2018 MISM Capstone Projects

Our capstone proposal is “You’re A Wizard”, an augmented reality game that will allow players to cast magic spells. For example, when the user issues a voice command (incantation), the user will be able to ‘cast’ a magic spell, which will result in corresponding animations and sounds. Animations will be holographic object creation/manipulation (floating, shrinking, growing). Whatever the spell, players will be able to experience the magic of The Wizarding World for the first time, in augmented reality.

The platform for the game is the Microsoft HoloLens, a pair of mixed reality smartglasses. This specific platform, and indeed mixed/augmented reality as a whole, is currently in the beginning phases; the HoloLens is still primarily available as a “Developer’s Edition” at $3,000. This capstone would seek to ‘get in on the ground floor’ of this major advancement in computing. Our game will show off what the HoloLens and mixed reality are capable of, thus exposing the world to the approaching future where HoloLens and mixed reality “will become that much a part of you” (Cook, T., Sep 2016).

Mixed reality has limited implementation as of the moment, but it is believed that augmented reality is an integral part of the future of computing; Gartner identifies augmented and virtual reality as one of its “Top 10 Strategic Technology Trends”[1]. Companies are in the early stages of getting the technology into people’s hands, and platforms such as Apple’s ARKit and Google’s ARCore enable developers to create apps for phones that many people already have, like the iPhone 8/X, Pixel 2, or Samsung Galaxy S8. The HoloLens has the immersive nature of a head-mounted display, while also getting rid of the tether usually associated with VR headsets and allowing the user to look through the smartglasses display. It also operates on Windows 10, bringing with it the advantages of being a PC and an AR device all in one. “You’re A Wizard” captures the aspects of augmented reality that we want to focus on: current technological
capabilities; adding business value; improving human-computer interaction; and in some cases, just making
dreams come true.

Mohoca

Name(s): Trevor Harmon, Jarrod Wilbur
Faculty Mentor: Jeff Jenkins, jeffrey_jenkins@byu.edu

Executive Summary

In the home building industry, home builders spend thousands of dollars having home interior designers make their model houses feel like model homes. Prospective home buyers walk through a model home and fall in love with the look and the feel of the home. But when they get to the end of the tour, they’re able to buy only the house, but nothing on the inside. Mohoca solves this problem: it acts as a marketplace that connects the home builder and home designer to the home buyer, enhancing the home buying experience and encouraging customer upsell.

Mohoca is different than any other online retailer because it uses the home builder’s name and the model home experience in order to reach customers. Home buyers associate their experience with a home builder (like Woodside homes), not with any designer (Kimberly Timmons Interiors, in this case). Additionally, many model home visitors remember the physical space and layout of the home. For example, they may remember that they loved a particular lamp in the living room of the home. Mohoca provides a home-builder-branded marketplace that has all of the products arranged according to the room, model home, and community. This provides a seamless transition from the physical home to the digital store, allowing customers to easily find the product they are looking for.

The technologies we are going to use for Mohoca are fairly straightforward–we plan to use AWS as our infrastructure and Ruby on Rails as our language/framework of choice. Because marketplace/e-commerce solutions already exist, we plan to follow existing conventions as we make technical decisions about Mohoca. The only exception is the way a user finds a product: our system is designed to be aware of the model home experience, and we will optimize around the user experience of finding a product in a specific home.

Comfort Cribs

Name(s): Tyler Bangerter, Nate Munk, Nathan Meredith
Faculty Mentor: Dr. Mark Keith

Executive Summary

Introduction
Our project is titled Comfort Cribs, which is an app to help prevent Sudden Infant Death Syndrome (SIDS). A monitor placed in the child’s room will inform parents of the current humidity and temperature. If the temperature or humidity falls out of the optimal range, the parents will receive an alert via text message or email.

The parent will then be able to optimize the child’s room for a healthier environment. The companion website application will display the temperature, humidity, historical data, and provide tips to decrease the likelihood of SIDS. We are still researching other factors that trigger SIDS and will implement them
into our app as we find them. We will also implement other features as they are discovered by researchers. Our target market is parents with a newborn infant, adoptive parents, and individuals (non-babies) who are concerned about their health.

Problem
SIDS and airborne allergies are a big concern for most parents. While SIDS is not common, many of us have friends or family members who have lost an infant because of SIDS. Doctors offer many recommendations to reduce the risk of both SIDS, and the development of allergies. There are other
solutions such as Owlet, which monitors heart rate and oxygen levels. The Owlet currently sells for $299. Comfort Cribs will retail for under $50.

Solution

Hardware
The baby monitor is built on a Raspberry Pi Zero with a combo temperature and humidity sensor that collects the needed data. We decided on this platform because of the low cost, availability, and ease of access to tutorials. The Pi will sit in a black case with a port for USB charging. A rechargeable battery will be available for purchase separately.

