Geek asks and answers

Automatic synonym detection methods

I am currently working on a neural network approach for classifying short documents, and since the bodies I work with usually take about ten words, standard methods for classifying statistical documents are of limited use. In this regard, I am trying to implement some form of automatic synonym recognition for matches provided in the training. My question more specifically is to resolve the situation as follows:

Let's say I have the “Food Attraction” classifications and one of the “Attractive Areas” and the data set as follows:

"Eating Apples"(Food);"Eating Marbles"(Spheres); "Eating Oranges"(Food, Spheres);
"Throwing Baseballs(Spheres)";"Throwing Apples(Food)";"Throwing Balls(Spheres)";
"Spinning Apples"(Food);"Spinning Baseballs";

I am looking for an incremental method that will move in the direction of the following relationships:

Eating --> Food
Apples --> Food
Marbles --> Spheres
Oranges --> Food, Spheres
Throwing --> Spheres
Baseballs --> Spheres
Balls --> Spheres
Spinning --> Neutral
Involving --> Neutral

, , , . , , , "", , , "" "". - , , ?

+5
4

, , .

, .

:

, , . ( , ..)

( ).

:

,

  • ,
  • ,

, , Apple .

.

, Apple, .

( ) . , , "" "" - .

, , , .

, , .

( , , .)

, .

Subj ate NounPhrase
NounPhrase are/is Moldy

The workers harvested the Apples. 
   subj       verb     Apples 

That might imply harvested is an important verb for distinguishing foods.

Then look for other occurrences of subj harvested nounPhrase

, , .

, , , , .

.

, .

Edit:

AutoSlog-TS, - .

+11

LDA, . LDA, ! Stanford , . , NN, . , LDA , ""

+1

(http://ronan.collobert.com/senna/) . , .

, , , , , .

, , , , 20 , 50-100 .

, , - . Word ( - 100k) 50. , .

WordNet (http://wordnet.princeton.edu/) , .

+1

, . , ?

, http://wordnet.princeton.edu/, . ( languges.) , , ( , ), hyponyms ..

/ . "" "" ""; . http://en.wikipedia.org/wiki/Hyponymy , "" "".

, . , , , , . -, , ? , . "", "", "", "", "" ""? (, .) , , .

0

More articles:


All Articles