Monday, February 20, 2017

Coursera courses - worth of doing?

Just finished Coursera course in Machine Learning.

It was interesting, I learn some interesting stuff.
Have something productive to do by last 6 weeks.

Just wonder, if it makes any difference ;-)

My guess is that this doesn't change this how attractive I'm on job market... I'm Senior Software Engineer so "they" are looking for me with or without certificate from Coursera.

But now I know at least some tools which may be used to play or even use machine learning.
I used it already in my ClickBaitaDetector. But still feel that this may be done much better.
I try to use it also to longer tests, but for now don't have idea how to work with text not looking at once on all words in this text, but on text as set of sentences.

"This is a test" and "This is a test?" looks the same if we are looking only at words.
And this is only tip of iceberg...

So, taking this course caused that I started to wonder how to do some things. So it seems that this was worth of my effort ;-)

posted from Bloggeroid




Similar postsbeta
How to detect clickbaits?
It's alive... still ;-)
Throw out 1 thing each day
Magic Toaster ;-)
OOo2GD 2.1.1 is ready :-)

Sunday, February 12, 2017

Happy Darwin's Day!

Today we have 206 years from day when Charles Darwin was born.
It is good cause to celebrate ;-)

And it is good time to celebrate.

Theory of Evolution was first (at least one of first) truly scientific theory.
It was from the beginning falsifiable, it predicted a lot of stuff which was found latter.

And it changed whole outlook on science, world and us.

Now when current administration in US undermines scientific data, when nationalists and racists raise theirs heads it may be good moment to remember about Darwin.

So, lets celebrate ;-)

We may celebrate with song:


And with visit to Natural History Museum (sadly this one was almost 2 years ago):




Similar postsbeta
How we may use Google Earth Plugin? :-)
2016...
4 steps of evolution of programmer ;-)
People doesn't read...
Coding time means often waiting/reading time ;-)

Monday, February 6, 2017

Roomba friendly furniture...

Tomorrow my Roomba will clean my flat.
He (his name is George) is doing it 3 times a week.

At night before his work day I must prepare flat for him.
Mainly ensure that boxes from old phones are placed in the gap under refrigerator:

And that armchair from Ikea is "hanging":

I need also move air cleaner to other place.

My computer chair and whiteboard are making problems too...

And because of this I'm calling for action ;-) for more Roomba friendly furniture!

Let's make first world even more comfortable! ;-)

posted from Bloggeroid




Similar postsbeta
Throw out 1 thing each day
OOo2GD 2.1.0 - filtering, sorting and more *nix friendly config file names
Whiteboard
My road to automation ;-)
It seems that meditation helps ;-)

Sunday, February 5, 2017

How to detect clickbaits?

From some time I'm working on Chrome extension which will help me identifying clickbait links.
I was reading article The Pot-Belly of Ignorance and realized that clickbaits are very addictive, but you don't get anything from those. So I decided to found a way to get them rid of my reading ;-)

My first idea was to use naive Bayes classifier to detect clickbaits.
It works, but isn't perfect.

So my second try was with logistic regression.

I'm currently talking on-line classes from Machine Learning, and what would be better training in this than using some real world example? ;-)

This try was better ;-)

But lets back to the beginning.

First problem was to find some data to train algos...
Happily I found it on GitHub :-)

Idea was that to detect clickbait it should be possible to use only title of link, without looking on url.
It is rather more probable to see "you" in clickbait link than in proper article, and to see "washington" in proper article is more probable than to see it in clickbait.

Next assumption was that it doesn't matter if words are in capital or lower case, YOU and you should be treated the same.

With those assumption I started to play ;-)

First I used naive Bayes classifier.
Here I done whole training by myself... with little help with Python code from "Machine Learning in Action" (I needed to translate if from Python with NumPy to JavaScript for node.js).

It worked, but even when numbers from test set looked good, it was sometimes too eager to classify something as clickbait.
I suppose that this was caused by this that model assumed at the beginning that for new title it is more probable to be clickbait than proper article.

So I moved to logistic regression.

Here I decided to use GNU Octave to do dirty work of calculating everything ;-)
JavaScript only prepares matrix with vectors build from articles titles, next Octeve is using gradient to find values in vector which will be used latter to classify new data.

It seems to work better. On training set it have 100% accuracy ;-) on test set it is about 97%.

If you are interested in looking on some code, you may find it on my GitHub ;-)
You may find there extensions, and some code for training.


Similar postsbeta
Which language is fastest? ;-)
My road to automation ;-)
Recursion is evil ;-)
It's alive... still ;-)
OOo2GD - surveys results

Friday, February 3, 2017

It seems that meditation helps ;-)

By last week I slept too little.
Mainly because it is almost impossible for me to go sleep early...
But other things expect that I will wake up earlier then I would prefer (for example I was conducting phone interview on 9:00 one day, so I needed to be conscious at this time ;-)).
Because of this today I was from start of the day really angry.
At some time I decided to make short break from work.
Sit on the couch and started to meditate.
It was short 4 minutes meditation, but after that I'm calm. And was even able to start to look on one of problems on my plate from different perspective and probably will be able to go with it.
So it seems that meditation works.. at least for me, and at lest sometimes ;-)
posted from Bloggeroid



Similar postsbeta
Throw out 1 thing each day
Correlation between traffic and time spent at work ;-)
My road to automation ;-)
Vacation in progress ;-)
Really first snow ;-)

Sunday, January 29, 2017

Coding time means often waiting/reading time ;-)

Several times a week I should spend one hour of my own time (private time) on coding my own things.
Google Calendar ensures that I will not forget about it ;-)

And strange thing, sometimes I'm doing rather big changes in code and don't write a line...
For example moving my own Android apps to new Jack compiler which will let me to use Java 8 features means that except small changes in gradle.build file latter I only ask IntelliJ to analyze project and latter do a lot of initial refactoring using mouse... for example changing anonymous classes to lambdas, and next lambdas to method references.

Other time I may spend most of this "coding" time reading docs which will let me do this what was intended...

This is cause why I like JavaScript so much.
When I'm creating something in JavaScript I usually don't use any tools, so in case of working on JavaScript projects almost whole hour goes to coding ;-)
And debugging ;-)

In general I found out that first tools help and make us more productive, but latter we spend more time waiting for tools, or working with tools than on working with code ;-)
Still I'm not sure if it is OK, or not.


Similar postsbeta
Which language is fastest? ;-)
Portable code smells list ;-)
How to detect clickbaits?
4 steps of evolution of programmer ;-)
My road to automation ;-)

Monday, January 23, 2017

Delete your Facebook account!

OK, first go to https://applymagicsauce.com/.

Can you see that somebody who have access to your likes on FB is able to say so much about you?

It isn't small thing.
Thanks to this in UK Brexit won, in US Trump won.

Knowing what kind of personality you have, which demographics you are, somebody may decide that it will be better if you will not vote, so they will show you some bad things about candidate you wanted to vote for. Or they may decide that if they hit in your "sweet spot" you will vote for theirs candidate.

This is manipulation on personal level.
They may touch your most hidden fears and use them against you.

So, delete your Facebook account.

Or at least, stop using Like and other reactions. At least your personality model may be little bit out of sync. And they will not hit you so much....


Similar postsbeta
OOo2GD help site
2016...
Google Buzz - let's mark some comments as not spam ;-) or how to unhide "hidden" comments ;-)
I want my Android back!!!!
How to create valid file name?