Machine Learning, Data-Visualization and Social Media Preferences

Hi, I plan on sharing more in-depth material here. However, to kick things off, I'll just share a few of the articles I've enjoyed this week.

I'm a big fan of Spotify and I've found some great music through Spotify's curated playlists. There were times I thought the playlists were becoming more and more repetitive and predictable, probably a content library issue. However, Spotify gets points in my book for employing machine learning algorithms to find relevant music. 

Spotify’s Discover Weekly: How machine learning finds your new music

Not so much of a blog post as a press release, but Google announced Teachable Machine, which lets really anyone with no technical knowledge learn more about machine learning. It's much more of a game than anything else, but I plan on exploring the tool further. 

Here's a screenshot of the tool:

Is Google warming us up for the post singularity apocalypse? 

I've updated The Digital Abstract with a few items this weekend. Here they are:

Facebook is now offering food delivery. They won't be replacing any service, but they serve the connection between the customer and service like DoorDash or GrubHub. 

Sorry, but there's a common threat here - machine learning. I've been fascinated by it and I'm really looking forward to seeing what machine learning will enable us to do in the next 5, 10, and 15 years. A big hope of mine is for medical cures. That's another blog post.

Throughout my research, Accenture has surfaced as a leader in the machine learning space. There's a pretty interesting SlideShare from Accenture Consulting that dives into machine learning and how it can improve banking, including detecting fraud, limiting risk, and staying compliant. I wrote about this here.

Another machine learning item that's really cool is the experimental "Living News" tool. To put it briefly, it's a data visualization showing the subjectivity and sentiment of news content delivered in real time from various sources. The real-time news feed is analyzed by a machine learning library to analyze the content. It was developed by Jibin Varghese at the National Institute of Design, India.

Here's the prototype that was published:

The last article I'll leave you with is Dead-End UX: The Big Problem That Facebook, Twitter, And Others Need To Solve. The author makes a solid point. Preferences upon preferences upon preferences back us into this corner where we are trapped. That's currently how Facebook and Twitter work. Sometimes we don't want, and especially need, information based on our past preferences. 

Social media activity is a good indication of how we've changed in time. Who we unfollow, the accounts we visit, and sentiments we communicate. Maybe machine learning will one day allow social platforms to adapt as we as adapt and give us a fresh persepctive when we've preferenced ourself into a corner.