Software
To gather the temperature and humidity we used a Python script on the Raspberry Pi. This code sends data in the form of Json to an AWS S3 database using IoT rules. From S3 we import the Json file into our website, which is running on Python, Django, and Django-Mako-Plus with a PostgreSQL database. If the temperature or humidity falls outside the optimal range, an AWS IoT rule will send the user a text message or email informing them of the current temperature or humidity.

The parents or guardians create an account when they purchase a Comfort Crib device. After they are logged into the site, they can see the current temperature and humidity being updated every 60 seconds. The user will also be able to see the historical data associated with their device. The website will also give information about Comfort Cribs, the technology, founders, and tips with associated research to help educate the user.

PTA Finance App

Name(s): Bronson Gardner, Nick Banks
Faculty Mentor: Dr. Tom Meservy

Executive Summary
This project addresses several problems related to financial management in small nonprofit organizations, such as PTAs and professional associations:
• Processing reimbursements takes a long time and requires a lot of paperwork, including signatures from several people in many cases.
• Treasurers are often volunteers with little to no accounting experience, and often aren’t sure quite how to handle things such as processing transactions or managing
• These organizations must prepare reports on their expenditures and income, which can be tedious and confusing when done by hand.

Software solutions exist for each of these problems, but most are priced out of reach for small nonprofit organizations. Those that are not out of reach only solve one of these problems and are often still difficult to use and/or integrate with one another smoothly. Thus, the problem effectively remains. Our solution is to build a mobile-first web-based application that addresses all three of these problems now, with the possible integration of more nonprofit-focused features in the future. The application solves these problems by:
• Simplifying the reimbursement process by allowing members of the organization to submit reimbursements by taking a photo of their receipt, and allowing leadership to approve the expense digitally
• Limiting confusion for inexperienced treasurers by providing a simple interface, tested on users who do not possess expertise in accounting or software development fields
• Streamlining the reporting process by providing the ability to produce printable reports automatically based on pre-defined parameters.

The application is built using the React Javascript framework for the user interface, with several Firebase services to provide back-end functionality: Firebase Cloud Firestore for the Database, Firebase Storage for storing images, Firebase cloud functions to facilitate email alerts and other simple backend functions, and Firebase Authentication to manage authentication and access controls. We also use Firebase Hosting to host the application.

MemryHub

Name(s): Brandon Colton, Greg Bird
Faculty Mentor: Jeff Jenkins: jeffrey_jenkins@byu.edu

Executive Summary

Problem
Current methods for storing and viewing memories – one’s own and those of family/friends – are difficult. These memories include recordings of grandparents, family photos, and written stories or journal entries. Most people have their memories stored on a computer or phone, and backups rarely exist.

Solution
MemryHub is the storage solution for personal and family memories. It becomes a digital memory chest with a two-fold approach: 1) connecting with a cloud storage solution (Google Drive) to save a user’s memories, and 2) providing a timeline to view the stored memories. Unlike its closest competitors (Ancestry.com and FamilySearch), MemryHub will focus on stories of living and recently deceased individuals.

Components/Technologies
MemryHub will be a web application that integrates with a user’s Google Drive for memory storage. The user would perform all of their uploading and interaction with the files (memories) through the web application. A user would create a timeline for each individual person for whom they have a memory to store. Audio, video, written, and photo memories would then be added to those timelines and stored in the user’s GoogleDrive in an organized manner.

Our web application will be built using Python/Django and hosted on Heroku. The database for the application will be a PostgreSQL database. The PostgreSQL database will only be used to stored information on how timelines are organized. All other files will be stored in the user’s personal Google Drive. MemryHub will create a folder within the user’s Drive and will utilize the folder directory to store timelines and memories. Because the app stores the files in a user’s Google Drive, MemryHub avoids significant storage costs and ensures the user still has full access/control of their files.

AHJAX

Name(s): Andrew Sanford & Troy Bradshaw
Faculty Mentor: Dr. Conan Albrecht

Executive Summary

When people want to learn something new, they look for an instructor who presents the concept in a clear, easyto-understand way. Currently, most ethical hacking video tutorials are very difficult to follow and learn from, whether it be from their distracting death metal music to voice modulators to lengthy videos. This leaves most potential students unable to learn what they need to for their schooling, (future) careers, etc.

We set out to create high-quality ethical hacking video tutorials that anyone with basic knowledge of networking and the command-line can follow and understand. To facilitate viewing, we hosted the videos on YouTube and integrated the videos into a static website, named Ahjax (https://ahjax.com). These high-quality video tutorials have aimed to be:
● Succinct
● Easy to follow and understand
● Appropriately use music

In order to make the tutorials:
(1) Succinct, we presented only necessary information (leaving out “fluff”). This was be done by:
(a) Studying out the concepts ahead of time (which usually required us to learn more about the
concepts and associated tools)
(b) Planning out what we we presented and how we presented
(c) Dividing tutorials into multiple videos (if necessary)
(2) Easy to follow and understand, we included scripts with each video. This way, those who can’t hear will still be able to learn. We also included files and links to materials the student needs to follow along with the tutorials
(3) Appropriately used music, which for the scope of this capstone project usually meant no music at all. Our capstone project focuses primarily on the video tutorials; however, the website is still an important part of our final deliverable.

