The AI technique that could imbue machines with the ability to reason by Karen Hao via @techreview

“At six months old, a baby won’t bat an eye if a toy truck drives off a platform and seems to hover in the air. But perform the same experiment a mere two to three months later, and she will instantly recognize that something is wrong. She has already learned the concept of gravity.” Yann LeCun, the chief AI scientist at Facebook, hypothesizes that a lot of what babies learn about the world is through observation. And that theory could have important implications for researchers hoping to advance the boundaries of AI.

I definitely agree with his observation on the number of pictures needed for learning to generally take place which makes it NOTHING like the way a baby or young child would learn things in real-life. So unsupervised learning it is then.

Small example of k-means in R:

km <- kmeans(iris[,1:4], 3)
plot(iris[,1], iris[,2], col=km$cluster)
points(km$centers[,c(1,2)], col=1:3, pch=8, cex=2)
table(km$cluster, iris$Species)

How Etsy taught style to an algorithm by/via @FastCompany

Is it romantic or rustic? Boho or minimal? Etsy needed to offer searchers a way to find goods that matched their style aesthetics, but since descriptions aren’t uniform and don’t always describe the style, text mining the descriptions wasn’t enough. Colour and patterns don’t reliably predict style, so image recognition alone didn’t do it either. Enter a model that blends text analysis with image recognition based on 43 human-identified styles.

I love this real-life example detailing the steps they took to work out how to do this. Definitely, a methodology that could be used by other organisations to do a similar type of thing.

MIT: We're building on Julia programming language to open up AI coding to novices by @LiamT via @ZDNet

MIT claims a win with probabilistic-programming system Gen in democratizing AI and spreading innovation for all.

This sounds really interesting and I'm interested in how this will work if it is easier than TensorFlow. I do have a word of caution though - coding it is one thing, getting the design and all the other aspects are something far more complicated so I'm fascinated how this is going to work.

PayPal-backed blockchain aims to help banks verify digital IDs by Penny Crosman via @infomgmt

A new project backed by the government of Luxembourg could ultimately be influential in the U.S., where banks have been slow to develop a shared platform for digital identities.

This sounds like n interesting project which can definitely have a far-reaching impact on the way we are identified in the future for banking and any other financial transaction. However, there are alternatives in development so it's a bit of a race. You can refer to this CCN article to read a little more about them.

A Turbulent Year: The 2019 Data & AI Landscape by/via @mattturck

Matt Turck offers his seventh annual in-depth look at the data ecosystem. Part one focuses on the landscape and issues of privacy and regulation, and part two looks at data infrastructure. He also includes a link to a spreadsheet with hundreds of additional companies.

This is brilliant and needs bookmarks printing out and referring back to frequently.  I love that there are lots of links throughout both parts so that you can follow them and find out more about that subject. Please go and leave him some comments to express how amazing this is.

Cloud savings, simplicity and data insights still elude most firms by Bob Viola via @infomgmt

A major expectation gap exists between what IT managers hoped the public cloud would deliver for their organizations and what has actually transpired, according to a new report.

This is interesting and could give some kudos to those who are currently resisting making the move the Cloud.

More than half of all data remains untagged and unclassified by Bob Violino via @infomgmt

A new study finds that companies have limited or no visibility over vast volumes of potentially business-critical data, creating a ripe target for hackers.

This is definitely an untapped resource within any company and exactly what an unstructured database would be good for was you start to investigate and sort out the format of the data and how to get sensible information from. You might find that with a little investment in the right resource it could become a valuable data source.

WEBINAR: What Is Inclusive Data Storytelling, And Why Does it Matter? - 30 July 2019

Data Science Central Webinar Series Event

What Is Inclusive Data Storytelling, And Why Does it Matter? Join us for this latest DSC Webinar on July 30th, 2019 You’ve spent weeks foraging for, shaping, and analyzing the perfect data set and you’ve crafted your dashboards with care. What could possibly go wrong? Have you made unconscious design decisions that will turn off your audience before you even get started? What message is your audience hearing that you didn’t plan to send? This latest Data Science Central webinar will introduce you to three areas where your design choices might be telling your audience a different story than intended, how be more aware of potential pitfalls, and make adjustments accordingly. Speaker: Jenny Richards, Senior Program Manager -- Tableau Hosted by: Stephanie Glen, Editorial Director -- Data Science Central   Title:  What Is Inclusive Data Storytelling, And Why Does it Matter? Date:  Tuesday, July 30th, 2019 Time:  9:00 AM - 10:00 AM PDT   Space is limited so please register early: Reserve your Webinar seat now

Where We See Shapes, AI Sees Textures by Jordana Cepelewicz via @QuantaMagazine

Deep learning vision algorithms often fail at classifying images because they take cues from textures, not shapes. This is a really interesting look at how machine vision actually processes the world.

