wordnet2sql is what is needed to populate a relational DBMS with data from WordNet. There are many reasons for doing that. Most importantly, this gives you a way of easily extending or modifying the data structures (for example by statistics). A DBMS allows you to formulate complex queries over the data (for example to extract certain statistics about wordnet), and to optimize the indexing structures to optimally support the queries needed. For certain tasks, a lexical database needs to be modified, which DBMSs allow you to do (even in a distributed fashion).

wordnet2sql consists of a few scripts to create database tables according to wordnet2sql's relational model, and a simple converter written in python that takes input from WordNet's standard datafiles ("dict/data.*") and converts them into an SQL script containing INSERT statements.

wordnet 3.0

WordNet is a lexical database. English nouns, verbs, adjectives and adverbs are organized into synonym sets, each representing one underlying lexical concept. Different relations link the synonym sets and words. wordnet2sql 0.99.4 uses data from WordNet version 3.0 released in December 2006.

WordNet was developed at the Cognitive Science Laboratory at Princeton University by G. Miller, C. Fellbaum, and many others. For details see http://wordnet.princeton.edu/ .

license

wn2sql is free software. In a nutshell: You're free to use this software and redistribute modifications, as long as you acknowledge contributions and don't show your gratitude by suing any contributors. But note that the licensing terms, in their legal smallprint, supersede and prevail over this paragraph.

download

tables

word

wordno: is the primary key of this table. A wordno uniquely identifies a word.

lemma: is the string that represents the word.

synset

A synset can be roughly viewed as an entity of semantic information. It is a set of words that can be substituted for each other in some sentences (but not necessarily all possible sentences) without changing its truth valuation. A word can also be present in more than one synset, because a word can have several meanings.

synsetno: is the primary key for this table. It uniquely identifies a synset.

definition: The definition of the meaning according to WordNet's glossary

lexfilenum: Number of the lexicographer-file the synset is in. Ths is a foreign key referencing the explanatory table "lexname".

sense

A sense in this context is the association between the syntactic entity (a word) and the semantic information it carries (a synset).

wordno: is a link to a word. This is the word this mapping refers to.

synsetno: is a link to a synset. This is the synset this mapping refers to.

tagcnt: Number of times this sense-tag occurs in the semantic concordances. (from CNTLIST).
NOTE: The semantic concordance have not been maintained since WordNet 1.7.1.

lexrel

Lexical relations are relations between senses, that is between certain certain words as used in particular linguistic contexts. This is saved in the direction from (wordno1,synsetno1) to (wordno2,synsetno2). So the association (wordno1,synsetno1) -> (wordno2,synsetno2) has the type indicated by reltypeno, but the association (wordno2,synsetno2) -> (wordno1,synsetno2) does not necessarily have that same type.

wordno1: is the link to the source word.

synsetno1: is the link to the source synset.

wordno2: is the link to the target word.

synsetno2: is the link to the target synset.

reltypeno: is the type of the relation. Possible values can be found in the explanatory table reltype.

semrel

Semantic relations hold between synsets. This is saved in the direction from synsetno1 to synsetno2. So association synsetno1 -> synsetno2 is of type reltype, but the association synsetno2 -> synsetno1 is not necessarily of the same type.

synsetno1: is the link to the source synset.

synsetno2: is the link to the target synset.

reltypeno: is the type of the relation. Possible values can be found in the explanatory table reltype.

sample

WordNet's glossary contains a definition and/or one or more samples, illustrating the use of the synset. This table contains the samples from the WordNet-glossary.

synsetno: is the link to the synset.

sampleno: unique number identifying the sample within the synset.

samp: the text of this sample.

adjmod

Adjectives have modifiers that give information about their position within a sentence. This relates to a sense, i.e. the use of an adjective in a particular context.

synsetno: is the link to the synset.

wordno: is the link to the word.

modifier: 'p': predicate position, 'a': prenominal position, 'ip': immediately postnominal position.

lexverbframe

Verbs have frames that give information about their usage. A lexical verb frame refers to the usage of a verb in a particular context.

synsetno: is the link to the synset.

wordno: is the link to the word.

frameno: Possible values can be found in the explanatory table frame.

semverbframe

Verbs have frames that give information about their usage. A semantic verb frame refers to a synset and applies to all words in it.

synsetno: is the link to the synset.

wordno: is the link to the word.

frameno: Possible values can be found in the explanatory table frame.

(c) Copyright 2007 -- 2009