Looking for suggestions of sources for data science project

Question

I'm currently taking a first step into Data Science, and studying a specific computational method, I'm a math student, but I'm looking for reliable data sources, and if related to biology (e.g public health numbers) or information spread (this one i guess is harder to evaluate, but maybe like spreading of news or sharing of information on internet) it would be very appreciated.

My point here is where to look up for this kinda of data, in a reliable manner, and preferably for free.

Any suggestions are welcome!

---

As suggested:

Data: Mainly data from numbers of case of disease in a region over time, or sharing of fake news (nothing correlated i know but not sure yet which path to pursue)

Context: Trying to correlate data via some kind of regression, so for modeling some specific situations, method is developed but looking to try it on real data, and it would be better if a large amount data is available.

Region: Preferably in Brazil, but at this moment any region should be enough.

License: Very situational whether willing to pay for it or not, if for free it is easier to start a work.

Format: Will problably use it on Mathlab or Phyton, so right now just raw data in txt or any format like this should do it

Authority: Not really open to dubious data

Non Answers: I did look for it on governamental sites from Brazil, it has some data but wanting other sources also (still acepting hints on websites from Brazil, might not have looked for all possible sources yet).

Answer 1

I think this is a good candidate for the UCI Machine Learning Repository, which is often used here as a resource: link to threads.

With the "View All Data Sets" view you can filter and find specific, prepared and cleaned data sets. For learning this is quite beneficial, because there will surely be others developing with these data sets. Each data set is labeled with what kind of analysis is suggested, for example "_regression / numerical / multivariate _".

Additionally, by exploring with the portal and trying various filters, you'll get familiar with the terminology of data science and machine learning.

---

Example usage:

Click on "Regression" on the left menu bar, and you get this view: http://archive.ics.uci.edu/ml/datasets.php?format=&task=reg&att=&area=&numAtt=&numIns=&type=&sort=nameUp&view=table

Then click on "Numerical": http://archive.ics.uci.edu/ml/datasets.php?format=&task=reg&att=num&area=&numAtt=&numIns=&type=&sort=nameUp&view=table

Keep playing with the filters until you find a good one that looks like fun to explore.

For example, here's the data set "Fertility"

100 volunteers provide a semen sample analyzed according to the WHO 2010 criteria. Sperm concentration are related to socio-demographic data, environmental factors, health status, and life habits

which has these technical properties:

---

You can also search the repository with custom search terms, like this search for Brazil.

Answer 2

For Biology, depends the kind of biology. For Ecology, probably you'd better check the Scientific publications from a journal, and check the associated data.

• Dryad, one of the repositories used for data linked to some scientific papers: https://www.datadryad.org/

• If you mean Molecular Biology you can try European Molecular Biology + European Bioinformatics Institute EMBL-EBI: https://www.ebi.ac.uk/

• If you mean Health related Biology you can try World Health Organisation: https://www.who.int/nutgrowthdb/database/en/ or the US Center for Disease Control (CDC) https://www.cdc.gov/surveillancepractice/data.html

• If by biology you mean Agriculture you can try the Food and Agriculture Organisation (FAO) http://www.fao.org/statistics/databases/en/

• If you are looking for Biology as a proxy for Healthcare and Medecine, there are some regulations about personal information which changes depending on the country you are on, and you'd better talk to a real Reseacher in the field you want to study data from. They will tell you where to find data of good quality.

Databases in biology are pretty much domain(-ish) specific. Formats for information are also quiet variated, mostly in ASCII files.

Answer 3

Glad to hear that you are interested in using data for your science project. There are plenty of public resources that you may utilize including GitHub, Kaggle, and Google Open Data.