To succeed in our goals, we divided responsibilities between ourselves as follows:
● Andrew Sanford
○ Penetration testing tutorials
○ Network mapping tutorials
○ Password cracking tutorials
○ Any additional videos deemed necessary or interesting (if time permitted)
● Troy Bradshaw
○ Website development*
○ Video tutorials:
■ Social engineering tools
■ Network sniffing
■ Any additional videos deemed necessary or interesting (if time permitted)

To record the videos, we used QuickTime, YouTube video hosting, DSLR, and VideoPro Mic. The website is hosted on AWS, registered at GoDaddy, and uses the Laravel framework and Bootstrap.

User Interface and Design Thinking Class

Name(s): Emilie Knecht Martinez & Alicia Wilkin
Faculty Mentor: Bonnie Anderson, email: bonnie_anderson@byu.edu

Executive Summary

We created the User Interface (UI) and Design Thinking course that was taught the first term of Winter 2018 and has been recorded and uploaded to Udemy and Skillshare.

Over the past year and a half, many of our peers have lamented their lack of visual design skills. They could not find a class at BYU that fit their needs that allowed students outside their major to take the course, and struggled to find similar courses online. Many expressed that they would be willing to invest time to cultivate design skills, but could not find an easy or effective place to start. These students are not alone in wishing they had more design skills. Multiple recruiters of BYU IS students have also expressed that IS students are lacking in this area. In short, IS students do not have a foundation in UI or Design Thinking, and students and companies alike could benefit from them having these additional skills. This past semester we taught several students visual design
skills through a UI and Design Thinking course and created an online course so future students can benefit from this knowledge as well.

The UI and Design Thinking course was taught the first term of Winter 2018 on Tuesdays and Thursdays from 9:30 a.m. to 10:45 a.m. The 1.5 credit course taught students important design principles and how to use design software, like Sketch. The class was composed of Information Systems, Information Technology, and MBA students. Before creating our course, we conducted user research to find what topics students would be most interested in covering.

Here are the topics we covered based on their feedback:
1. Visual Fundamentals (Color Theory, Layout & Composition, Typography)
2. Practice with Industry Tools (Sketch, Sigma, AdobeXD)
3. Interactive Prototyping
4. Brand Management
5. Combining UX, UI, and HCD
6. Logo Design

Throughout the term, instructors and students alike contributed to a Design Resources Google Doc so that students could share design inspiration and tools to help them design better. We also gave them access to all class slides for reference. As design needs arise, students will be able to reference these resources and continue to benefit from them.

The second phase of our capstone has centered on packaging the materials we used to teach our class so that we can deliver the content online through the learning platforms Udemy and Skillshare. Based on the feedback we received from students in our class, we adjusted our course materials to include more follow-along tutorials.

Our technology stack was simple. We used Sketch, InVision, InVision Studio, a camera and a microphone, Learning Suite, BYU Tech Podiums and our laptops.

TrashTalk

Name(s): Lucas Carvalho, Spencer Shumway, Alexandra Shumway
Faculty Mentor: Jeff Jenkins

Executive Summary

TrashTalk is a business idea that aims to streamline the garbage collection process. Matthew Liddle came to the IS program to present his idea, and selected our team to build the corresponding web application. He brought homegrown IoT sensors which, when placed on a dumpster, record multiple times each day how full the dumpster is. The goal for our team was to create a database to host this data, and an accompanying web application to serve it up for waste management admins.

Our application was built to address three main needs that Matthew expressed: store data about a company’s waste containers, provide visualizations about that data, and optimize collection routes based upon that data. The application was built using Ruby on Rails. APIs were integrated both from Google and Particle. The database is a MySQL DB. We used Amazon Cloud9 for development and deployed with Heroku. The graphic below illustrates the path the data takes.

RESEARCH – Machine Learning for Cursive OCR

Name: Grant Dexter
Faculty Mentor: Dr. Tom Meservy

Executive Summary

INTRODUCTION
Statistics from a well-known genealogical site show that over 1.3 billion genealogical records have been manually digitized (indexed), still with millions more to be processed. In the LDS Church History Library, tens of people are employed to manually create digital transcripts of historical documents. Ancestry, Kindex, and State historical sites have thousands of imaged documents with cursive writing to be digitized. This research project explores tools, libraries, and algorithms to implement a solution for automating the digitizing of cursive (or other non-print) handwriting on historical records.

PROBLEM
Cursive handwriting on historical records involves two unique problems not solved by current OCR(optical character recognition) systems. The first is that cursive characters all run together, making current edge detection systems ineffective for segmentation. The second is that background colors/noise, handwriting scripts, and faded images render simple image thresholding useless. Many OCR technologies are available for print writing. Google’s Vision API detects location of handwriting and interprets with decent accuracy. Dropbox, Microsoft, and others have similar products, but these services also fail on cursive writing, especially with historical documents.

