FrogCrew is a comprehensive web-based system designed to simplify the management of TCU Athletics sports broadcasting crews. Traditional manual methods of scheduling, tracking availability, and assigning roles are inefficient and prone to errors. This often leads to miscommunication and scheduling conflicts. To solve these challenges, FrogCrew provides a unified platform for administrators. It enables them to manage game schedules, assign roles based on availability and qualifications, and automate notifications efficiently. Key features include customizable crew member profiles. These profiles allow users to update essential information such as availability, roles, and qualifications. The system also offers an automated scheduling tool that simplifies the process of creating game schedules and assigning roles. Additionally, FrogCrew includes a shift exchange feature. This feature allows crew members to request shift swaps, with automated notifications sent to administrators for approval. The system's reporting tools provide financial reports, position-specific insights, and individual performance analyses. These tools help administrators assess crew utilization and manage costs effectively. By automating core functions, FrogCrew reduces manual workload and minimizes errors. It also improves communication between administrators and crew members, ensuring optimal staffing - ultimately enhancing the execution of our TCU sporting events; Go Frogs!

Wearable smart devices, which continuously capture physiological signals such as heart rate, respiratory patterns, and blood oxygen levels, offer significant potential for the early detection of serious health conditions. Timely diagnosis of diseases such as arrhythmia and sleep apnea can greatly enhance patient outcomes by enabling early intervention. However, extensive collection of diverse, real-world wearable sensor data faces challenges due to privacy concerns, data scarcity, and logistical constraints. This research introduces a novel deep learning framework that integrates publicly available wearable sensor data with synthetic physiological signals generated by large language models (LLMs) to create comprehensive and privacy-compliant hybrid datasets.The proposed framework leverages convolutional neural networks (CNNs), optimized for time-series data analysis, alongside advanced machine learning techniques to identify early signs of arrhythmia, sleep apnea, and related health conditions from physiological data. The integration of synthetic data generated by LLMs addresses critical challenges of limited data availability and privacy concerns, enriching the training datasets with diverse scenarios and physiological variations. Preliminary experimental results demonstrate that the hybrid approach, combining publicly accessible wearable sensor data and LLM-generated synthetic signals, significantly enhances the model's accuracy, generalization capability, and resilience to data variability. Models trained on hybrid datasets consistently outperform those relying solely on real-world data, suggesting that synthetic data provides meaningful supplementation to traditional datasets. This study further highlights how synthetic physiological data can enhance the scalability and efficacy of AI-based health monitoring systems, reducing dependency on extensive clinical data collection. By exploring and validating this innovative data synthesis approach, the research contributes significantly to developing more effective, accessible, and proactive healthcare diagnostic tools, ultimately advancing AI-driven solutions in preventive healthcare.

The iPELiNT project develops an AI-powered patent analysis dashboard designed to streamline the patent prosecution process for attorneys and practitioners. This web application leverages modern technologies including Vue.js with Nuxt3 framework for frontend development, NodeJS with Express for backend services, MongoDB for database management, and integrates AI models from OpenAI to analyze patent documents.

The system features a user-friendly dashboard that allows practitioners to upload patent applications, analyze document health, view CPC prediction analytics, examine keyword relevance, and identify potential prior art conflicts. Key functionality includes document parsing, automated health checks, Art Unit prediction, and generation of actionable reports. The solution also incorporates user account management, notification systems, and specialized document generation tools.

Development followed an iterative process with clearly defined milestones and tasks distributed across team members. The project addresses a critical need in the patent industry by providing an all-in-one platform that simplifies complex patent analysis, replacing traditionally fragmented and cumbersome tools with a streamlined, intuitive interface.

The completed iPELiNT dashboard enhances efficiency for patent professionals, improving application quality through AI-powered insights, and ultimately streamlining the patent prosecution workflow with modern design principles and cutting-edge technology.

