Show more

@Surasanji @SecondJon
As Dave said "The vaccines work best when everyone takes them and there is a herd immunity."
That is the best practice.

@Surasanji @SecondJon what I know about it is that, at least for some diseases, there is a low % of people for which the vaccine doesn't work. Causes might be different, I know the case of Hepatitis A and B, in one case after the first attempt the patient made a test and the result was that no antibody had been produced. After the second vaccination attempt the vaccine worked only for Hepatitis A, but not B. Only one case between more than 100 people in this clinic.
I suppose that for other diseases it can be similar.

@Surasanji The anti-waxxers are a real epidemic. What you see in Israel now I know very well from European countries like Italy or Germany. Just here the repercussion in the media is higher, probably because of the size of the country.

@ckling Thanks a lot Christoph, what is ANEW?

Machine Learning Yearning A Revolutionary Book From Andrew Ng

Almost impossible to find available annotated corpora for aspect based sentiment analysis... If somebody can recommend someting... I will be thankful.

I have the feeling that after the first hype is has lost momentum as concurrent to in Data Science.
It is difficult, regardless how good is the language, to compete with a established programing language like Python with a broad community and lots of good open source libraries.
Anyway, an article with the opposite opinion: "Julia vs. Python: Julia language rises for data science"

More consumers turn away from Apple in
An increasing number of Korean consumers are turning their backs on Apple based on the notion that iPhones offer a limited scope of functions in addition to the inconvenience of the iOS operating system, while continues to maintain its high price policy.

Archives and modern mythology: The use of archival records in comic books

In some áreas of NLP I'm not really impressed. I have see good results if sequencial models in Sentiment Detection, convolutional in Named Entity Extraction, or some DL approaches in text classification... but with not a lit of effort I was able to improve them just with Logistic Regression (for sentiment), CRF for entities or SVM for classification.
In other NLP areas like dialogue models or machine translation DL is offering better results.

K. J. R. boosted

@alan I also feel the same way. My issue is the interpretability of DL models. Especially, when developed by people who lack knowledge in some basic ML concepts. This has frustrated me enough that I started a movement around what I call Shallow Learning. Most of what people think they need a DL model for, could easily be a good fit for logistic regression or some simple decision tree logic.

@arjen Hey Arjen, yes, I know. Just I prefer to use for other reasons. cheers!!

Will Bots Really Replace the Human Customer Support Role?

Linear Regression Vs Logistic Regression - a basic and nice article

5 Important Artificial Intelligence Predictions (For 2019) Everyone Should Read
Artificial Intelligence – specifically machine learning and deep learning – was everywhere in 2018 and don’t expect the hype to die down over the next 12 months.

How to Master the Art of Freelancing | The Startup Magazine

Freelancing is becoming increasingly popular around the world.

K. J. R. boosted

מעצר של בכירה בחברת וואווי מאיים לחדש את המלחמה הכלכלית בין סין וארה”ב

K. J. R. boosted
K. J. R. boosted

I am really disappointed with the results of the sentiment analysis of the Google Cloud NLP.
It is really expensive, and the results are not really better than what we get with tools like TextBlob.

Show more
QOTO Mastodon

Welcome to the QOTO instance

QOTO: Question Others, Teach Others

"I know you won't believe me, but the highest form of Human Excellence is to question oneself and others." - Socrates

A Mastodon instance for scholars in Science, Technology, Engineering, and Mathematics (STEM) and all yearning to learn.

A free speech space. No censorship here.
We do not silence or block any instances.

Unique Features
  • Inline math Latex support - Use \ ( and \ ) for inline LaTeX, and \ [ and \ ] for display mode.
  • 65,535 character limit for toots (usually 500)
  • 65,535 character limit for profile bio (usually 160)
  • Full text searches - usually you can only search hashtags and usernames
  • Halcyon / Twitter clone front-end - Want the interface to look exactly like the Twitter interface? After you signed up just go here to login:
  • PeerTube - We host our own PeerTube server with open registration for up to 100MB, but all of our active users get unlimited quotas. Sign up here:
  • Professionally hosted with nightly backups
  • light modern theme with full width columns (not fixed)
  • Several extra themes - including mastodon default and mastodon default with full width columns (not fixed)

Message to New Members

Welcome to the Qoto Mastodon instance, its like twitter but better!

Please don't hesitate to Direct Message me or tag me in Toots (what we call tweets here) if you have any questions or even if you just want to chat.

If you are new to Mastodon you may want to check out this official getting started guide:

We are a free speech, no censorship zone. Feel free to talk about whatever you want, say whatever you want, as long as it is legal you wont ever get banned from the server. With that said we do have a few rules: No spam, and no using multiple accounts to circumvent personal bans.

This server isnt a typical mastodon server either. I have personally modified the code on this server to give it several unique features you wont find on other servers.

If you are new to Mastodon here is some useful info.

First off, if your coming from Twitter and you'd prefer an interface that looks exactly like twitter, then just go here and login with your Qoto credential: No need to create an account there, your account on this server will work just fine. Just enter your username in the following format @[username] so for me it would be @freemo But personally I think our default skin here is way better.

Basically Mastodon is a decentralized twitter. That means anyone can run a server (like Qoto) that people can sign up to and use just like twitter. The unique part is how the servers talk to each other.

If you want to tag someone on the same server as you then you can just do @user however if you want to specify someone on a different server you can do @user@server. If someone follows you or mentions you on any server on the internet you'll get notified right here. So my full tag is

So now a bit about the timelines, there are three (unlike twitter there is just one): Home, Local, Federated.

The Home timeline is just like what you'd expect on twitter, that is, everyone you are following.

The local timeline is any post made by anyone on your local server.

The federated timeline is what is cool. It combines the timeline of everyone on the server into one. This way you can see the posts of everyone that is followed by anyone on the server (assuming the post is public). So this timeline is a close representation of everyone on the internet who has Mastadon (though not really). Moreover we have a bot here that goes around and federates with everyone it can find, so our Federated Feed is rather robust covering thousands of instances.

Only other thing that is unique is listed vs unlisted posts. You have normal privacy settings like on twitter but you also can set a post to be listed or unlisted. If it is listed it acts just as i described. If it is unlisted then it acts just as a normal tweet would but it doesnt show up on local or federated timelines anywhere.

Also this goes without saying, but don't be racist, sexist, hateful, and don't harass people. Its a pretty shitty thing to do, don't be shitty!

The following is the list of QOTO administrators/moderators.
Note: The account listed as the administrator is NOT an actively monitored account. It was selected so as not to show favoritism among our moderator group, we are a democracy and even our administrators and moderators are held to the same high standards of decency as the rest of the community.

***We will never advertise on QOTO or sell your information to third-parties***