What are people really doing with machine learning?

Machine learning is being built into just about everything these days. At this point, the largest user base of people accessing machine learning directly will probably be from Microsoft Excel.[1] A lot of other office productivity software can invoke machine learning directly. Outside of machine learning being built into applications and software the next place to go looking to figure out what people are really doing with machine learning is to start looking at common use cases.

During the course of my research and analysis related to machine learning about 26 use cases ended up making my list. That list is not exhaustive by any means. It is a good start in terms of thinking about what people are really doing with machine learning out in the wild. A lot of attention has been spent on computer vision machine learning related to the self driving cars use case. Autonomous vehicles are an interesting pursuit and a fine use case for machine learning. During the course of driving a lot of it is very definable and repeatable. A lot of special cases and unique scenarios exist. 

1. Product recommendations

2. Image recognition

3. Account prioritization

4. Lead prioritization

5. Monitoring systems

6. Speech recognition

7. Marketing targeting

8. Fraud detection

9. Inventory management

10. General forecasting

11. Customer experience

12. Voice assistants

13. Dynamic pricing

14. Email management

15. Sentiment analysis

16. Time series forecasting

17. Image analysis

18. Geospatial analysis

19. Streamline data

20. Optimize operations

21. General chat bots

22. Targeted chat bots (Service/Help)

23. Health check monitoring

24. Anomaly detection

25. Sound pattern matching

26. Sales trending

Part of my research efforts were to place those use cases on a chart that evaluated scale vs. maturity. 

Links and thoughts:

I have been really enjoying the Machine Learning Street Talk and this week's episode is interesting “#52 - Adversarial Examples Beyond Security (Hadi Salman, MIT)”

I’m not entirely sure why, but I did watch Linus build a computer this week, “The Fastest Gaming PC... Is THIS BIG???”

I watched a bit of the Microsoft Developer “AI Show Live-Episode 11-New to Anomaly Detector: Multivariate Capabilities”

Top 5 Tweets of the week:

Dr. Nels Lindahl @nelslindahl

I gave this post a read over @Medium today its called @nuxeo : How to approach a version upgrade with a SQL to MongoDB data migration by Mariana Cedica

Andrew J. Hawkins 🚇🚌🚲🛴 @andyjayhawk

so i wrote a story last week about Tesla jacking up the price of its solar roof, and it led to a flood of emails from other aggreived customers who had a lot to say about the company, its products, its mission, and its customer service. so i wrote another story

The Verge @verge

Solar flare: why Tesla’s price hike has customers seeing spots https://t.co/EEQzyj2m34 https://t.co/KOovaWj7uj

David Carroll @profcarroll

The manipulation is right there on the surface. UX industry’s reckoning looms large as designers squeezed out of labor chain in favor of automated algorithmic data-driven design optimized for outcomes without regard for consumer protections. #DarkPatterns

nytimes.comOpinion | Stopping the Manipulation MachinesRegulators should crack down on “dark patterns,” widespread digital practices designed to deceive consumers.

TensorFlow @TensorFlow

🎶 Tune any musical instrument with @ml5js and Crepe. Michelle Sun joins us on #MadeWithTFJS, to tune instruments live in browser (even your voice) without any hardware. She can even play games 🐍 with instruments 🎸 as an input device! See how → goo.gle/32PNbiN

Dr. Nels Lindahl @nelslindahl

From @FortuneMagazine: This A.I. startup is saving Walmart and other big companies millions by automating negotiations. fortune.com/2021/04/27/a-i… via @googlenews

fortune.comThis A.I. startup is saving Walmart and other big companies millions by automating negotiationsThe two-year old company counts prominent members of Europe’s “Skype Mafia,” among its investors. It says its software is saving customers in millions of dollars each month by tackling the “long tail” of suppliers.

Footnotes:

[1] Data Analysis, Machine Learning and Microsoft Excel or check out this link from Venturebeat

What’s next for The Lindahl Letter?

• Week 16: Ongoing ML cloud costs

• Week 17: Figuring out ML readiness

• Week 18: Could ML predict the lottery?

• Week 19: Fear of missing out on ML

• Week 20: The big Lindahl Letter recap edition

I’ll try to keep the what’s next list forward looking with at least five weeks of posts in planning or review. If you enjoyed reading this content, then please take a moment and share it with a friend.