Designed to empower students with transparent, real-time insights, this innovative digital platform provides comprehensive reviews of classes and instructors, enabling informed academic decision-making. It aggregates detailed evaluations of course content, teaching effectiveness, workload, and overall classroom experience, offering a dynamic alternative to traditional end-of-semester surveys that frequently deliver delayed or insufficient feedback. Backed by survey research underscoring the vital role of timely, honest assessments in shaping successful academic journeys, the platform bridges the gap between institutional data and the practical needs of students. Its intuitive, user-friendly interface allows seamless navigation through a wealth of peer-generated feedback, making it easier for students to select courses that align with their educational goals and personal learning styles. Moreover, by establishing a constructive feedback loop, it provides educators with actionable insights to refine their teaching methods and foster continuous improvement. This collaborative environment not only enhances individual learning experiences but also contributes to building a more effective, accountable educational community. Through open dialogue and shared knowledge, the platform drives positive change, promoting excellence and ensuring that every academic decision is supported by reliable, student-centered information. By continuously evolving based on extensive user feedback, the platform remains dedicated to advancing educational quality and student success.

Public property tax data is often presented in raw formats, making it difficult for the average user to interpret. Our client initially developed a product that provided access to Kern County property tax information only. Our project enhances accessibility by developing ParcelSearch.com, a platform that centralizes property tax data. With this rebranded system, we have expanded coverage to include Kern, Monterey, San Luis Obispo, and Tulare Counties, with plans for further expansion. Users can create accounts and choose from various subscription plans to conduct property searches using multiple search criteria: owner name, parcel number, and legal descriptions. With the development of a user-friendly interface and expanded search functionalities, the platform caters to realtors, investors, and homeowners seeking property insights. This system was built using modern web technologies, including Vue.js for the frontend, Java and Spring Boot for the backend, and PostgreSQL for database management, to name a few. Future plans include expanding nationwide to create an all-encompassing and user-friendly property data platform.

Academic advising presents significant challenges in both time management and complexity. Currently, students navigate between two advising options: generic online resources and personalized consultations with professors and advisors. While personalized advisement offers tailored advice, professors cannot be expected to meet with every undergraduate in their major, especially as enrollment grows, and academic advisors may lack specialized knowledge required for some majors. Echelon addresses this gap by creating a middle ground between generic and personalized advising, offering students an effective supplement and saving time for all parties involved. Echelon functions as an intelligent chatbot assistant powered by large language models such as Llama 3 and Mistral. Upon signup, students share their academic records, enabling Echelon to create individualized profiles that consider key factors such as major/minor selection and performance in critical courses. The project is being built using TypeScript and Rust with Svelte and Axum frameworks, respectively. Echelon utilizes PostgreSQL for user account and conversation storage and Qdrant for vector storage and retrieval. Designed with flexibility in mind, Echelon can be deployed at any university, given basic institutional information such as course catalogs and degree requirements.

FWDX BioBlade is a web-based system designed to automate genetic sequence comparison for Fort Worth Diagnostics (FWDX), a company specializing in high-quality diagnostic reagents. FWDX faces a significant challenge: ensuring its reagents remain effective as pathogens mutate over time. Currently, this process is manual, time-intensive, and costly, relying on external bioinformatic consultants to compare existing reagent sequences against national and international genetic databases like NCBI and GISAID.

BioBlade improves this workflow by automating sequence comparisons, detecting mutations or deletions, and generating real-time reports. This automation significantly reduces turnaround time, improves accuracy, and lowers costs, empowering FWDX scientists and regulatory personnel with timely and accurate information. Key features include:
- Automated sequence analysis for efficient reagent validation
- Customizable query intervals for up-to-date comparisons
- A user-friendly reporting dashboard for streamlined decision-making

As artificial intelligence and machine learning continue to evolve, the need for efficient search and retrieval mechanisms for unstructured data has grown exponentially. Traditional relational databases, optimized for structured queries, struggle with the high-dimensional nature of modern AI-generated embeddings. This challenge has led to the rise of vector databases, specialized systems designed to store, index, and retrieve data based on similarity rather than exact matching.

This symposium explores the fundamental concepts of vector databases, their key components—such as vector embeddings, indexing techniques, and similarity search algorithms—and their advantages over traditional databases. We discuss how vector search operates using distance metrics like cosine similarity and Euclidean distance and compare the roles of vector databases and standalone vector indexes.

