/sci/ - Science & Math » Thread #12301711

102KiB, 621x960, 120085063_2792006767791174_6592294012041153863_n.jpg

View Same Google iqdb SauceNAO

What is the Difference of Data Science and Computer Science?

Anonymous Tue 03 Nov 10:44:32 2020 No.12301711 View Reply Original Report

Quoted By: >>12301712 >>12302846 >>12303048 >>12303548 >>12303807

I believe in the racial discrepancy of white and black but how about computer science and Data Science? those fields have the highest level in society, especially in salary.

Anonymous

Anonymous Tue 03 Nov 2020 10:46:45 No.12301712 Report

Quoted By: >>12302947 >>12302965

>>12301711
>52-65% unhappy in their current job
Whew lad. Would be interesting to see how that compares to average but last time I checked job dissatisfaction wasn't even that high in the natural sciences.

Anonymous

Anonymous Tue 03 Nov 2020 18:56:51 No.12302846 Report

Quoted By:

>>12301711
Computer science is a broad field of various topics all related around computation. Some of these are more applied and closer to engineering, and some of them are heavily theoretical and close to mathematics. You could say computer science proper, ie not the undergrad software dev route, is the result of mixing together researchers from math, engineering, and physics.

Data science is an broad application area of CS and statistics. It borrows heavily from the massive data storage and retrieval, ML techniques, and numerical algorithms from the CS side, and then statistical learning theory, inference, and experimental design from statistics. Like CS, the actually interesting parts of data science have almost nothing to do with the entry level industry applications / undergrad degree.

Anonymous

Anonymous Tue 03 Nov 2020 19:26:09 No.12302947 Report

Quoted By:

>>12301712
wonder how many of them are career changers.
just gotta get into my field somehow before all these fuckers oversaturate hpc

>job dissatisfaction wasn't even that high in the natural sciences.
that probably discounts all the people in hard sciences that couldn't get a job in their field

Anonymous

Anonymous Tue 03 Nov 2020 19:30:32 No.12302964 Report

Quoted By:

Computer science includes various subdisplicines like artificial intelligence, data structures, human-computer interaction, computer graphics, programming language theory, automata theory, etc.

Data science doesn't mean anything. Literally anyone who does statistics on a computer calls themselves a "data scientist"

Anonymous

Anonymous Tue 03 Nov 2020 19:30:39 No.12302965 Report

Quoted By: >>12303028

>>12301712
>>52-65% unhappy in their current job
Yeah, can someone explain where this comes from? I have this naive notion that you as a data scientist you would occasionally do some hard but creative and useful work.

Anonymous

Anonymous Tue 03 Nov 2020 19:47:57 No.12303028 Report

Quoted By:

>>12302965
>I have this naive notion that you as a data scientist you would occasionally do some hard but creative and useful work
i'm a SWE and our perception of data science is that they just clean data and run it through a canned algorithm, or even just are programmers with a different title
also they probably feel annoyed that they don't get paid much compared to undergrad cs majors

Anonymous

Anonymous Tue 03 Nov 2020 19:53:52 No.12303048 Report

Quoted By: >>12303521

>>12301711
>Data science methods
>Top 1: logistic regression
AHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAH
This explains why the last time I spoke with a data """scientist""" he said I was making pseudo-intellectual memewords when I mentioned that the goal of supervised learning algorithms is just to statistically infer a mapping between two vector spaces.

Anonymous

Anonymous Tue 03 Nov 2020 21:58:50 No.12303521 Report

Quoted By: >>12303799

>>12303048
So is AI anything besides machine vision / value comparison for some sort of weird cost/benefit analysis at this point? I have a friend who does it for a living and he told me if I did LEETcode & grinded courses for a year I could get an entry level job doing some machine learning work. And he makes six figures already.

Anonymous

Anonymous Tue 03 Nov 2020 22:06:05 No.12303548 Report

Quoted By:

>>12301711
>logistic regression
How do you prevent this please

Anonymous

Anonymous Tue 03 Nov 2020 22:08:31 No.12303554 Report

Quoted By: >>12303673

CompSci is a lot of things. DataSci is just spicy statistics.

Anonymous

Anonymous Tue 03 Nov 2020 22:46:24 No.12303673 Report

Quoted By:

>>12303554
nah, data science proper has a lot of interesting problems for CS, like massive data retrieval on conventional systems. that being said, I think CS is more inherently interesting because its domain of problems is so much more general

Anonymous

Anonymous Tue 03 Nov 2020 23:36:01 No.12303799 Report

Quoted By: >>12303974

>>12303521
Most (((AI))) workers don't have any clue about the mathematics behind the stuff, they just use python libraries.

Anonymous

Anonymous Tue 03 Nov 2020 23:38:04 No.12303807 Report

Quoted By:

>>12301711
Data science can be thought of as data driven programming.
Some people call it software 2.0

Anonymous

Anonymous Wed 04 Nov 2020 01:14:38 No.12303974 Report

Quoted By:

>>12303799
Can you elaborate? I don't even understand how that is possible. They are basically just running the equivalent of graphing calc stat tests on their calculator? Why not just use MathLab? How hard would it be to become a data scientist then?

Anonymous

Anonymous Wed 04 Nov 2020 01:43:06 No.12303995 Report

Quoted By:

Data science is a pretty wide umbrella, but it's separate from computer science.

Computer science is the theory of computing. CS is also different from software engineering, which is the practice of building software systems.

DS in general covers all forms of data analysis and the computational tools used to do this analysis. There's also a strong engineering side of things. For analysis, this can be stuff like making dashboards, doing inferential statistics, building predictive models, machine learning, etc. The engineering side is building the infra to run data collection and model serving at scale in an automated way.

One of the hard things about DS is how many different things are held under the name. The guy who makes bar chart dashboards has the same title as the guy building the AWS backend to a recommendation system, even though those are completely different things with different skillsets.

Capcode	All Only User Posts Only Moderator Posts Only Admin Posts Only Developer Posts
Show Posts	All Only With Images Only Without Images
Deleted Posts	All Only Deleted Posts Only Non-Deleted Posts
Ghost Posts	All Only Ghost Posts Only Non-Ghost Posts
Post Type	All Only Sticky Threads Only Opening Posts Only Reply Posts
Results	All Grouped By Threads
Order	Latest Posts First Oldest Posts First

Your latest searches