## Where communities thrive

• Join over 1.5M+ people
• Join over 100K+ communities
• Free without limits
##### Activity
Shouvik Roy
@royshouvik
If any of you are struggling with the math prereq (Linear Algebra, Probability) for the data science path, I highly recommend Khan Academy
Mahdi Dibaiee
@mdibaiee

@royshouvik I'm on Calculus 1B: Integration, last chapter, and I'm excited for the next part!

I was just wondering, I haven't seen much integration in use in data science personally, differentiation, that's a key, but integration? Hardly seen it anywhere.
How does it apply to data science problems? (I understand that it's very good to know calculus inside out, and it really helped me with my math skills anyways, just asking out of curiosity!)

@royshouvik Also another question: I'm a programmer for 4 years now and I've worked with numpy, scikit and tensorflow a little before, should I take the python courses?
Thank you Shouvik! :+1:
Shouvik Roy
@royshouvik
You are right @mdibaiee Integration isn't needed as is in Data Science, it's kind of a basic math prereq (along with other Calculus topics) which would help in any technical field. It might help in visualizing the area under a curve (size of a population, give a bounding curve), what do you think?
If you are comfortable with numpy and scikit, I guess you can skip learning Python, its totally up to you and you have to take that call based upon how confident you feel, time available etc.. that's true for any topic not just Python
We should understand that this is just a guiding path, for everyone who is interested in learning about data science and machine learning. Some of us might have proficiency in one topic or the other.. but the guide shouldn't make any assumption about learners background (except maybe, willingness to learn) , so its possible that there are too many topics / courses in the path
Mahdi Dibaiee
@mdibaiee
@royshouvik I think understanding what Integration is and how it works does really help with mathematical thinking overall, and might actually be helpful in some fields, it has many applications in different fields, if not directly used in Data Science, I think it might be helpful in other stuff. I personally enjoyed understanding it :+1:
Shouvik Roy
@royshouvik
@mdibaiee exactly
Mahdi Dibaiee
@mdibaiee
@royshouvik I see, that's right and that's what makes me love this path so much! I'm free to learn what interests me and that prevents me from being bored! :D
Thank you so much for compiling this path! I don't know how I can express my gratefulness :D
Shouvik Roy
@royshouvik
Just by helping fellow people and submitting PRs :)
Mahdi Dibaiee
@mdibaiee
I'm willing to do that, I'll keep an eye for questions and issues :D
Shouvik Roy
@royshouvik
The path belongs to all of us, so feel free to submit a pull request if you would like to change anything about the path.
Great :+1: we need more people like you :+1: :+1:
:smiley:
Mahdi Dibaiee
@mdibaiee
@royshouvik sure, I will! It's been good so far, I'll see what I can improve along the path. :D :+1:
Srikant Mahapatra
@SassySamurai
@mdibaiee: Integration is used a lot in probability. And probability is one of the cornerstones of data science.
Mahdi Dibaiee
@mdibaiee
@SassySamurai oh, I'll have to see! I haven't reached probability yet, thanks for mentioning it! :D
Vincent Dania
@vincentdania
Anyone from Abuja, Nigeria here?
wsr13990
@wsr13990
i recently try to open edx course for statistics and it says enrollment is closed
is there any good alternative for that?
Srikant Mahapatra
@SassySamurai
@wsr13990, try the new one from University of Edinburgh: https://www.edx.org/course/statistics-unlocking-world-data-edinburghx-statsx
wsr13990
@wsr13990
@SassySamurai Thanks, gonna check that out
wsr13990
@wsr13990
Just as information, it seems that NLP course in coursera is no longer accessible
ionakathryn
@ionakathryn
does anyone know how i can learn R but not download anything and use an online IDE?
or learn data science
i was using datacamp but they want me to subscribe / pay which im not sure if i want to do yet
Xavier Sumba
@cuent
Hello, I'm working on finding relatedness of words. The problem is the following: given 2 groups of words, I need an output which should be some a confidence score that says whether there is or not a relationship between g1 and g2.
Anyone knows a way to that? or Where can I start? I was thinking on word2vect.
Mahdi Dibaiee
@mdibaiee

@cuent: You can definitely use word2vec and cosine distance to find relationships between words, just be aware, relationship does not always mean being synonym, or antonym, or anything like that, the human perspective of "relationship" is not necessarily reflected in word2vec, for example, bad and good are likely to be very similar to each other in word2vec, it's in the sense that they are both used in similar contexts.
I guess you could train a model to find the relationship you have in mind between words.

As a side note: If you are looking for finding similarities between words in terms of characters, you might find Levenshtein distance useful.

Xavier Sumba
@cuent
@mdibaiee Thanks
Mahdi Dibaiee
@mdibaiee
@cuent You're welcome :D
ionakathryn
@ionakathryn
@mdibaiee thank you!
:smile: )
Mahdi Dibaiee
@mdibaiee
:pray: :D
Mahdi Dibaiee
@mdibaiee
Hey, there are a few issues and pull-requests waiting for review, some of them are for a month ago. I would appreciate it if you take a look at them. :D
@royshouvik @ericdouglas
Shouvik Roy
@royshouvik
@mdibaiee Thanks for the heads up...was not able to check on DS repo of late... had some accidents in the family..
But I have merged all the PRs (except one which added paid content to the curriculum).
Mahdi Dibaiee
@mdibaiee
@royshouvik thank you! I also created the pull-request for Natural Language Processing! :+1: :D
ionakathryn
@ionakathryn
Hi I'm learning R via an edx course
and ive sorted my data from columns 13:24
and I need to take a random sample size of 1.....this data from columns 13:24 is called newdata
so why when i run the command ....
sample(newdata,1) all the results are showing from 13:24 when i only want to sample 1??