Timely and accurate disease prediction is crucial for effective public health response and outbreak mitigation. This project develops a predictive analytics model to forecast the incidence of diseases like measles, rubella, and hepatitis in a specific state all over the US. The model integrates historical epidemiological data with environmental factors such as temperature, humidity, and precipitation, which were collected through web scraping. Using machine learning techniques, the system analyzes patterns and generates forecasts to assist health officials in proactive decision-making. A key focus of this project is ensuring interpretability and accessibility for non-technical users by incorporating data visualization and user-friendly reporting mechanisms. By bridging data science and public health, this project aims to enhance outbreak preparedness and response strategies.

Trailspur Data Project Abstract
Background/Introduction – Our client, Trailspur Capital Partners, is a real estate investment company based in Texas. We assist the company by building a database about commercial / industrial real estate to manage the market more easily and better decision-making.
Objective – The business requires both the Geographic data from the County’s officials and the properties listings with vacancies information. Our goal is to design a database that can handle the aggregate data coming from both sources, which includes arranging and categorizing the properties, coming with several built-in functions namely identifying listings / vacancy changes, before deploying everything to the server.
Methods – Our frontend, built with Vite and Vue, provides a smooth and interactive user experience while on the backend, we utilize AWS Lambda with Python to automate essential tasks, including downloading official county appraisal data, performing spatial merges using GIS functions, and managing our Supabase database.
Results – After successfully aggregating real estate data from both sources into a structured database, which enables easier tracking of property status changes, the platform efficiently processes and visualizes real-time property listings, allowing our client to analyze market trends and make data-driven investment decisions.
Conclusion – Our project enhances real estate market intelligence for Trailspur Capital Partners. The system’s automated functions minimize manual workload and improve the accuracy of property tracking, providing a scalable solution for future expansion.

Cognitive Behavioral Therapy often relies on patients consistently completing therapeutic homework, regularly assigned by their therapist. A leading cause of Cognitive Behavioral Therapy failure for patients is non-compliance with their assigned therapeutic homework. About 20%-50% of patients fail to complete assignments due to inconvenience, a lack of clear instructions, or forgetting to finish the assignment. MENDmate is an online platform designed to solve this problem by providing a streamlined user experience for homework assignment and completion. MENDmate allows providers to assign homework to their patients and monitor their progress. It also provides patients with the ability to track and complete their homework assignments. Additional features of MENDmate include a learning library that gives patients an opportunity to learn about mental illness and practice coping techniques, a journal that allows patients to record their experiences and daily mood and anxiety assessments. MENDmate also tracks and reports the patient's data trends such as completed assignment, mood level, and anxiety level, which allows both the therapist and the patient to keep track of their progress over time.

Fort Worth PsychWorks, a leading psychiatry office, provides comprehensive neuropsychological and psychological assessments for patients across all age groups. Currently, after administering a variety of cognitive and behavioral tests, psychiatrists must manually input the resulting data into report templates, a process that is both labor-intensive and inefficient. This manual approach can take between 45 minutes to two hours per report, detracting from the time available for direct patient care and reducing the clinic’s overall operational efficiency.
To address this challenge, our senior design project introduces an automated report generation system named the PsychWorks Report Generation System. This software solution empowers psychiatrists to select or customize templates tailored to individual patient needs, add or remove specific tests, and automatically generate detailed report text, tables, and charts based on input scores. By maintaining the flexibility of the existing Excel-based system while significantly reducing the time and effort required for report creation, the PsychReport Automator enhances the user experience for clinic staff, boosts the potential for billable hours, and ultimately supports improved patient outcomes.

Obesity disproportionately affects underserved communities due to systemic barriers such as limited healthcare access, socioeconomic challenges, and a lack of culturally relevant health resources. Under the leadership of Dr. Christina Robinson and her team of medical students—Rumaila Hussain, Kavita Patel, Joice Song, and Fatema Jafferji—we are developing a mobile health application designed to support individuals in managing their health more effectively. This app will provide users with tools to track biometrics, manage chronic conditions, and set SMART (Specific, Measurable, Achievable, Relevant, Time-bound) health goals. A key feature of the app is a motivational text messaging system that encourages users to stay engaged with their health objectives. By integrating personalized and accessible interventions, this project aims to bridge healthcare gaps and empower individuals to take proactive steps toward healthier lifestyles.

