bas-rustenburg
commented
8 years ago This website you mentioned uses an xml format for recipes. https://www.directaccessrecipes.com/index.jsp?groupId=1&catId=34
Open sonyahanson opened 8 years ago
bas-rustenburg
commented
8 years ago This website you mentioned uses an xml format for recipes. https://www.directaccessrecipes.com/index.jsp?groupId=1&catId=34
So this is a bit tricky, didn't check reddit yet, but here are some ideas:
1) See if we can become involved with IBM Watson and use whatever database/resources they have handy: link. 2) Screen-scrape some html websites, such as this one: link. 3) Download a fancy recipe manager app or two, and see if they give us a downloadable text version of their recipe databases: link and link. 4) Similar to 2 and 3, but these are fancier websites that must have databases, maybe they have a way of downloading things or we can also try to screen-scrape these: link to best salad recipes on big oven, link to best salad recipes on BBC food, link to salad recipes in the new york times, and link to salad recipes on allrecipes.com. 5) There are a handful of websites that people seem to have worked hard on and then have fizzled, that have tried to programmatically access, use, and/or collect recipes: gourmet recipe manager, fridgetofood1, fridgetofood2 (which happens to have been an old project of a good friend of mine since I was 2 y/o), direct access recipes, and I bet there are others. Possible they have githubs with their data or something, or we could collaborate, by sending emails, etc.
And then we can make an awesome histogram of frequency of ingredients found!