BrainStation Project Showcase

By BrainStation August 26, 2020
Share

After 12 weeks of accelerated learning, collaborating, and creating, BrainStation Diploma students are able to showcase everything they’ve accomplished at our Virtual Demo day. 

Here are some of our favorite projects from Spring 2020. Don’t forget to take a look at all of the latest graduate’s work on our hiring page.

Bi-Modal Sentiment Analysis

Syed Fatir Ali, Data Science

As humans, we perceive the world through a myriad of different senses. The “data” that these senses provide to our brain helps us create our “reality”.

Ali designed a machine that could replicate how human’s perceive the real world. To create more accurate Natural Language Processing (NLP), machine learning models should be able to utilize multiple sources of data, so Ali’s project aimed to create a real-time sentiment analysis classifier using text and audio data concurrently.

Watch Ali demo his project on the our hiring page.

Tabby

Brendan Kirby, UX Design

Tabby is a crowd-sourced cat-sitting service and community intended to improve the accessibility, availability, and affordability of pet-sitting exclusively for cat-owners.

Kirby’s app offers direct messaging, cashless payments, and on-demand appointments to ensure a street-free process from beginning to end.

See more mockups on our Hiring page.

YYRO

Weizhen Liang, Web Development

Liang created YYRO, a fully functional and ready-to-deploy gaming website. The site has all the essential elements required for a gaming website and is dedicated to bringing players together in an active, friendly community.

The site is fully responsive for mobile, tablet, and desktop and compatible with iOS and Android devices.

See what the gaming site has to offer.

ImageCap

Varsha Rai, Data Science

ImageCap is a tool that generates captions from images using neural networks. Rai fed image feature vectors and partial captions to the model so it could predict the next word of the caption.

To process images, Rai used a Convolutional Neural Network (CNN) and the GloVe model to process words in the caption.

Partial caption and image feature vector pass through the Recurrent Neural Network (RNN) and the model picks up the next possible word with the highest probability and generates the next sequence.

Watch Rai explain her full process.

WeClick

Samiya Dalwai, UX Design

WeClick is an app that helps newcomers to Canada connect with one another through interest-based activities and 1-on-1 meetups. The designs show that activities and people are recommended to users based on compatibility and interests as a way to help those who’ve recently emigrated to Canada make meaningful connections and find their community.

Watch Dalwai explain her design process and review her portfolio.

JOGA

Steven Martins, Web Development

JOGA is an app that allows users to find sports pick-up games created by other users. They can also create their own games for other users to join.

To leverage local park infrastructure and green space, Martins created JOGA to cater to those looking to play soccer, basketball, and street hockey, but as the community flourishes, more sports can be added to the platform.

See more features on JOGA.

HUGGZi

Bonney Halani, Digital Marketing

New parents don’t have a lot of time on their hands, and choosing the right products can be a daunting endeavour.

To ease some of that burden, Halani developed a marketing strategy for HUGGZi, a monthly subscription box service specially tailored to fulfill your baby’s needs with a curated selection of the best products.

Find more digital marketing final projects.

Airbender Project – Airbnb Listings Recommender System

Rey Reynaldy, Data Science

Rey loves travelling, but found it difficult to select the best place to stay on Airbnb.

For his project, Rey built a recommendation system for Airbnb listings. Using content-based filtering methods, user preference, similar content, and similar user preference listings he was able to curate a preferred Airbnb list. NLP process was used during exploratory data analysis and sentiment analysis technique was utilized to determine the score of the user’s review content.

Learn more on Rey’s Github.

SpellBee

Jessica Kwok, Web Development

As a new mom, Kwok wanted to make use of technology in a fun, educational way, so she developed SpellBee to help kids learn and practice how to spell.

Kwok recognized that parents are seeking tools that they have control over, but also expose their children to learning opportunities. SpellBee does just that by allowing parents to create and edit their own vocabulary decks.

See more web development projects.

Noeme AR

Wynnie Chung, UX Design

Noeme AR is an augmented reality application to help patients record and showcase their pain symptoms as a way to enhance patient-doctor communication and collaboration.  

Chung’s application reimagines the traditional paper-based, pre-assessment process into three-step, digital AR experience that captures the location, intensity, and direction of pain on the patient’s body. By providing a baseline of quality information on a patient’s condition, Noeme AR helps doctors and patients quickly enter into meaningful conversations towards diagnosis and plan of care.

See all the features of Noeme AR on the Hiring Showcase.

Movie App

Sana Krichen, Data Science

Sana designed an app that predicts movie genres and detects spoilers in reviews. The technical process requires scraping data from the IMDB website, using supervised machine learning techniques along with natural language processing and implementing some text similarity techniques.

The end result is an app that can predict what genre a movie falls under by reading and learning from movie descriptions.

Meet Sana on our Hiring page.

Svara

Ryan Todd Garza, Web Development

Svara is a generative music app based on the traditional rules of Hindustani ragas, a melodic framework for improvisation largely used in Indian classical music.

Music-lovers using the app are shown frequency and amplitude feedback along with any other relevant information about the raga that has been randomly generated.

Watch Ryan demo his project on our Hiring page.

micdrp

Angus Ryer, Web Development

Micdrp is a voice-to-confidence converter. The idea is that if you sing specific notes, you’ll strengthening specific muscles in the process. Angus’ app, micdrp guides singers through what notes to practice, and the result is power, clarity and confidence while belting out tunes.

See more on our Hiring page.


Want to create your own tech portfolio? Speak to a Learning Advisor to find out how you can take the first step towards your new career in data, development, or UX.