
AI Blueprints
By :

In all of the previous chapters of this book, we have developed AI solutions that operate behind the scenes, in a way that does not allow direct user interaction. For example, in Chapter 3, A Blueprint for Making Sense of Feedback, we showed how we could measure the overall sentiment of users based on their tweets and comments, but this user feedback was collected passively rather than directly asking users for their opinions. In Chapter 5, A Blueprint for Detecting Your Logo in Social Media, we developed a technique to detect a company's logo in random photos, but the people who took those photos did not have any direct interaction with our logo detector. Up to this point, we have explored several ways that AI and ML can help make sense of a big pile of data, be they images, tweets, website clicks, or song plays, for example. But the interactive prospects of AI have remained unaddressed...