 2018 in Calgary, Alberta, Canada. Teams that make it to the Final Competition will …

Optimizing Tensegrity Gaits Using Bayesian Optimization, James Boggs

Honors Theses

We design and implement a new, modular, more complex tensegrity robot featuring data collection and wireless communication and operation as well as necessary accompanying research infrastructure. We then utilize this new tensegrity to assess previous research on using Bayesian optimization to generate effective forward gaits for tensegrity robots. Ultimately, we affirm the conclusions of previous researchers, demonstrating that Bayesian optimization is statistically significantly (p < 0:05) more effective at discovering useful gaits than random search. We also identify several flaws in our new system and identify means of addressing them, paving the way for more effective future research.

A Better Way To Construct Tensegrities: Planar Embeddings Inform Tensegrity Assembly, Elizabeth Anne Ricci

Honors Theses

Although seemingly simple, tensegrity structures are complex in nature which makes them both ideal for use in robotics and difficult to construct. We work to develop a protocol for constructing tensegrities more easily. We consider attaching a tensegrity's springs to the appropriate locations on some planar arrangement of attached struts. Once all of the elements of the structure are connected, we release the struts and allow the tensegrity to find its equilibrium position. This will allow for more rapid tensegrity construction. We develop a black-box that given some tensegrity returns a flat-pack, or the information needed to perform this physical …

Attracting Human Attention Using Robotic Facial Expressions And Gestures, Venus Yu

Honors Theses

Robots will soon interact with humans in settings outside of a lab. Since it will be likely that their bodies will not be as developed as their programming, they will not have the complex limbs needed to perform simple tasks. Thus they will need to seek human assistance by asking them for help appropriately. But how will these robots know how to act? This research will focus on the specific nonverbal behaviors a robot could use to attract someone’s attention and convince them to interact with the robot. In particular, it will need the correct facial expressions and gestures to …

Effective Ann Topologies For Use As Genotypes For Evaluating Design And Fabrication, John R. Peterson

Honors Theses

There is promise in the field of Evolutionary Design for systems that evolve not only what to manufacture but also how to manufacture it. EvoFab is a system that uses Genetic Algorithms to evolve Artificial Neural Networks (ANNs) which control a modified 3d-printer with the goal of automating some level of invention. ANNs are an obvious choice for use with a system like this as they are canonically evolvable encodings, and have been successfully used as evolved control systems in Evolutionary Robotics. However, there is little known about how the structural characteristics of an ANN affect the shapes that can …

Elliptic Curve Cryptology, Francis Rocco

Honors Theses

In today's digital age of conducting large portions of daily life over the Internet, privacy in communication is challenged extremely frequently and confidential information has become a valuable commodity. Even with the use of commonly employed encryption practices, private information is often revealed to attackers. This issue motivates the discussion of cryptology, the study of confidential transmissions over insecure channels, which is divided into two branches of cryptography and cryptanalysis. In this paper, we will first develop a foundation to understand cryptography and send confidential transmissions among mutual parties. Next, we will provide an expository analysis of elliptic curves and …

Accurately Simulating The Battle Of Thermopylae To Analyze "What If" Scenarios, Josh Wasserman

Honors Theses

The Battle of Thermopylae (480 BCE) was a last ditch effort to stall the Persian army as it marched south toward Athens. Led by Leonidas and his personal guard of 300 Spartans, a citizen army of Greeks was able to delay a Persian army of over 100,000 soldiers at the town of Thermopylae for several days. Although the Greeks were ultimately defeated at Thermopylae, the battle provided enough time for the Greek states to regroup and plan a counter attack, eventually defeating the invading Persians. This battle was crucial not only for the preservation of Greek independence, but also the …

Multi-Agent Simulation Of The Battle Of Ankara, 1402, Ruili Tang

Honors Theses

In 1402, at the north of city Ankara, Turkey, a battle between Ottoman Empire and Tamerlane Empire decided the fate of Europe and Asia. Although historians largely agree on the general battle procedure, the details are still open to dispute. Several factors may have contributed to the Ottoman defeat, such as the overwhelming size of Tamerlanes army, poisoned water, the tactical formations of the military units, and betrayal by the Tartar cavalry in the Ottoman left wing. The approach is divided into two stages: the simulation stage, which provides data to analyze the complex interactions of autonomous agents, and the …

Effect Of Deictic Gestures On Direction-Giving In Virtual Humans, Anthony Pham

Honors Theses

Virtual agents are animated characters that use speech and gesture to interact with human users. They can serve as an intuitive interface for a variety of purposes. I am investigating the use of deictic gestures by a direction-giving agent. Deictic gestures are pointing gestures that humans often use in direction-giving to help clarify the route and destination. In my experiment, I developed a virtual agent to give directions to people to six different locations with the following spatial relationships to the starting point: left, right, left behind, right behind, left up, and right up. Three versions of the virtual agent …

A Graph Based Departmental Spoken Dialogue System, Julia Isaac

Honors Theses

Spoken dialogue systems are automatic, computer based systems that are a great way for people to receive important information. In this project, I created a spoken dialogue system that people can use to learn about the Computer Science Department at Union College. The system was built by populating an open source dialogue system using a graph based dialogue manager. I improved upon a previous working dialogue system by making the conversations sound more natural, improving the flexibility of the system and making the system more robust. To help with this process a corpus was created using about 200 different dialogues …

