WoW HackathonCompanies and people around the world are developing game changing solutions with applications powered by Watson and AlchemyAPI services. We recently hosted a hackathon at IBM Watson’s signature event, World of Watson (WoW), in Brooklyn, New York. Almost 170 developers water-taxied across the East River to the Duggal Greenhouse, where they were given 48 hours and full access to all Watson and AlchemyAPI services to build innovative cognitive applications. These developers leveraged any of the 15+ services, individually or in combination. The result was 40 unique applications that combined APIs such as Trade-Off Analytics, Personality Insights, and Sentiment Analysis. The use cases for these applications range from empowering middle schoolers to choose the best high school based on their personality, to facilitating connections between like-minded people based on their geographic location. In just 2 days, developers created powerful cognitive applications that will improve lives. These services are not solely limited to hackathon projects. In addition to anyone having access to our APIs on Bluemix, our Watson Ecosystem team supports partners in using these same cognitive services to build commercialized applications. In addition to teams who were having their first experiences with Watson services, some of our existing partners were keen to experiment with new and existing services. Amanda O’Malley caught up with one of our partners to get the scoop: Philip Mortiboy, Software Developer of Decibel Music Systems Ltd What were your expectations pre-hackathon and what are your reactions post-WoW? Our expectation was to get hands-on experience with tools that we had not yet had the chance to use during our time as Ecosystem partners, and share ideas with fellow competitors. Having competed in the hackathon, we’ve come away with a lot of interesting concepts to develop further and a greater understanding of the tools available to help us do so. What did you build and with which APIs? We used AlchemyAPI’s Concept Tagging and Keyword Extraction tools to find key concepts and themes in song lyrics. The most relevant of these were then sent to the Music Geek API, a music recommendation service that Decibel is developing with IBM, which generated a playlist based on these concepts. What surprised you about the APIs? The depth and breadth of APIs available was a surprise. We’ve previously been focused on APIs that directly contribute to the work we are doing with Watson as an Ecosystem partner, so it was great to have the opportunity to explore tools such as the Visual Recognition API that we wouldn’t normally get the chance to use. Seeing these tools in action has certainly given us an appetite for finding interesting ways to use them within Music Geek. What was challenging and what was easy about the building process? The Bluemix platform made starting development very easy. The ‘plug and play’ nature of the tools let us try out concepts and experiment with new ideas without much overhead. Understanding the outputs and limitations of certain tools was at times challenging, however, the advice and support of the on-site IBM Watson and Alchemy staff was incredibly helpful. Does your experience during the hackathon influence your partnership with the Ecosystem? (ie: generating new ideas, feeling more at ease with the services, etc) Our experience at the hackathon has given us several new ideas that we hope to incorporate into the development of Music Geek. We are currently exploring the possibility of taking our use of AlchemyAPI’s Keyword Extraction tool further to help us identify key themes of a song. We are also looking at using Personality Insights to help identify and manage author bias in album reviews. As an Ecosystem partner, it was great having a chance to talk to IBM Watson and AlchemyAPI developers about upcoming products and future ideas. Our API services are constantly evolving and expanding – and as Philip mentioned – the unlimited access to all of our services comes with bounds of support. Interested in seeing what you can build with Watson? We have written sample code for you below so you can begin building your cognitive application: Want to try keyword and concept extraction yourself? Use this script with the AlchemyAPI Python SDK to get started! [code language=”python”]from __future__ import print_function from alchemyapi import AlchemyAPI import json alchemyapi = AlchemyAPI() demo_text=’Yesterday dumb Bob destroyed my fancy iPhone in beautiful Denver, Colorado. I guess I will have to head over to the Apple Store and buy a new one.’ response = alchemyapi.keywords(‘text’, demo_text, {‘sentiment’: 1}) if response[‘status’] == ‘OK’: print(‘## Keywords ##’) for keyword in response[‘keywords’]: print(‘text: ‘, keyword[‘text’].encode(‘utf-8’)) print(‘relevance: ‘, keyword[‘relevance’]) print(‘sentiment: ‘, keyword[‘sentiment’][‘type’]) if ‘score’ in keyword[‘sentiment’]: print(‘sentiment score: ‘ + keyword[‘sentiment’][‘score’]) print(”) else: print(‘Error in keyword extraction call: ‘, response[‘statusInfo’]) response = alchemyapi.concepts(‘text’, demo_text) if response[‘status’] == ‘OK’: print(‘## Concepts ##’) for concept in response[‘concepts’]: print(‘text: ‘, concept[‘text’]) print(‘relevance: ‘, concept[‘relevance’]) print(”) else: print(‘Error in concept tagging call: ‘, response[‘statusInfo’]) [/code] Code contributed by Zach Walchuk

Join The Discussion

Your email address will not be published. Required fields are marked *