The rapid advancement of artificial intelligence (AI) presents a unique opportunity to revolutionize education through personalized learning experiences. Traditional teaching methods often fail to address the diverse learning needs of students. This research explores the application of AI in education, focusing on machine learning algorithms, intelligent tutoring systems, and adaptive learning models to create personalized educational experiences. By analyzing student data, AI can optimize learning pathways, improve comprehension, and enhance engagement. The study discusses the potential, challenges, and future directions of AI-driven personalized education.

During the college admissions process, students are presented with an overload of information from each school they are applying to and accepted by. A critical aspect for deciding on a school is the estimated Cost of Attendance (COA) and the financial aid package. Each school calculates their COA differently and thus offers a unique financial aid package. It is important for students to have a way of comparing and evaluating a school's cost with financial aid. While college counselors have developed excel sheets with algorithms that compare personalized cost with financial aid and scholarships, not all students are familiar with excel which may result in an inaccurate analysis. Transparent Tuition is a tool for students to accurately compare financial aid options from each university they are applying to. This project was developed using React.js and Spring Boot. These are two development libraries that will make Transparent Tuition scalable in the future. By creating a user-friendly web tool, students can better understand the school’s information and make a more educated decision when deciding on their university. Students will be able to connect with a college counselor to receive advice regarding their options when choosing a university. This will allow students to make an educated decision on their college based on both the short-term and long-term financial impact.

College students go through many transitions during their time at school. They learn to live on their own, manage household tasks, and balance their academics. A specific change in college is to learn how to grocery shop and cook for yourself. When students move off campus, they go from a dining plan where most of their meals are provided to needing to make all their meals. This results in many students relying on fast food or the same easy meals. Easy Bites, in partnership with TCU’s Nutrition Department, is designed to help students find quick, cheap, and nutritious meals. All our recipes are designed by Nutrition students on campus for college students to add variety to their diet. Easy Bites is composed of two aspects: an online portal for nutrition students to submit recipes for approval, and a mobile app for college students to view recipes. Our mobile app is connected to the Kroger database to provide users with accurate information about specific ingredients prices and availability. By working with the Nutrition Department and connecting with the Kroger database, we are making it easier for students from the deciding on recipes, shopping for the ingredients, and making the meal. With this, Easy Bites makes it easier to make nutritious meals as a college student.

49 million people in the United States have suffered from anxiety disorder in the past year, and 80 million have suffered in their lifetime. Many traditional methods of treatment, while often helpful, are sometimes inaccessible, time-consuming, expensive, intimidating, or overall impractical. In a world where people are increasingly in need of care and therapists are increasingly burnt out, technology bridges the gap and increases accessibility for those who previously would have been excluded. What CognitV strives to create as a solution is a Virtual Reality Exposure Therapy experience where patients can face their anxiety in a safe, controlled environment through a VR headset. Geared towards players with Social Anxiety Disorder, this treatment method allows patients to safely expose themselves to public speaking and confrontational scenarios from the comfort and privacy of their own homes. This treatment method would be faster and more accessible, is preferred by younger patients, and fills the treatment avoidance gap, all while providing a realistic, immersive experience that can effectively aid in treating mental health disorders, either with or without an accompanying clinician.

Using Virtual Reality and Artificial Intelligence, CognitV creates an immersive environment geared towards Players with Social Anxiety Disorder which allows them to safely expose themselves to public speaking and confrontational scenarios from the comfort and privacy of their own homes.

“AI Powered Patent Analysis Software”
Patent AI is an online patent analysis tool which gives feedback on uploaded patent application documents and provides a likelihood of it being accepted by the USPTO.
This tool is meant to reduce the rate of rejected patents –being at 90%– and the wait time associated in getting a response from the USPTO.
Our application is informational, accurate, intuitive, and will simplify the patent application process.

