The web is full of data/information, but it is hard to find the relevant information useful for our interest. What if we could not only narrow it down to relevant data, but also gain personalized set of information?
What it does
Our app scrapes the articles, patents, social media posts on the web and identify relevant keywords from them. Our app predicts what technologies will become commercial relevant in the near future.
How we built it
We preprocess the collected data from the web using the Datapoint utility provided by Viacom. We thought it is challenging to use the Datapoint, so we used the GraphQL's syntax to utilize the Datapoint. Then, the google cloud platform's Natural Language API uses machine learning process to identify key terms that we are interested in, and it also tells us the salience of the word. We used these keywords with high salience as search term to perform deep, recursive search for expanding our search range and orienting the search towards a direction. We used predictive model to determine which technology would be more commercial relevant than the others.
Challenges we ran into
We had to come up with a novel algorithm to evaluate the tailored data we prepared to tell the user unique and meaningful information about the technologies future.
Accomplishments that we're proud of
We deal with large quantity of data using datapoint utility and manage to use recursive queries to expand the range of search. We implemented datapoint utility using the GraphQL's syntax.
What we learned
We learned how to work as a team moving towards to a goal quickly.
What's next for Salience
We are going to promote the data integrity of the web to easily identify meaningful information. We wish to promote diverse, healthy expressions of people.