SOLUTION
This capstone project has involved research, experimentation, and implementation of tools to automate the digitization of cursive writing on historical documents. The solution, to-date, is built in Python, running on a Windows Server, leveraging OpenCV, and Keras with a Tensorflow Backend. The system detects writing locations, extracts snippets to be classified, and finally classifies individual letters within snippets based on a convolutional neural network (with future expandability to LSTM layers). A dataset was created specifically to train this model, from a set of imaged documents gathered from the Joseph Smith Papers Project (late 1800s, five different script authors), augmented with additional training images. The dataset currently includes over 26,000 images.

Hype Lights

Name(s): Jeffrey Niemann, Cameron Pattberg, Benson Weeks
Faculty Mentor: Professor Meservy

Executive Summary
Introduction: Hype Lights is an entertainment experience enabling events to engage attendees in a crowd-based light show utilizing Apple and Android mobile devices. Through coordinated planning, each device will receive coordinated instructions to turn the flashlight on or off in precise intervals based on location within the crowd.

Problem: There are hundreds and thousands of devices at sporting events and concerts that are completely untapped for their combined potential. While many stadiums feature advanced lighting and sound equipment, they forego the biggest potential source of lighting equipment sitting right in front of them, as well as failing to engage the crowd in a highly engaging way.

Solution: Hype Lights designer application empowers event staff to create and deploy a lightshow to crowd devices through iOS and Android mobile applications. Crowd engagement and participation increases the quality of experience, the memorability of the event, and the likelihood that attendees will want to return and
invite others. This drives up the value of the event. It requires no additional equipment on the part of the event coordinators as the management app can be used from any mobile or laptop device. Event coordinators can choose their songs, and then go on the application and choose corresponding light patterns to be displayed on the crowd.

Major Components/technologies: Hype lights uses ultrasound detection capabilities in participants phones and other devices to coordinate the light show. Other web based technologies are used to coordinate location of devices and the configuration required for each device. Mobile devices will use either an iOS or Android
mobile application of minimal size to communicate between the central web application and the device flashlight. The web application uses the Django framework and a JSON API as the medium for communicating with mobile applications.

Absolute Task

Name(s): Thomas Christensen & Chase Bleyl
Faculty Mentor: Jeff Jenkins

Executive Summary

Absolute Task is a task management application designed specifically for fast food restaurants. Multi-unit restaurant owners often have difficulty gaining transparency into their operations. They can ensure the store they are visiting is operating well, but the other N restaurants typically struggle without hands-on
management. Hiring and training minimum-wage workers is expensive, and extreme rates of attrition make establishing communication procedures almost impossible. These factors lead to tasks not being done when they need to be done, leading to diminishing profits at an already low-margin business.

Absolute Task aims to increase efficiency and transparency while reducing costs. By re-designing the standard checklist, Absolute Task has simplified task management for a fast-paced business. An intuitive design reduces training and increases efficiency. SMS notifications are sent out to management when tasks are not completed, increasing accountability. This enables owners to be in N places at once, knowing where the problems are when they happen, not when they affect their profits.

A management application gives operators the ability to manage their schedules and tasks, as well as see reports for their units. These features are currently being developed and rolled out to users when they are finished.

Technology Incorporated:
– Mobile App – React Native Android
– Management App – ReactJS
– Authentication: AWS Cognito User Authentication
– API: AWS API Gateway, Lambda (Python)
– Data: AWS RDS – MySQL
– Notifications: AWS SNS

LDS Scripture Citation Index UI Research

Name(s): Brandon Buchanan, Mitch Hussey
Faculty Mentor: Bonnie Anderson

Executive Summary

Dr. Liddle created an app called LDS Scripture Citation Index. This app enables users to explore scriptures by providing them the ability to see historical and current LDS talks that have included the scripture they are reading. This database of information is used by users to study the Gospel, prepare talks and help people explore historical publications that aren’t as reliably available to the typical individual.

The Problem
The app has seen relative good user growth over the past few years as more users have installed and actively use the app. The app has relied solely on the development skills of Dr. Liddle with no assistants from any formal development team. While the app in its current state contains excellent content along with good design, Dr. Liddle has not been able to perform any formal user testing due to time and resource constraints. Because of this, there is a potential to develop an improved user experience by performing user testing and applying the recommendations from testing to the app. Project Overview

Our Capstone project included several iterations, each providing new data about user interactions and data. The project had three major sections: data analysis of current user data, a first iteration of low fidelity user testing and a second iteration of high fidelity user testing based on updated user interface. We analyzed current user data collected utilizing Flurry, the current analytics engine monitoring the app. We used this data to help us understand historical patterns of user interaction with the app. We found a pattern of users initially engaging with the app and then slowly loosing engagement after a week or two. This data and other
patters we discovered helped us understand that users were having a difficult experience navigating around the app and understanding how to best utilize the app’s many features. We decided to direct our attention to helping users navigate and use the app more efficiently.

We developed low fidelity mockup designs of the app with alternative navigation and bookmarking functions. We then used these designs in A/B testing with a group of users, majority of which had never used the app prior to our testing. The testing provided us with valuable results helping us better refine the user interface and app navigation.