IMPROVING COLLECTION MANAGEMENT IN THE MONNIG METEORITE COLLECTION.
R. G. Mayne, J. Huther, Y. Tyagi, B. Danysh, R. Perkins, M. O' Connor, T. Truong, and B. Wei.
Monnig Meteorite Collection, Texas Christian University, 2950 W Bowie Street, Fort Worth, TX 76109 (r.g.mayne@tcu.edu)
Department of Computer Science, Texas Christian University, Suite 341, 2840 W Bowie Street, Fort Worth, TX 76109.

Introduction: Collection Management Software is a vital tool in sample-based science and a key part of any scientific collection of meteorites. However, this software is often designed as a one-size-fits-all solution, which can be used for all collections within a museum. As a result, much of the commercially available software for collections management is not ideal for the curation needs of extraterrestrial materials. Platforms are often vendor-specific, contain redundant and unnecessary functionality, and require significant time to be invested in staff training.

Over the past two decades, The Monnig Meteorite Collection has utilized FileMaker Pro for the management of the Collection. FileMaker Pro was chosen as it allows the user to design a custom solution to fit their specifications. However, this either requires that the administrator stays current on all updates and functionality of the software, or continual investment in external support for the system. The current database was designed in 2014 and is no longer meeting the needs of the Monnig Collection or the users of the database, who are primarily sample-based scientists and collectors. After consultation with industry experts, curators, and users of the database, it was decided that an update of the current database was not the best approach for the Collection, instead a new custom database that meets the needs of both the Curator and the user was commissioned.
This project introduces the development of a comprehensive database and user-friendly web application portal, marking a substantial improvement over the existing legacy system.

Project Overview: The primary aim of the Monnig Meteorite Database Project, hereafter referred to as MMDP, is to offer a detailed and robust database for the Monnig Meteorite Collection. It will feature an enhanced catalog search portal, enabling users to explore and search the collection through various parameters and filters. The system is also designed to aid gallery curators and administrators by providing detailed views of collection items, tracking sample history, and managing loans, all within a secure and user-friendly interface.

MMDP seeks to preserve the wealth of knowledge encompassed within the Monnig Meteorite Collection. The digital database and search tool will facilitate research and offer broad access to the collection for researchers, collectors, educators, and students. This initiative is set to serve as a valuable educational and scientific resource, equipped with extensive functionalities.

The database is being developed as a senior design project in the Department of Computer Science at Texas Christian University (TCU). The senior design project is a year-long program required of all Computer Science and Data Science graduates, where they work with external clients to develop and implement workable solutions to the briefs provided.

System Development and Preparation: in the Fall 2023 semester, the MMDP Team focused on data preparation and outlining the project scope into needs (must have features for launch), wants (features that are not required at launch but the capability to add them later is required), and wishes (features that are not required). Inconsistencies in the legacy data were identified and corrected; these included repeated entries, varied date formats, typographic errors, and missing fields. Python was utilized for data cleaning, and the team standardized data and organized it into relational database tables using PostgreSQL, hosted on Azure cloud for maintenance and backup.

Development will continue throughout the Spring 2024 semester and the outdated and insecure legacy portal will be replaced with a newly developed web application. This application is being built using Spring Boot for backend operations, and HTML5, CSS, and the VueJS Framework for a responsive front-end UI, ensuring accessibility across various devices. The current launch date for the new collections management system is May 2024.

Functionalities: MMDP will address the need for functionality for both the administrators of the database (primarily the Curator in this case) and the external user (Figure 1). The required parameters for both of these audiences are described below.
All users of the database will be able to:
1. perform parameterized searches using criteria such as Name, Monnig Number, Class, Group, Clan, Country, and Observed Fall or Found (Figure 2a).
2. filter and modify search results directly on the search result page (Figure 2b).
3. Find accessible detailed information about each meteorite sample, including images, from the search results via individual display pages for each sample.
4. download all the search results based on the given constraints with a single click from the search results page.