This is absolutely fascinating and a great approach as to how relatively minor changes might make all the difference to your algorithms and outcomes.

WEBINAR: Maximizing Data Science Applications And Model Development - 24th July 2019

Data Science Central Webinar Series Event

Maximizing Data Science Applications And Model Development Join us for the latest DSC Webinar on July 24th, 2019 In this latest Data Science Central webinar, you will learn the value of an optimized workstation for Data Scientists. We will demonstrate NVIDIA CUDA-X AI software stack and how it has enhanced data science workflows. Featured Speakers: Tim Lawrence, Founder and VP of Engineering and Operations -- BOXX Technologies Allen Bourgoyne, Senior Product Marketing Manager, Quadro -- NVIDIA Hosted by: Stephanie Glen, Editorial Director -- Data Science Central   Title:  Maximizing Data Science Applications And Model Development Date:  Wednesday, July 24th, 2019 Time:  9 AM - 10 AM PDT   Space is limited so please register early: Reserve your Webinar seat now

The Pentagon has a laser that can identify people from a distance—by their heartbeat by David Hambling via @techrieview

It detects a person’s unique cardiac signature with an infrared laser. While it works at 200 meters (219 yards), longer distances could be possible with a better laser. “I don’t want to say you could do it from space,” says Steward Remaly, of the Pentagon’s Combating Terrorism Technical Support Office, “but longer ranges should be possible.”

Wow - this is amazing. 

What 70% of Data Science Learners Do Wrong by Dan Becker via @Medium

Corporate data science is still a new field. Many academics haven’t worked on real problems for real businesses yet. So they teach textbook algorithms in a way that’s separated from data and business context. This can be intellectually fun. But, students are mistaken if they assume these courses prepare them well to work as data scientists.
Short, solid read.

Worth reading and understanding what he means - my advice would be to log onto Kaggle and do some competitions as some are for real companies and they provide realistic data. If you can do one of those and get a good result you have a better chance in the real world of work. Also, it is useful to have the code for them in Github and available for any potential interviews.

The quality of its data can make or break an organisation by Bob Violino via @infomgmt

High-quality data can improve decision making, customer service, business processes and competitiveness. Poor quality data can potentially lead to financial ruin.0

Time and time again I've explained the need for quality data. If the data is not in a structure you understand or clean the results will not be reliable and will be complete rubbish.

SLIDESHOW: 8 steps to recover quickly (and well) from a data breach by/via @infomgmt

Cyber attacks are happening with increasing regularity, and health organizations need a recovery plan.

This is very useful to have and could be helpful in drawing up for organisations own plan with more detailed information.

AI and machine learning will require retraining your entire organisation by Ben Lorica via @OReillyMedia

“Implementing and incorporating AI and machine learning technologies will require retraining across an organization, not just technical teams,” says Ben Lorica. Here’s why.

Some interesting stats in this article by Ben which might be very useful when justifying your own organisation's training provision. I think this is going to be required in organisations more and more int he future and that it will not just be an IT sill.

6 Data Analytics and Business Intelligence Trends to Fire Up Your Business by @eSparkBiz via @Datafloq

Nowadays, when business intelligence and data analytics are in huge demand, this blog will help you to be aware of its latest trends with ease.

Time and time again we come back to the need for quality data and proper data management. We really need to learn the lessons and actually start sorting it out. Yes it will cost money, yes it will use resources, but it is an INVESTMENT into your company just as buying a new software package or piece of machinery is.

Microsoft will fix your sexist PowerPoint presentations with AI by Mark Wilson via @FastCompany

Microsoft PowerPoint’s new AI tool, Presenter Coach, lets you practice your presentation in front of an AI. Along the way, it will offer helpful suggestions like: “avoid reading the slides,” “don’t use so many filler words,” or pointing out when you’ve unnecessarily gendered your pronouns—“that could be culturally sensitive in some cases.”

This is a great use of AI and a great help to anyone working on a presentation (and if we are honest we all get nervous before giving one).

Machine learning and data science applications in industry by Derek Snow via FirmAI

This curated list of applied machine learning and data science notebooks and libraries across different industries is pretty interesting.

This is incredibly useful and you can give feedback and suggest new ones to him too.