account groupicon_account_group_smallShapeleft arrowlong left arrowright arrowbinocularsbook open 3.3.3.1 2222Open Book Iconbook open 4.3.3.1Open Book with Words iconicon_browser_talkbrowser windowicon_browser_window_smallIcon / Digital Businessicon_cloud_outlineicon_cog_bulbcompatible devices 2.3.3.1computer media.3.3.1icon_credit_cardscompatible devices 2.3.3.1icon_envelopesFacebookIcon / Digital BusinessIcon / Product Engineeringicon_graph_baricon_graph_bar_smallgraph line 2.3.3.1account group 3.3.3.1icon_heart_handshakeInstagramlayout 8.3.3.1LinkedInPage 1 Copy 2mobile phone portrait.3.3.1Mobile Phone (iPhone) iconIcon / Digital BusinessOutline_Icons_1_network.6.3.1icon_network_circlenewsletter iconOutline_Icons_1_account group 1.3.3.1Presentation IconProduct Iconquote 1.2.3.1quote 1.2.3.1 copyregister machine.3.3.1shopping cart 4.3.3.1Twittervideo clip 1.3.3.1Video player iconGroupGroup 2

Timeframe

5 weeks

Challenge

We wanted to explore uses for machine learning, so we decided to create a tool to predict availability of Capital Bikeshare stations depending on location, time of day, and weather.

Solution

We used the large amount of trip data DC's Capital Bikeshare has made available to the public, combined with other historical data, and applied statistical methods. To increase accuracy, we added weather data to the learning system.

Blog Posts

Results

We quickly built a web page and a REST API that collects user input and displays predictions for the given location, date, and time. Next, we decided to add a conversational interface so we can access the API using natural language. We also added a Slack bot integration so that users can access the solution right in the chat window.

Our Latest Projects

Augmented Reality Apps

We created a prototype AR app for the iPhone that shows users where the nearest Capital Bikeshare station is relative to the user’s location and directional orientation. The AR overlay shows the name and distance of each station.

Read Our Lab Notes

Slack + Beacons

We got a few RadBeacons to play around with, so we integrated them with Slack to create a "check-in" feature that alerts team members via Slack when someone gets into the office.

Read Our Lab Notes
No results match this filter.