4 years ago5,000+ Views
I was on LinkedIn today and saw this useful list of big data sources shared by Bernard Marr. All of these have completely free data that you can use for data mining and other forms of data analysis. I was pretty excited to read about it because I'm currently working on a data analysis project and was looking for a data set to work with. • Over 134,101 free government data sets to choose from agriculture to finance or education. • US Census Bureau US population/geographic/education datasets. • European Union Open Data Portal Similar datasets but based in European Union • Data from UK government that includes British National Bibliography • The CIA World Factbook Data of history, population, economy, government, and more • US healthcare data • NHS Health and Social Care Information Centre UK Healthcare data • Amazon Web Services public datasets 1000 Genome Project, NASA database of satellite imagery of Earth, and more • Facebook Graph Basically all the user data Facebook can get away with sharing. Those privacy statements are something to think about. • Gapminder Data from World Health Organizations and the World Bank • Google Trends Stats on search volume for given terms since 2004. Pretty fascinating. • Google Finance • Google Books Ngrams You can search and analyze full texts of millions of books. • National Climatic Data Center Environmental/meterological/climate data sets • DBPedia Wikipedia data • Topsy Social media data • Likebutton Mines facebook public data • New York Times Searchable and indexed archives of news articles dated back to 1851 • Freebase Community compiled database of structured data (over 45 million entries) • Million Song Data Set Metadata on over a million songs and pieces of art Hope you find these data sources useful! (via linkedin) @AreliCanales you might want to save this one!
Time series analysis? of what? Sounds like you are trying to do a regression analysis to find the current state of something due to its past... am i correct?
@Goyo, Yes, I could use regression analysis you are correct :-)
@Goyo time series analysis. I wanted to use a cool dataset but it looks like I'm sticking to earthquakes. Well, not that earthquakes aren't interesting haha
@TechatHeart what kind of data project are you working on?