2018 Projects

Is your project missing? Edits? Corrections? Fork our repo.

Emotiquote (Won Best Visual Award!)

Made by Abigail Hodge, Michelle Pine, Sharon He, Sam Price

Our interests in machine learning and web dev/UI prompted us to develop a web application which handles the Indico API and Mediawiki API to enable emotional analysis of very large and interesting datasets, namely, famous quotes (drawn from Wikiquotes).

CoOperate (Won Hack to Give Back Award!)

Made by Campbell Alden, Kimberly Whitney, Jessica Rouzan, and Adrian Kant

Like TRACE reviews for co-ops but anonymous

Nudge (Won Rookie of the Year Award and Genuine’s #InventTogether Award!)

Made by Sebastian Bernal, Alejandro Baez, Marc Estrada, Grigory Kimatov, and Rocktem Bhowmik

Have you ever fallen asleep on a bus/train and then missed your stop? Sleep deprivation is a common problem, especially among students. So, we created a mobile application called Nudge that is able to wake you up with a customizable alarm shortly before you reach your destination.

Scout Robotic Pet (Won Most Technically Challenging Award and TriNetX’s Best Use Of Data Award!)

Made by Devendra Shivraj and Mark Yamarone

We made a small robot with a camera and a couple sensors on it, to allow it to find and follow it’s best friend: you!
The inspiration was a robotic pet that can’t leave you alone.We made it because we wanted to do a challenging hardware hack that would be fun for us to build and code. We used a Raspberry Pi, Arduino Uno, Amazon AWS and lots of jumper wires. (plus love <3)

cap.py (Won PowerAdvocate’s Most Useful for the Everyday and Rough Draft Venture’s Most Entrepreneurial Hack!)

Made by Daniel Chen, Abel Shin, Elizabeth Cho, Caitlin Wang, Brandon Yip, and Duk Hwan Kim

Uploading a photo to your favorite social media platform is always a struggle when you can’t think of the perfect caption. With cap.py, never worry again that your caption is holding back your image from getting the attention it deserves.

Start by choosing the picture you want to caption. After uploading, cap.py will pair your image with a relevant quote and hashtags (+ bonus GIF while the caption is generating). If you think that the first caption doesn’t fit your needs, you can regenerate another.

Built with Flask/Python, Google Cloud Vision, BeautifulSoup, HTML/CSS/JS.

Bringing Genius to Spotify Desktop

Made by Shane Timmerman

Not many people know that the Spotify desktop app has a designated full screen mode. Even fewer people actually the bland and tasteless feature. On accasion I have found myself listening just putting on a song on computer and just wishing I could do something more with my screen space, similar how Spotify’s android apps use their space. Putting a highlight on album art and real time lyrics display, similar Genius feature on their mobile app, I built a minimal player to help proof out the concept. The backend is written in Python on a Windows platform, while the front-end was coded using Kivy. This helped me play around more within the front-end development space.

Viewed in another light this could be a karaoke application synced to your Spotify account. I don’t really know how I feel about that.

Genuine Dogs

Made by Michael Fleisch, Becker Ewing, JdD Mullally, Travis Bergmann, and Brandon Gioggia

We made an app to recognize the breed/breeds of your dog using the Indico API. We have future plans to expand this idea to allow people to reconnect lost dogs with their owners.

Scavvy

Made by Kate Rupar and Mia Fryling

We created a game using Apples CoreML and Vision framework. The game is essentially a scavenger hunt, where the user is given the name of an item to find in 30 seconds and they only gain a point when the image recognizers sees it. There are 3 modes available, each based off of a different mlmodel; Food, Emotion, and Objects.

Aerodust

Made by Dennis Giese and Andrew Tu

We used recently rooted COTS vacuum robots to map Wifi networks and rooms. In addition to that we played around with different usecases (webradio, etc).

Connect Zero

Made by Kevin Luo and David Zhao

Alpha zero based connect 4 ai

ZodiHacks

Made by Kelsey Esposito, Olivia Petrillo, Caitlin Rossi, Willa Shiomos, Anjali Thakkar, and Sarah Witt

Our project is a website that predicts a person’s zodiac sign based on their text input. We like to have fun and thought this was a fun idea for a project! This is our first hackathon so we tried to think outside of the box and do something different and doable given our limited experience and the time constraints. We started this project by using Indico’s text analysis API for personas for Python. Then, we used Flask, html, and css to get the website up and running.

RedditMyersBriggs

Made by David Malakh and Skyler Szot

It uses the Indico API to analyze Myers Briggs personas of self text posts on specific subreddits. (Reddit data was collected through the PushShift Reddit API). This project allows anyone to see some interesting data about their favorite subreddits and what the personas are of people posting on them.

Glenn’s Game

Made by Catalina Berretta, Felicia Zhang, and Jenn Der

For our friend/roommate’s 20th birthday, we decided to code her a game based on her life, sort of a “Choose Your Own Adventure” filled with inside jokes and silly references (i.e: the final boss is Barry Dwight, a porcelain duck one of the team members brought from Portland that everyone hates and who they’ve all started a conspiracies against) None of us have ever worked with Unity before but it was the best framework for what we wanted to do so we just dove in. What we have is the basic skeleton of it and we plan to add more scenes before her actual birthday (March 11).

