Enjoy a tipple or five? You might need this AI system to tell you when it's time for a new liver Machine-learning algorithms might be able to work out if your brain is turning to mush from end-stage liver disease just by reading your emails, according to preliminary experiments. ESLD, also known as chronic liver failure, slowly progresses over months and years. It’s often caused by cirrhosis, where scar tissue in damaged …
Ignoring that there are non-alcohol related sources of cirrhosis...
If you have 2 heavy drinkers with similar education levels, then you are comparing like with like. If one shows more impairment than the other there is a higher chance they need medical intervention.
Of course, if this does work out then we with be incentivising people to write long, ranty, meandering emails to their doctors.
I admit this experiment is totally idiotic but it can always be used as a negative example.
Machine Learning technology is when you search without knowing what you're looking for.
As Yogi Berra used to say, You’ve got to be very careful if you don’t know where you are going, because you might not get there.
"...NLP may provide opportunities to detect cognitive impairment in ESLD..."
Machine Learning is not NLP! These are two different things: NLP just helps structure information, translates it into a computer-friendly and readable format, that's all.
AI uses structured information (texts) to search for another information, formulating (expanding to several hundred and thousands of patterns) search queries. In turn, feedback-based Machine Learning refines these searches (based on attracting new texts).
Finally, the AI acts or does not in accordance with the information found.
They, however, simply compare emails: "...where they compared the emails..."
The idea of Artificial Intelligence:
1. A personal profile is created, based on structured tests; this is NLP part, where you need it as such.
2. A search query, which may consist of 1-2 words, is expanded to complete-and-meaningful patterns; the technology of which is outlined in my US Patent 6.199.067 (PA Advisors v Google).
3. This search pattern is then filtered through the profile, it's enriched with hundreds and thousands of explanatory patterns.
4. Data is searched.
5. The information found is used by AI. (For example, by Waymo and Uber driverless cars.)
Machine Learning technology helps to refine the queries by the addition of texts.
That's it, nothing more or less.
Other factors can be ignored? Like time of day, work load pressures and/or caffeine/blood ratio imbalances (without enough caffeine to revive the dead I don't do the brain meats thinkie time thing good). On that note, time for more tea.
Tramp because thats what I feel like before the first (coffee) hit of the day.
Now I'm terrified when composing emails knowing that an algorithm created by a bunch of programmers with too much time on their hands will diagnose me with I don't know what disease.
Seriously, I would suggest those guy to turn their attention to plants or rocks and allow us to carry on with our happy lives.
Depending on your target audience, the "After" version might be more suitable, since it gets to the point and avoids a load of flannel that isn't actually up for debate anyway. So, people who have honed their communication skills over the years and no longer sound like some Apprentice Twat who is trying too hard and lacks self-awareness ... these people have liver disease?
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