Administrators will be able to
1. have access to specialized functionalities that are secured and restricted from regular users. Upon logging in, they are presented with a portal offering various database management options.
2. view more detailed information about samples than regular users, including the sample's history and loan information. They have the ability to add new meteorite samples or create subsamples.
3. perform data manipulation tasks, such as deleting or modifying existing sample records.
4. have control over the media associated with samples, allowing them to add or delete media.
5. administrators able to create, view, update, and delete history entries for each sample. This historical data management is a key new feature not possible in the current system.
6. Access loan management capabilities include adding, modifying, archiving, and accessing archived loan entries for samples.
7. print labels for samples, which can be used for curation in the vault.

Summary: The MMCD stands as a model of integration, combining domain expertise, data best practices, and user-centric design. This project offers a template for other universities, museums, galleries, and research centers aiming to enhance their functionalities and provide a seamless, user-friendly experience for accessing and managing meteorite data collections.
Embodying the spirit of scientific collaboration, this initiative is open to opportunities for collaboration to expand the platform's capabilities or to implement similar solutions in other institutions.

Acknowledgments: We are grateful to the Department of Computer Science at TCU for their continued support of the Monnig Meteorite Collection through the Senior Design project. We also thank Dustin Dickens for his advice and assistance in the discovery portion of the database redesign.

The Department of Computer Science at Texas Christian University offers a course where senior students, in teams, collaborate with clients to solve real-world software problems. Students handle every project phase: definition, analysis, design, implementation, testing, deployment, and documentation. However, in these teams, there's a variation in how much each student contributes. Some are very active, while others are not. Communication issues can also arise. To handle these challenges and improve team efficiency, there is a Student Performance Tracking system in place that includes Weekly peer evaluations where each student evaluates their own teammates in accordance with the rubric defined by the professor and Weekly Activity Reports (WAR) where each student writes down their own contributions for the week.

While this system works and improves team efficiency, these tools are too manual and thus time consuming. For the WAR, each student has to edit the Google Docs document for the week which is then reviewed by the professor. This can lead to human error, meaning some students might not get the right credit if they make mistakes while filling out the Google Docs document. For the Peer Evaluation, each student must review the WAR for the week and then make an excel spreadsheet to evaluate their teammates and then upload it to TCU Online. Once all students have turned in their peer evaluation report for the week, the professor has to download reports of all students and then run these through a Java program which then calculates the results for all students. Then the professor uploads the results to TCU Online (a course management tool used by TCU). Not only does this leave room for human error on the students' side (spreadsheets must have the right columns), but it is also very time consuming for the professor as they have to download all reports manually from TCU Online and then run the Java program and finally upload the results back to TCU Online.

The automated Student Performance Tracking system (Peer Evaluation Tool) streamlines the evaluation process by providing a centralized website where students can directly fill out their Weekly Activity Reports (WARs) and complete peer evaluations. It also enables them to view their own submitted WARs and received peer evaluation scores from their teammates. For the instructor, the system offers the functionality to create and customize evaluation rubrics, which ensures consistency in peer assessments. Instructors can access and review all peer evaluations and WARs, allowing them to monitor team dynamics and individual contributions efficiently. This comprehensive solution eliminates the manual handling of documents and the need for external spreadsheet software, thereby reducing human error and saving time for both students and instructors.

This presentation investigates the learning process of artificial intelligence by training a model to play the game of Go using an AlphaZero-type algorithm. Through evaluation of 12 Go models, the authors reveal the split personality many exhibit, much like the famous Schreiber book Sybil. The best models appear indistinguishable from human players in the early stages of the game before devolving into self-destructive tendencies in the endgame. Possible remedies for this behavior are explored through modifying training data generation, hyperparameter tuning, and optimizing neural network input dimensions.

In the dynamic environment of venues with large seating capacities, efficient management of seating occupancy emerges as a critical challenge. Traditional manual monitoring methods are often cumbersome and prone to inaccuracies, hindering optimal seat allocation and event management. Addressing this issue, our senior design project introduces an AI-based solution tailored to revolutionize real-time seating availability reporting for event organizers.
This project aims to provide a comprehensive tool that enables event organizers to track seating occupancy in real-time, facilitating the identification of peak attendance periods and enabling data-driven decision-making. By harnessing the power of artificial intelligence, our system offers a detailed analysis of seating patterns, thereby enhancing the efficiency of event operations and optimizing resource allocation. The ultimate goal is to improve the event experience for both organizers and attendees by ensuring a seamless flow of information regarding seating availability, leading to more effective management of large-scale events. Through this initiative, we endeavor to set a new standard in venue management, where technology and data converge to create smarter, more responsive event environments.

