MIT Natural Language Processing First Lecture: Introduction and Overview (Part I)

Natural language Processing: Background and overview
Natural Language Processing:background and overview
Author: Regina Barzilay (Mit,eecs Department,september 8, 2004)
Translator: I love natural language processing (www.52nlp.cn, January 3, 2009)

The question to be answered in this class (Questions that today's class will answer):
1. What is natural language processing (what is Natural Language processing (NLP))?
2, why natural language processing is more difficult (why NLP are hard).
3, we can build a program that can be learned from the text. (Can We build programs that learn from text)?
4. What does this course contain (what would this course is about)?

One, what are natural language processing (what is Natural Language processing).
1. The computer takes natural language as input or output:
The picture is slightly ...
The input corresponds to the natural language understanding (Nlu:natural Language Understanding);
The output corresponds to natural language generation (nlg:natural Language Generation);
2. Various viewpoints about NLP:
A, computational model of human language processing (computational models of human language processing):
--Within the program according to Human Behavior (Programs that operate internally the way humans do)
B. Computational model of human communication (computational models of human communication):
--Programs interact like humans (Programs that interact likes humans)
C. Efficient processing of text and speech computing systems (computational systems that efficiently process text and speech)
3. Application of NLP (NLP applications):
A, "Baby Fish" MT (machine translation with Babel fish) ....
B, MIT translation System (mit translation systems) ...
C, Text Digest (textbox summarization) ...
D, dialogue System (dialogue systems) ...
E, other applications (other NLP applications):
--Grammar check (Grammar Checking)
--Emotional classification (sentiment classification)
--ets Composition score (ETS essay scoring)

Second, why natural language processing is more difficult (why NLP are hard).
1. Ambiguity (ambiguity)
"At last, a computer this understands you like your mother"
The understanding of the sentence:
A, it understands you just like your mother understands you (it understands your as well as your mother understands);
B, it understands your liking for your mother (it understands (that) your like your mother);
C, it understands you as if you understand your mother (it understands as well as it understands your mother)
D, let's look at Google's translation: Finally, a computer can understand that you like your mother (the translator added, it seems that Google's understanding is more like B).
A to c these three kinds of understanding is good or bad. (1 and 3:does this mean well, or poorly?)
2, different levels of ambiguity (ambiguity at many levels)
A, voice level ambiguity--speech recognition (at the acoustic Level-speech recognition):
--"... a computer that understands you like your mother"
--"... a computer that understands you lie cured mother"
B, syntactic ambiguity (at the syntactic level):
Entries
Different structures result in different interpretations (Different structures leads to Different interpretations)
More examples of syntactic ambiguity (more syntactic ambiguity) ... figure
C, semantic (meaning) levels of ambiguity (at the semantic (meaning) level):
Definitions of "mother":
--a woman who had given birth to a child
--a stringy slimy substance consisting of yeast cells and bacteria; is added to cider or wine to produce vinegar
Here's an example of the ambiguity of the meaning of the word (this was an instance of words sense ambiguity)
More examples of Word ambiguity:
--they put money in the bank
= Buried in mud?
----saw her duck with a telescope
D, Discourse (multi-language) level ambiguity (at the discourse (Multi-clause) levels):
--alice says they ' ve built a computer that understands your like your mother
--but She ...
... doesn ' t know any details
... doesn ' t understand me at all
This is a instance of Anaphora, where she co-referees to some other discourse entity

To be continued: Part Two

Attached: Courses and courseware PDF download mit English page address:
http://people.csail.mit.edu/regina/6881/

Note: This document is published in accordance with the MIT Open Course authoring and sharing specification, reproduced please specify the source "I love Natural Language processing": www.52nlp.cn

from:http://www.52nlp.cn/mit-nlp-first-lesson-introduction-and-overview-first-part/