Connect Partify

Made by Durward Benham and Amina Ly

A service that allows you to take an existing Spotify playlist, and optimize it’s play order for parties and other events

Raven Messenger

Made by Kyle Sferrazza, Ian Wolterstorff, Trevor Stenson, Bryce Thuilot, and Ryan Drew

We (almost) made a messaging app for Android that generates secure One Time Pads (OTPs) as scannable QR codes, and then allows secure messaging with friends using those OTPs.

Emosic

Made by Karan Tyagi

Used Microsoft Azure Emotion Detection API and python.

Polarizing Opinion Sentiment Analysis

Made by Olivia Wan, Crystal Rhee, and Sarah Zhang

Uses an indico.io model trained on for/against arguments for various divisive topics such as gun control, climate change, etc. to predict the viewpoint of a given text passage.

Blockchain Messaging

Made by Edward Shen, Jacob Bernhardt, Nithin Saripalli, and Khalil Haji

Java instant messaging with blockchain encryption

Nature Calls

Made by Josh Hoffman, Justin Konecny, Matt Martin, and Steve Setteducati

Imagine you are touring NYC, you have been enjoying the day, eating a ton of food and staying well hydrated. Suddenly, Nature Calls, and you have no idea where the nearest bathroom is. We hope to solve that, we attempted to make an app that will take your location, and show you the nearest bathrooms. So when nature calls, you know where to run to.

Mood Song

Made by Joseph Gabay, Axel Simms, Emily Hontoria, and Anjaneya Naik

Using a webcam, this app takes a picture of your face and uses the Indico API to detect your mood. From here, it matches your mood with a song from a pool of over 180 top songs that have had their lyrics analyzed with the Indico API. It then plays your ‘mood song’ through spotify.

Lucid VCS

Made by Milo Turner, Zach Casey, and Kira Evans

Version control system, controlled by interacting with a text file on emacs, uses “patch theory” to calculate merge, commute etc.

Dadabots

Made by CJ Carr, Zack Zukowski, and Bryan Robinson

Used a recurrent neural network to train a raw-audio model on punk band NOFX, generated a song that definitely sounds like NOFX but with nonsense lyrics.

Get Together

Made by Colton Donelly, Aditya Jalan, and Daniel Melcer

An event management app which covers every aspect of an event, ‘from organizing an event to reaching the venue on time’ The app, after 2 nights of work, boasts of a login page with persistent login capability and secure password storage on server, as well as a task assignment functionality, aimed at making an event more efficient by distributing tasks among the attendees. While the client side of the application does not yet support task assignment, the server is prepared to assign tasks within the group to specific members or to the group as a whole. The application gives the user the control to create his own private ‘party-space’ to which he can invite other members through a personalized code, and assign tasks and requirements for the party to fellow attendees.

Future prospective improvements include but are not limited to event decisions within events, for e.g a Pepsi v/s Coke poll; Venue mapping and Uber/Lyft integration within the App; ‘Wishlist’ integration; option to look up and buy tickets to public events; dedicated group chats and inter-personal chat bubbles; calendar integration for appropriate reminders of events; and even the functionality of booking venues through the app, in the long run, among tons of other features.

CourseOn

Made by Prisca Joseph, Sarah Wessel, Rucha Khanolkar, and Megan Lau

CourseOn is a web application to help students to organize which courses they need to take throughout their time at Northeastern. This allows students to generate their grad plan easily, dynamically, and interactively.

pupDate

Made by Anna Choi, Andres Rodriguez, Gabriella Roman, and Christopher Trinh

A web application for dog owners to find their dogs some nearby playmates, since Anna owns a dog and is always looking for playmates! Made in PyCharm.

Joe & Pedro’s

Made by Pedro Silva and Joe Donovan

Our project identifies your music preferences based on how your twitter compares to those of artists’ fanbases

Live sentiment analysis

Made by Divyavijay Sahay, Brian, and Tram

We created a system which can perform sentiment analysis on live video and sound. This system can be used in numerous ways like for psychiatrist consultation, audience reaction and also with some more functionality blind/deaf person can use this system to analyse their surrounding.

Chlobber

Made by Derrick Duncan

Command-line utility that helps you restore your lost Chrome browsing sessions and tabs

PaperStx

Made by Jakob Hain

An editor for block syntax, like the syntax in Scratch. It’s a webpage/webapp whose custom logic is entirely client-side (e.g. you don’t need a server to run it, you can just load the HTML file). The block syntax has basic types. Not every block can be connected to every other block. The blocks are colored, and they’re have colored “holes” for other blocks - the only blocks which can go in a hole are the ones with one of the holes’ colors. First, you create a template file. You follow a very simple DSL to either create your own block syntax or recreate an existing one (e.g. Scratch). Then, you upload this file to the webpage. Then, the webpage will give you all of the different block syntax forms, grouped by class. You can drag them onto a canvas, and create your block syntax.

jgka

Made by Megan Larson, Dan, Julian, and Tyler

dawd


Follow the nice people at @hackbeanpot for more updates about HackBeanpot!