Efficient teacher-student interaction analysis is essential for educators to enhance teaching quality. Traditional manual review methods are excessively time-consuming and can yield subpar feedback. ClassifAI offers a streamlined solution for educators to gain insights without sacrificing work hours, utilizing the OpenAI Whisper model for transcription and a fine-tuned Gemma model for question categorization.

ClassifAI is advancing existing tools by addressing four key improvements: transitioning to local hosting for cost savings and data security, integrating the WhisperX model for improved transcription accuracy, automating Costa's Three Levels of Thinking question classification via Google's Gemma, and upgrading the web interface for better user experience.

ClassifAI's architecture comprises a user-friendly web server with ExpressJS and React, a local MongoDB database, a fine-tuned Gemma model for question categorization, and WhisperX for speech-to-text. ClassifAI offers speech recognition, diarization, question categorization, and analysis, delivering enhanced performance. Educators easily upload their teaching audio/video on our platform via a file or YouTube, which is then processed by our GPU server for transcription and analysis. The resulting transcript, graphs, and metrics are accessible for review and can be exported in various formats.

Speech impairment ranks among the world's most prevalent disabilities, affecting over 430 million adults. Despite its widespread impact, many existing video-conferencing applications lack a comprehensive end-to-end solution for this challenge. In response, we present a holistic approach to translate American Sign Language to subtitles in real time by leveraging advancements in Google Mediapipe, Transformer models, and web technologies. In March 2024, Google released the largest dataset for the problem domain with over 180 GB in size, containing ASL gesture sequences represented as Mediapipe numeric values. Our methodology begins with the implementation and training of a Transformer model using preprocessed Google dataset, followed by the establishment of a back-end server equipped with the trained model. This server handles video input preprocessing and real-time inference, communicating with client services as a REST endpoint. To demonstrate the practicality of our approach, we developed a video conferencing application utilizing the AgoraRTC SDK, which communicates with our back-end server to transcribe user gestures to text in real time, displaying them on the receiving end. Through this end-to-end system, we enable video calls enhanced by the real-time transcription of fingerspelled gestures with low latency and high accuracy, effectively bridging the communication gap for individuals with speech disabilities. With a growing imperative for AI applications engineered for human well-being, our project seeks to promote the integration of AI in applications designed to enhance human wellness, thus bringing the broader awareness and adoption of this endeavor.

Chalk Mountain Services of Texas, LLC. is a trucking company whose business is transporting raw materials, such as fracking sand, to various oilfield sites in and around west Texas. With over 1,300 assets in their fleet, they’re presented with a number of logistical problems, like optimizing a driver’s time to make as many trips between drill sites and raw material depots as possible in a day. Such routing and scheduling applications must have accurate data—the assets are either in or out of service and their location—to schedule sensible routes.

Should an asset break down in the unforgiving terrain of west Texas, the appropriate employee should have the ability to take note of such an incident so that routing and scheduling applications have correct, up-to-date data. The company’s current solution allows for any user to make changes to any asset, regardless of authorization status. Inconsistencies in assets’ statuses can lead to an employee having to manually intervene in the scheduling process, which decreases the company’s overall efficiency. Additionally, their current application is not mobile-friendly, but a sizable portion of users nevertheless interface with the current website from their phones.

The company’s expectations come in either one of two forms: a website and a companion app or a reactive website that can be used on a desktop or mobile device. The application shall use CRUD—create, read, update, and delete—methods to keep track of the assets, and the application shall provide different users with different access levels with Active Directory authentication. We have created a reactive website that can be used from either a desktop environment or mobile one, and our implementation of their requirements exists as a three layer architecture: a Microsoft SQL Server database, a backend developed in NodeJS, and a React front end. To make the deployment as simple as possible, we did not pursue developing the application on cloud providers; the application depends on a connection to an in-house SQL server and Active Directory service both of which cannot be accessed outside their intranet and are critical to the application’s functionality.