Python has NLTK , Java has Stanford's NLP tools , probably PHP also has a bunch of em, is there any such collection of tools for C++?
I have found some but either they were not open source (freeware/shareware) or discontinued.
If there are none , could anyone direct me to the resources which can explain how named entity recognition works? I think I can try to implement one if there is none (my training/testing set is limited).
I would highly likely would find a POS tagger , but it would be helpful if I can know how this thing works.
And lastly I would request Keshav Dhandhania , if he can share some his projects on Machine Learning especially Natural Language Processing with us.
I already told you why I am using C++ , as I the current project requires me to either use PHP or CGI , I dont like PHP , I hate the $'s in front of variable names , the fastest option would be C++ (for me at least).
Natural language processing, or NLP, is a field concerned with enabling machines to understand human language. This small guide by the folks at YCombinator is well written from a perspective of offering beginners some guidance on getting started with NLP.
I am currently going through the Stanford NLP course , particularly the N-grams part , fortunately thanks to Python's NLTK I can generate the frequency distribution and than probabilities , but doing this every time a query is fired (search suggestions) wont be economical. So I researched a bit on how to store them , but havent found any nice resource for it. Do anyone have some good resources on this topic? Also if some have some other resources for learning NLP , please share.
[Paper summary] Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
The comments on Open Review provide a really good summary of the paper (copy-pasting the highlights and summary below)
Comments by Azalia Mirhoseini (author)
The main idea of our paper can be summarized as this: Massively increasing the capacity of deep networks by employing efficient, general-purpose conditional computation. This idea seems hugely promising and hugely obvious. At first glance, it is utterly shocking that no one had successfully implemented it prior to us. In practice, however, there are major challenges in achieving high performance and high quality. We enumerate these challenges in the introduction section of our new draft. Our paper discusses how other authors have attacked these challenges, as well as our particular solutions.
While some of our particular solutions (e.g., noisy-...