We used the results from our low fidelity testing to developer higher fidelity mock-ups of the app. We used these mock-ups in testing with more users. The users we tested provided us with further insight into how we could design a better user experience for the app.

The results from our project were provided to Dr. Liddle as a series of recommendations of how to improve the LDS Scripture Citation Index app for users.

IS 555 Revamp

Name(s): Becky Abbott, Chase Sowards, Johnny Liao
Faculty Mentor: Doug Dean

In addition to a foundational understanding of data science principles, industry also demands technical expertise and experience with technologies used in data science. Now, more than ever, these hard/technical skills are a priority for budding technical professionals.

In Information Systems (IS) 555, students learn many valuable insights about data modeling algorithms and how to interpret data. However, upon completion of the course, past students lacked the hard technical skills for data science, like R programming and Azure. Students that have a desire to enter a career in data science find it difficult to become marketable because recruiters are searching for those that not only understand the algorithms, but also know how to create automated processes using R and similar tools. Our capstone project helps students gain these skills and knowledge that are in demand for data science/analytics positions.

Our capstone project includes creating high-quality data science tutorials and lessons for Dr. Dean to teach in his data mining course during the Winter semester of 2018. The course will include the use of the R programming language to teach students concepts of data cleaning, creation and interpretation of data visualizations, forecasting, the use of JMP to support the teaching of data mining algorithms, and the use of Azure to give students experience in further data science technologies.

Our final goals/deliverables will be (1) course content, tutorials, and source code, (2) a student evaluation of the effectiveness of the course, and (3) plans to continue the improvement of this course in the future. We will work with Dr. Dean to achieve these goals.

Bambi Data Engine

Name(s): John Lim, Aaron Wilson, Sean Burke
Faculty Mentor: Gove Allen

Executive Summary

Background

Merchant Labs is a technology company founded this year in Bentonville, Arkansas. Their aim is to bring the culture of technological innovation that exists in Utah Valley to the hub of Walmart headquarters. They currently provide reporting services for suppliers that sell product into Walmart.

Problem

The system that supports Merchant Lab’s reporting services was built several years ago for a handful of clients. Since then, Merchant Labs has more than tripled their customer base and is in need of a modern data engineering system to support their data engineering services.

Specific issues include:
• The database was designed by a student without any prior experience and the data model is not scalable to include store level data.
• The current reporting solution employed by Merchant Labs relies on an employee waking up in the middle of the night to refresh the data found in some spreadsheets. We are going to eliminate this responsibility and automate the reports so that they can be automatically created.
• The data workflows are managed by database tables that contain records about when to run queries to Walmart’s database. Modern companies that do this kind of work employ a solution such as Apache Airflow to programmatically manage their data pipelines. That development environment needs to be versatile enough to deploy on local development machines and easily deployed into the cloud.

Solution

Our propose analytics solution for Merchant Labs can be divided into three major pieces:
Linker API: Walmart does not provide an API to interface with their data. Pulling data requires third-party software as well as a REST API to manage the requests that we send to Walmart programmatically.
Dockerized Airflow: Not only do we need to deploy Apache Airflow, we need to write Docker containers that can be run through Docker so that we can work on a local machine as well as in deployment. Otherwise various configurations will easily break the Airflow environment. Airflow is an open-source project built by Airbnb to manage the production and scheduling of tasks, such as our needs to extract, transform, and load data from Walmart’s Retail Link.
Web App: Once the data has been harvested and stored, we need an interface for clients to login that allows them to request their report data immediately. This request needs to be processed instantly since we are trying to solve the problem that Retail Link is slow and unprocessed data.

Board Game Pal

Name(s): Brayden Connole, Stephen Thomas
Faculty Mentor: Conan Albrecht

Executive Summary

Our project is called Board Game Pal. We want to provide players of niche games companion apps to simplify their play and help them have a more enjoyable experience.

Families across the world and board or card game players everywhere all face these problems at some point or another. They sit down to play a game and realize that they aren’t ready to play; they still need to get pen and paper to keep track of their scores and stats. Once these players obtain a scrap paper and something that may qualify as a writing utensil they begin to play the game. The scoring of these games is sometimes quite complex, and at the end of a game the result is often a messy paper with scribbles, eraser marks, and hopefully math done correctly. There should be an easier and cleaner way to score a simple game of Pinochle. Another problem some board gamers experience is when the game they’re playing has hundreds of pieces and components of varying shapes and sizes. These pieces sometimes get lost and they are always a mess and a hassle to set up and clean up. If these pieces could be digitized it could solve this pain point. Finally, many gamers want their own analytics system to measure their performance after each game. They ask questions like how many times has this player or
faction won? Should I alter the game in some way to make it fairer? If these players had an app it could give them these stats.

