The team drew their inspiration from the desire ot aide local small- and medium-sized businesses in their business and marketing models, via streamlining their advertising process and providing a service that offers the greatest possible outreach based on their advertising budget.
What it does
A platform to connect with companies that require advertisement to companies who own physical advertising real estate, such as billboards. This platform will allow businesses to make data-centric decisions before they choose a billboard location for advertisement.
How we built it
Data was collected from intensive research, data mining, and data manipulation from the internet. This data was then fed into several machine learning algorithms developed in ML Azure and Python, to train a model designed to optimize outreach based on budget capacity.
Challenges we ran into
As a team, we lacked an experienced front-end developer. So, we struggled quite a bit with the front end aspect of the design, and spent the majority of the time data mining for information to feed into the machine learning models, as well as trying to integrate a front end component into the design.
Accomplishments that we're proud of
We're very proud of the clever machine learning algorithms, as well as the team's cooperativeness throughout the whole hackathon. Joining a team of perfect strangers is sometimes hit or miss, but our work ethic complimented one another well, which helped drive us to productivity.
What we learned
We learned the process of setting up the front end website component of a system, as well as subtle hacks on GitHub using Markdown to create a faux "website interface", for demo purposes.
What's next for Billboard-AI
Next steps would be to implement the model throughout more cities in Ontario, and across Canada, once the data becomes available to us. Finding data was a very difficult and time consuming aspect of this hack, and limits the hack from achieving full potential if it was driven into production.