http://www.nlp-class.org/

“The course covers a broad range of topics in natural language processing, including word and sentence tokenization, text classification and sentiment analysis, spelling correction, information extraction, parsing, meaning extraction, and question answering, We will also introduce the underlying theory from probability, statistics, and machine learning that are crucial for the field, and cover fundamental algorithms like n-gram language modeling, naive bayes and maxent classifiers, sequence models like Hidden Markov Models, probabilistic dependency and constituent parsing, and vector-space models of meaning.”

Excellent!  I’ve signed up.    I love the online courses by Stanford.

-Adeena

Great. Thanks for posting this. I’ve fallen behind, but the machine learning course has been really interesting.

I’m going through the lectures from this stanford nlp class, taught by Manning…

Favorite quotation so far, from Lecture 1:

“Lately using probabilistic models has become quite trendy and widely used in a lot of areas of artificial intelligence.”

:)

Also it’s interesting that Good-Turing smoothing is still used in nlp (see lectures 2 and 3…), showing once again how Turing anticipated a lot of AI work.

The lecture slides for the “Introduction” from https://courseware.stanford.edu/pg/courses/lectures/214428 lists, in the second column of page 3, two “garden path” sentences which for kicks I submitted to my bot’s linkagent.

> link: svo: The man who hunts ducks out on weekends.
S=The man; V=ducks out on; O=weekends

> link: svo: The cotton shirts are made from grows here.
S=The cotton; V=grows; O=NA

Basically link grammar interprets the sentences correctly (although for my purposes I include the adverb and preposition in the “V”, and the “O” is really the object of a prepositional phrase.)

I also have a MontyLingua agent, which interprets both sentences naively (i.e., follows the “garden path” the sentence leads you up before reaching a point where you should backtrack and re-interpret):

> monty: what is the verb in “The man who hunts ducks out on weekends.”
hunts

> monty: what is the subject of “The cotton shirts are made from grows here.”
The cotton shirts

—-

I’d like to add a statistical nlp parser; i’ve downloaded the stanford lexparser a few times in the past but since it only provided input/output through text files, and since it used a lot of resources, I haven’t yet made it into an agent.

The online version of the LexParser is at http://nlp.stanford.edu:8080/parser/index.jsp

Checking the above two garden path sentences, it gets the first, but interprets “cotton shirts” as the subject of the second (as MontyLingua did :)

(ROOT
  (S
  (NP (DT The) (NN cotton) (NNS shirts))
  (VP (VBP are)
    (VP (VBN made)
      (PP (IN from)
      (NP (NNP grows)))
      (ADVP (RB here))))
  (. .)))

Excellent! I have been catching up on the courses.