The solution to these problems is to use what every individual has on their person at all times, their cell phone. It might sound cliché, “There’s an app for that!” but right now, there isn’t! Some companies have tried to make generic apps that solve the problem for all games. These solutions are rarely useful as each game is unique and different. They are not in the “one-size fits all” category. Games are unique, and so the apps that serve as a companion to them should be as well. Our goal is to provide an individualized solution for different games that will specifically smooth out pain points in each game. We want to hit a niche market, and we’re going to start with the games Pinochle, Sushi-Go, and Legendary. We will provide seamless scoring mechanisms in the app. Say goodbye to the hassle of keeping track of all the game pieces as they will be digitized. Finally, our apps will provide the statistics and analytics players have been searching for to enhance their gameplay and make gaming more enjoyable.

We want to be able to reach gamers in both the android and iOS markets as quickly as possible. React Native will allow us to reach both simultaneously. React Native also offers us the unique ability to create components that can be used in multiple apps. For example, each companionship app will have a menu of some sort. We can build this menu once and with a little tweaking make it unique for each app individually. Other components will include player and team tracking and game play screens. We are going to use local data storage provided by the devices of the end users. The data is easily accessible and readily available. There is a risk of users losing their data if their device gets destroyed, so we have considered implementing API’s to popular board games sites like boardgamegeek.com which will allow users to store their results online.

Tri-Change

Name(s): Buster Townsend, Johnny Welker
Faculty Mentor: Mark Keith

Introduction/Problem Statement

Poor mental health is a problem most Americans (and the world) fight today. There are a variety of self-improvement apps and other apps trying to tackle the problem. Some of these apps and programs help and others, well, tend to fall a short. The question we are trying to solve is how can we help users break bad habits that feed poor mental health and overall improve lives.

Proposed Solution/Approach

Our project is intended to focus on creating aids to better treat mental health. The idea is to build a web application (mobile first design) around a new therapy model. The purpose of the app is to help users achieve a more healthy mental state. The interesting part of this app is that it can also be used in other facets of life. As we build and test the app, we can prodce results to see if the user is in fact improving or not.

Our goal is to build a web app around a new therapy model that can be applied to a variety of habits and mental health problems. The new model has had success in therapy sessions and hopefully now can be adapted to fit on a mobile device to serve a larger audience.

The Tri-C model is focused on helping the patient identify a core issue and an associated goal that the patient wants to change. The model then follows a triangle of three principles: Capacity, Connection, and Commitment.
● Capacity is geared to helping the patient understand why he/she wants the change and what are the
needed actions and subgoals in order to meet that better future state.
● Connection focuses on helping the user connect and create emotional ties and bonds that will help
reinforce the new desired behavior set in capacity section.
● Commitment’s goal is continue to develop a set of actions that will enlarge the user’s capacity and
strengthen the user’s connection.

As a user moves from one section to another, the user re-enforces the positive behavior, emotions, and actions developing the user’s core values

Darrio

Name(s): Jake Wagstaff, Matt Gardner (iOS), Jordan Widdison, Scott Chambers, Seth
Weinheimer (React / API)

Faculty Mentor: Conan Albrecht

Executive Summary

The Problem

Year after year family and friends’ birthdays come and go. A 20$ amazon gift card here, a new pair of socks there, and a birthday has come and gone. Birthday gifts are often a wasted opportunity. The goal is simply to celebrate friends and family in a fun and generous way, but often our best efforts don’t leave the giver or receiver fulfilled. There are many reasons why the experience of birthday gift giving isn’t always what it could be. One of the contributors is forgetfulness. We sometimes forget to buy birthday presents for those we love until the last minute. Another reason is budgetary constraints. With the number of friends and family to give gifts to most of us are only able to afford $10 – $20 and no one would expect any more. Combining the forgetfulness problem with a $20-dollar limit and we just can’t afford to give what we would like to give.

The Solution

Darrio will keep track of the important birthdays in your life. When it’s time for a birthday Darrio solves the gift giving problem by letting the user choose their most desired gift and the fun begins. Family and friends are notified through social media channels and they collectively “crowd fund” the birthday. It’s a fun, easy way to remember birthdays and give the gifts you’ve always wanted. If 15 people contribute 20$ you can give that shiny new Xbox instead of a token pair of socks!

The Team

Our team has a proven track record of successful iOS and Web projects. We have the experience to be able to create and lead this product from an idea into reality. Every day around 200,000 people use the products we’ve helped create. Jake and Matt are mobile experts with skills in Android and iOS. Jake is most famous for having been one of the few developers of “Mutual” the LDS singles dating app. Matt is the well-rounded jack of all trades and Android genius. Scott, Jordan and Seth lead out the web. Scott is the database guru. Jordan has never seen a problem he can’t handle and Seth is an experienced rails engineer. Together we are a well-rounded full-stack team will a portfolio of successful projects more than capable of delivering the technology Darrio needs to scale and grow.

