We want to make optimized and efficient search for customer analytics. We solved it with a Byte. Saving TONS of storage space and many times faster than conventional approach.
What it does
Our hack simplifies database search reducing the time complexity and space complexity. We leveraged the flexibility of NoSQL databases and created a cutting edge Byte based database where values are just one BYTE and are specific to each document in the collection.
How we built it
We classified every possible column value into one BYTE sized value taking the advantage of flexibility of NoSQL. Reducing the space complexity to greater than 50%.
Challenges we ran into
We had problems with parsing data from MongoLab. One of us had issues starting MongoDB on Command Line which took up a good amount of time.
Accomplishments that we're proud of
We designed an efficient search algorithm a reverse engineering approch of machine learning classification problem.
What we learned
The hack had us learn how to use mlab and highcharts. Kaushik, an aerospace engineering major and beginner programmer used the time to learn the basics of MongoDB from absolute scratch.
What's next for ByteSearch
We are looking into furthuer optimize the problem to a bit level, and improve upon our current progress and implement a proper visualization. For testing purposes, the main focus is generating random information for each of the twelve data categories as opposed to manually inserting data for testing. With a fully functional project, we hope to have ByteSearch working with small to mid-range data driven organizations and potentially expand ByteSearch, possibly into Air Traffic Control and replace Flight Progress Strips.