Inspiration
In this era, information surround us forging our sense about the world's truth. But the problem with information, is that it can be manipulated by those with large communication reach. This means, that the people usually give the quality of "vox populi" to sources that control the information. The traditional information spreading schema can drive us to develop unfunded biases and make decisions that harm us based on manipulated information. The ideal scenario for information spreading would be to provide enough sources to individuals and let them build their own criteria while identifying the different poles and elements of an information corpus. The problem with that scenario, is that humans are not good processing large amounts of data. For this reason, I think that technologies should help people understand any subject in order to build an objective criteria for the social good.
What it does
Crowd Status helps you understand Twitter's crowd opinion about any subject by:
First: Classifying the information between different topics about your queried subject providing you the main entities for each of them (Available topics: COMMERCIAL ITEM, DATE, EVENT, LOCATION, ORGANIZATION, OTHER, PERSON, QUANTITY and TITLE)
Second: Sorts the information by popularity showing you the most popular information first and determines the sentiment polarity for each topic (i.e. information containing words like "good", "great" or "happy" will be rated as positive, while information containing words like "bad", "hate", or "death" will be rated as negative)
Third: For each topic, it lets you know the most popular phrases or if you want to go deeper and spend more time understanding the topic it lets you listen the full tweets
How I built it
The Alexa Skill was build using NodeJS and AWS Artificial Intelligence services. The skill retrieves information from tweeter, process it and delivers the condensed data to the users.
Challenges I ran into
The biggest challenge that I ran into, it was to fulfill the technical and certification requirements for the Alexa skill. For example, the response size has a limit and you need to manage your response size in order to avoid errors.
Accomplishments that I'm proud of
I am proud of being able to process text with natural language processing technologies in a matter of milliseconds. I found that fact really impressive since it allows me to bring rich information to users really fast.
What I learned
A learned a lot about chatbots development and once again, the importance of unit testing during the chatbot development process. As the users intents could be very variate it is important of keeping track of conditional statements, error handling and basic functionality. Unit testing is the least method that must be applied in order to succeed in with the chatbot development.
What's next for Crowd Status
Due to the time spent in Crowd Status development the current version is only a minimal viable product. Further efforts will be focused on premium features towards its monetization bringing the opportunity to access more valuable information to users.
Log in or sign up for Devpost to join the conversation.