Monetization

The long-term goal for any project has to be monetization. Without upfront capital, this project needs to be selfsustaining from the beginning. Fortunately, there is a very profitable business model worked into the platform. Darrio will charge a platform fee of 5% on every transaction in the system. An immediate 2.7% goes to Stripe our credit card processor and a small portion will go to server costs. We anticipate a 2% percent profit margin. As is the case with most technology companies we will become very profitable at scale. Initially, most of the profits will be reinvesting in SEO to continue to drive traffic. Our year 1 goal will be 20,000 birthdays processed per month. At that volume with an average birthday fund of 100$ our projected gross annual revenue would be $2 million with an estimated profit of $500,000.

BeautyHub

Name(s): Heekyung (Casey) Yang, Heesu (Stacey) Yang
Faculty Mentor: Bonnie Anderson, Stephen W. Liddle

Executive Summary

Majority of Google or Yelp searches display local brick and mortar businesses. Therefore, it is challenging for customers to locate nearby home-based beauty service providers who are talented and reliable since there is no platform where people can share reviews. Also, for home-based beauty service providers, it is difficult to keep steady workload due to limited knowledge and resources on advertisement. Communication between customers and home-based beauty service providers is ineffective. Resolving this problem would generate more business opportunities for home-based beauty service providers and high chances to find better beauty services for customers.

BeautyHub is a B2C platform where people can easily post and/or search for home-based beauty services such as eyelash extension, hair, and nail care. Through BeautyHub, home-based local beauty service providers would be able to promote their business without needing to have an expansive advertising budget and post their job availability effortlessly. Also, BeautyHub is designed to facilitate effective booking system between home-based beauty service providers and customers with simplicity. Additionally, a reliable beauty service community can be established by sharing customer reviews between users.

The main parts of BeautyHub are as following:
● Appointment(scheduling) system
● Search system
● Review/rating system

In order to implement these main parts, the following technologies and services will be used:
● Domain – Heroku
● Web application deployment model – Heroku
● Database – MongoDB
● Web development languages – HTML, CSS, Bootstrap, JavaScript, Node JS, Express, jQuery
● Control repository hosting service – Github
● APIs: Google Maps

Neiybor Growth Features

Name(s): Sean Bird and Michael Hansen
Faculty Mentor: Jeff Jenkins

Executive Summary

Overview

Neiybor is a peer-to-peer self-storage platform with over 2,000 users in 25 states. Recently featured in TechCrunch and Business Insider, we connect people who have extra space and want to earn easy additional income with those who like to save 50% on self-storage. We already have a functional platform but are looking
for improvements by adding features that will facilitate higher conversion rates.

Problem

We are looking to grow quickly. That means we need more traffic, higher listing conversion rates, and higher renter conversion rates.

Solution

Our solution is to create the most growth we can in 300 hours of work, by doing the following:
1. Build a localized page intended to rank highly on search engines for “storage in city-name”
2. Build an interface that allows hosts to easily market their space on Facebook, Twitter, or LinkedIn
3. Build an interface that allows potential renters to record their storage needs and then advertise their need on Facebook, Twitter, or LinkedIn and bring more users to our marketplace
4. Build a listing earnings calculator to generate interest and improve listing signup rates
5. Integrate the Facebook mutual friends API with our search feature to display common connections, create a sense of trust, and help more potential renters convert
6. Building a new homepage that is cleaner, has better messaging, and will increase conversion rates. Gamify the referral program with a rewards tracker.

Tech Stack:
• Ruby on Rails
• React.js with Typescript
• Hosted on Heroku

Dollar for Good

Name(s): Blake Dowling, Hokino Armstrong, and Jacob Fredericksen
Faculty Mentor: James Gaskin

Executive Summary

Our project, Dollar for Good, is a web application that enables users to easily make donations on a one-time or recurring basis to nonprofits.

The problem that this project will solve is that donating to nonprofits can be difficult and time-consuming. Many people want to do good in the world but do not want to take the time to find nonprofits, navigate nonprofits’ websites to figure out where the money would actually go, decide which nonprofits are preferred, and make donations. Even if someone does go through the process of selecting a nonprofit and making a donation, it is very easy to make a contribution once and never come back to make another donation (or to only come back infrequently, despite a willingness to donate again). In short, there is no one-stop shop for accessing nonprofits’ information and for making small, painless donations frequently with minimal configuration required. With no place to make these types of donations, the maximum potential for nonprofit donations is not being met.

To solve this problem, we created a mobile-responsive web application that enables users to easily donate one dollar (or an amount of the user’s choice) each day to the feature nonprofit or one of their favorited nonprofits. Donations will be made automatically on a recurring basis if the user chooses this donation setting. Nonprofits are categorized and browsable, and each has a profile with relevant information about the nonprofit. Featured nonprofits are also available. Users are able to view their own donation activity as well as that of their Facebook friends and of the public (to the extent that other users allow their donation activity to be visible).

We developed the web application using Ruby on Rails, building on a project that was previously started using Ruby on Rails and later abandoned. We developed the front end using HTML, CSS, and JavaScript. We integrated with Facebook for authentication to enable users to easily sign up and see friends’ donation activity, and we integrated with Braintree/PayPal to process transactions.

Building this application and promoting it efficiently will enable all kinds of different people in the world to donate to nonprofit organizations and contribute to the good of society.

