Ask HN: What are some good resources to get started with Information Retrieval?

https://news.ycombinator.com/item?id=1945819

by ideamonk • 16 years ago

2 4 16 years ago

news.ycombinator.com

I haven't been in touch with hardcore academics since around 2 years. Recently coming across a few IEEE papers on twitter(finding hot trends, ranking tweets, etc), I'm thinking of teaching myself some information retrieval from scratch, but from this standpoint topics like NLP, IR seem quite hard to fathom, or even to begin with in the first place.

Any pointers to make the journey smoother, essential foundations? trusted/good books? fun experiments for self learning? IR specific wikis/discussion groups?

Thanks in advance.

nostrademons2• 16 years ago

Managing Gigabytes is one of the more well-known textbooks...

Really though - if you want to get started with IR, your best bet is probably to get a job at Google. This is one of those fields where industrial practice is light years beyond academic research, and once you get a job at Google in basically anything, it's relatively easy to transfer to Search and learn how the system works.

l0nwlf2• 16 years ago

I am not convinced though. I don't think nooglers are assigned to Google Search team. They won't expose their search-algorithm to a rookie.

nostrademons2• 16 years ago

It's significantly easier to get into Google and then transfer to search quality than it would be to get into a graduate program, get the Ph.D, and then get the necessary data and machines to do top-level work in IR. Just look at the number of tenured professors that accept jobs at Google.

FWIW, I've been working in Google Search Quality since I joined the company nearly two years ago. I don't work in ranking, but I have high-level knowledge of how the whole system works that's better than you'd get out of a textbook.