Blending Two Automatic Playlist Generation Algorithms, James Curbow

Honors Theses

We blend two existing automatic playlist generation algorithms. One algorithm is built to smoothly transition between a start song and an end song (Start-End). The other infers song similarity based on adjacent occurrences in expertly authored streams (EAS). First, we seek to establish the effectiveness of the Start-End algorithm using the EAS algorithm to determine song similarity, then we propose two playlist generation algorithms of our own: the Unbiased Random Walk (URW) and the Biased Random Walk (BRW). Like the Start-End algorithm, both the URW algorithm and BRW algorithm transition between a start song and an end song; however, issues …

Computing The (Un)Computable: A Computationally-Augmented Perspective On The Yasukuni Shrine Controversy, Ryan Muther

Honors Theses

Computational methods have been used with increasing frequency in the social sciences and humanities, due to the availability of digital sources and computing power to study everything from changes in the meanings of words in Latin texts to how knowledge was categorized in eighteen century encyclopedias. Recent trends in the fields of digital humanities and computational social science include statistical methods like machine learning, requiring large pre-tagged and annotated sets of documents which in turn necessitates a great deal of prior work to create data to use with such methods. This reliance on large corpora of annotated data limits the …

Approaching Humans For Help: A Study Of Human-Robot Proxemics, Eric Rose

Honors Theses

In order for a robot to be effective when interacting with a person, it is important for the robot to choose the correct person. Consider an example where a robot is trying to perform a task but it isn’t capable of doing a subtask, like going up a flight of stairs. In this case, the robot would need to ask a person for help with the elevator, in a socially appropriate way. We have conducted an experiment to determine who would be the best candidate to approach in a situation like this. Should the robot choose to approach someone who …

The Effects Of Early Confidence Interval Training On User Efficacy In A P300 Brain-Computer Interface Spelling Task, Adam Starkman

Honors Theses

Brain-computer interface (BCI) technology can provide communication for individuals suffering from degenerative neuromuscular disorders. The present study sought to demonstrate improved BCI performance in healthy individuals using confidence interval training with a P300 BCI spelling program. In this BCI interface, electroencephalographic (EEG) activity was recorded as participants attended to a specific target character within a matrix of flashing letters and numbers presented on a computer screen. The BCI uses the P300 Event Related Potential to select the intended character. In a prior patient case, use of a confidence measure that rejected questionable selections improved that user’s spelling efficiency. The present …

What It Is To Be Conscious: Exploring The Plasibility Of Consciousness In Deep Learning Computers, Peter Davis

Honors Theses

As artificial intelligence and robotics progress further and faster every day, designing and building a conscious computer appears to be on the horizon. Recent technological advances have allowed engineers and computer scientists to create robots and computer programs that were previously impossible. The development of these highly sophisticated robots and AI programs has thus prompted the age-old question: can a computer be conscious? The answer relies on addressing two key sub-problems. The first is the nature of consciousness: what constitutes a system as conscious, or what properties does consciousness have? Secondly, does the physical make-up of the robot or computer …

Classifying System Call Traces Using Anomalous Detection, William Doyle

Honors Theses

We used data mining techniques to detect intrusions among system call traces and have outlined our results. Recent work at the intersection of security and machine learning has lead to better understanding of anomalous intrusion detection. There is a need to more thoroughly understand how anomaly detection can be used because of its potential applications and advantages over current standard methods. In this thesis, we report on a new approach of anomalous detection using system call traces. Our goal is to be able to create a system that can accurately detect hacking attacks by analyzing the sequences of system calls …

Analysis Of The Peerrank Method For Peer Grading, Joshua Kline

Honors Theses

Peer grading can have many benefits in education, including a reduction in the time instructors spend grading and an opportunity for students to learn through their analysis of others work. However, when not handled properly, peer grading can be unreliable and may result in grades that are vastly different from those which a student truly deserves. Therefore, any peer grading system used in a classroom must consider the potential for graders to generate inaccurate grades. One such system is the PeerRank rule proposed by Toby Walsh, which uses an iterative, linear algebra based process reminiscent of the Google PageRank algorithm …

Evolving Soft Robots With Vibration Based Movement, Andrew Danise

Honors Theses

Creating effective designs for soft robots is extremely difficult due to the large number of different possibilities for shape, material properties, and movement mechanisms. Due to the lack of methods to design soft robots, previous research has used evolutionary algorithms to tackle this problem of overwhelming options. A popular technique is to use generative encodings to create designs using evolutionary algorithms because of their modularity and ability to induce large scale coordinated change. The main drawback of generative encodings is that it is difficult to know where along the ontogenic trajectory resides the phenotype with the highest fitness. The two …

Aggressive Monetization: Why The Pay For Currency Model Is Dominating The Ios App Store Today, Theodore Kitchen

Honors Theses

The Apple iOS App Store has only been around for 5 years, and yet it has completely changed the way that mobile software is distributed. In this brief period, the online marketplace has seen dramatic shifts in the most successful strategies used by iOS software developers and, more specifically, game developers to gain revenue. As of March 14th 2014 fifteen of the twenty top-grossing iOS apps feature some form of in-app currency that users may purchase with real money, eighteen are mobile video games, and all twenty of these apps are free to download. This paper explores a new business …

Cellular Automata And Music: A New Representation, Richard French

Honors Theses

For millenia, we’ve thought of musical composition as a purely human activity. However, we once also thought of an activity like chess to be purely human, but Deep Blue was able to defeat Kasparov in 1995 all the same. Could there perhaps be some tool or algorithm for musical composition that can replicate to some extent what human beings can do with music? This project explores this idea through the use of a tool called a cellular automaton. A cellular automaton is a grid space with a finite number of states for each of the ”cells” or ”squares” where a …