Inspiration
While working on projects with a team, it can be hard to keep up with around-the-clock electronic communication. It is difficult to enter a conversation midway without feeling lost, and searching through an old conversation for specific information can be tiresome. We set out to create an app that would make team communication a more efficient and enjoyable experience.
What it does
The app takes all the information of a group chat and synthesizes it into a chat summary. The components of the summary include common topics that have appeared in the chat, dates that have been scheduled, and tasks that members of the chat can assign to themselves or other people. These functions put all of the important information in one place, making group work and discussions simpler.
How we built it
We started by brainstorming ways to quantify the information in a conversation and thinking of how we could synthesize it into an easy to understand summary. With the algorithm in mind, we began creating a template for the chat login and interface pages using HTML and CSS, with styling from Google's Material Design specification. At the same time, parts of our team worked on integrating the template with Firebase to provide user authentication and a database for group conversations. Finally, we pulled it all together, creating a responsive, clean interface with robust error-checking and unique chat-summarizing features.
Challenges we ran into
We had challenges using the Watson Concept Insights, which would have allowed us to analyze text and come up with keywords of the conversation to show the user. We also were originally planning on allowing users to login with Facebook, and we were able to prompt a Facebook login box, but were ultimately unable to integrate the login with our website due to recurring errors. Thus, we developed our own user account entry system with Firebase.
Accomplishments that we're proud of
We are proud of the clean interface of our app, the ease of use, the asynchronous updating and tracking of online users with Firebase, and the innovative idea.
What we learned
We learned that it's hard to predict what can go wrong when using an external API, and that user experience and functionality are both important; we should never focus on one at the expense of the other.
What's next for Chatl;dr
We want to integrate better machine learning algorithms and natural language processing, using MEAN stack to allow for Bluemix product functionality. Additionally, we would like to create a full-featured chat interface that will reach the full capacity of our innovative idea.
Log in or sign up for Devpost to join the conversation.