fchongg / chum-bucket

-# Chum-Bucket

Inspired by: Our own difficulties finding links from past conversations, scrolling for days on end, we decided to tackle building a solution to our problem.

What it Does: Chum-Bucket pulls up links from a conversation that specific to a topic the user gives it, using natural language to parse through HTML pages using Azure Search.

How We Built It: Using Luis.AI for its natural language capabilities, hooking it up to a Bot framework, and using Azure Search to parse through the data from the links.

Challenges We Ran Into: Linking all services together.

Accomplishments that I'm Proud of: Learning new technologies in only 24 hours.

What We Learned: A whole lot from helpful mentors and peers, especially about the tech services that Microsoft offers.

What's Next for Chum-Bucket: We are hoping to add more functionality in the future, like searches where we can filter using time date frames. We are hoping to integrate the cognitive services that Microsoft offers, where we can search up topics using Bing Search, and save the links that Bing Search finds on Slack.

Built With: Azure-search bot-framework luis.ai javascript node.js

How to View if you are a UBC-MSFTHacks2017 Attendee:

• Send a direct message to frances-bot on Slack or invite frances-bot to join your channel.
• Type in links (ex. https://en.wikipedia.org/wiki/Butterfly or https://en.wikipedia.org/wiki/Muffin) in to your conversation.
• Add more links to your conversation if you'd like.
• Type "search links for caterpillars/monarch/butterflies (or whatever topic your links pertain to)" and WATCH THE MAGIC HAPPEN as it parses the link to check if the topic is relevant to the page.