Light Hat Security

Name(s): Ben Schulthies and McKay Daines
Faculty Mentor: Jeff Jenkins

Executive Summary

The problem we attempted to solve with our capstone project is that most people don’t know much about computer security, how to protect themselves, or have a place to learn. We set out to make cyber security education simple, free, and accessible to everyone.

Our project is a simple WordPress site that contains educational videos we created about various cyber security concepts. These videos teach basic computer security principles in a very simple and straightforward manner. Content is free and accessible without requiring an account or logging in. Unlike competitors, our content is not lodged into a long and complicated course. Our goal was to make the videos simple enough that users can jump in anywhere and still comprehend the topics presented—and we believe we have done just that.

The technology stack we used includes WordPress, VideoScribe, Audacity, and YouTube to create videos and host them online. Videos are uploaded to YouTube and embedded into the site.

Each video is about 5 minutes long (plus or minus a couple minutes) with off-screen narrating, simple animations, and occasional subtitles to help viewers maintain focus.

AWS Course

Name(s): Stephen Godderidge, Jason Smith, Drake Loud
Faculty Mentor: Tom Meservy

Executive Summary

Problem

Information Systems students are graduating BYU without a deep knowledge of Amazon Web Services. In addition, some 2nd -year MISM students lack adequate AWS knowledge to actually implement their capstone projects.

Solution

We designed a project-heavy course to prepare students both for their capstone projects and for the industry itself. This class will give students exposure to ~10+ projects on a variety of AWS topics and services, such as Lambda, CloudFormation, Machine Learning, Kinesis, Mobile Hub, IOT, Serverless, and others. This class will also give the students the beginnings of a knowledge base to become an AWS Certified Solutions Architect. While this class won’t directly incorporate the certification into the course work, it will prepare students to take the test on their own with additional work on the end.

Project Details

We start the project off by 1) surveying prospective students to find out which topics they are most interested in, 2) meeting with industry professionals about what skills would be most useful, and 3) refining the schedule accordingly. In the following iterations, we created, tested, and refined labs on individual topics, and finally created guidelines for a final project.

Need For Feed

Name(s): Devin Rasmussen, Tyler Doermann, Spencer Stolworthy
Faculty Mentor: Conan Albrecht

Executive Summary

Food truck ownership and operation is all about location. However, most food truck owners have little to no interest or expertise in bargaining for and pursuing appointments at locations. Because of the tediousness and scaling challenges of coordinating between food trucks and venues and events, a separate market exists: scheduling. Currently schedulers are equipped neither to communicate with nor to efficiently schedule trucks to locations in a timely manner.

In Salt Lake City, there is a dominant player in the scheduling market, but trucks don’t like working with the company because of its high percentage cuts and uncooperative spirit. Newer scheduling companies are beginning to gain traction, but are operating in google calendars and rudimentary websites with google forms. Furthermore, they are having to spend hours calling, emailing, and texting different trucks to try and fill slots.

Need For Feed aims to produce a minimal viable product that will address the following:
1. Scheduler calendaring needs in the browser
2. Food truck owner notification of an invitation on mobile
3. Site/Venue Representative needs to request a food truck be present

The technologies we will use are as follows:

1. ReactJS for the web portal that serves schedulers and site representatives
2. Loopback, an open-source Node.js API Framework
3. PostgreSQL

Leveraging these technologies, we will introduce the possibility of a new ecosystem for schedulers and food truck operators and make the food truck scheduling more streamlined and enjoyable for all parties.

House Rules

Name(s): TJ Murphy & Michael Case
Faculty Mentor: Dr. Greg Anderson, greg_anderson@byu.edu

Executive Summary

House Rules is a web-based application that allows users to play card games in a virtual environment. We have found that people enjoy playing games online or on mobile devices because it often speeds up gameplay, allows the user to play with other users in separate locations, and multiple games can be played from one device. We have also observed that people like playing games at home because they have more interaction with friends and family, players can create and implement “house” rules, and people like the feel of a real-life game.

Our goal is to combine the best features of virtual (online and mobile) gaming and traditional card games. We will do this by providing a platform where users can create, modify, and delete rules while still being in separate locations from their friends and family. This solution will also significantly speed up gameplay while still allowing for interaction between players.

We plan to create a web-based application that is mobile-first. In order to use our application, users will need to have at least two devices: 1) a device (preferably a computer or tablet for the larger size) that functions as the game board and 2) a device (preferably a smartphone) that functions as the player’s hand. The user (host) will create a game and if desired, allow other players to join. Gameplay then continues as players use their smart phones to interact with the virtual game board. We will focus on speed as we realize that this is a necessary contributor to the success of the project.

Key technologies include: Firebase, Vue.js, AWS (S3), HTML/CSS, Bootstrap, and Material.io. We have implemented a serverless architecture. We hope to provide a quality solution to families and friends that wish to stay connected with those they love and care about. We hope to provide a quality solution to families and friends that wish to stay connected with those they love and care about.