Apparatus and computerised method for determining constituent words of a compound word

• US 20050222998A1
• Filed: 03/30/2005
• Published: 10/06/2005
• Est. Priority Date: 03/31/2004
• Status: Active Grant

• Current Assignee: Océ-Technologies B.V. (Canon Inc.)
• Original Assignee: Océ-Technologies B.V. (Canon Inc.)
• Inventors: Iijin, Pavel M., Driessen, Samuel J.
• Granted Patent: US 7,720,847 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates

Abstract

An apparatus, a computer program and a computerized method for determining constituent words of a compound word are provided. Constituent words constitute a compound word. When the constituent words comply to split decision criteria then the constituent words can be used in a separate form. The separate form of the constituent words is used in the search to retrieve the related documents from the document collection.

Claims

1. An apparatus for determining constituent words of a compound word, the apparatus comprising:

a document collection;

means for determining, from the document collection, a number of documents containing the compound word;

means for determining, from the document collection, a number of documents containing constituent words constituting the compound word;

means for determining a ratio between the number of documents containing the compound word and the number of documents containing the constituent words constituting the compound word; and

means for splitting the compound word into the constituent words when the ratio is smaller than a threshold value.

2. The apparatus according to claim 1, further comprising:

means for detecting a linking morpheme in the compound word.

3. The apparatus according to claim 1, further comprising:

means for finding the constituent words by using a word list derived from the document collection.

4. The apparatus according to claim 1, wherein the means for determining the number of documents containing the compound word determines the number of documents containing the compound word from a word list derived from the document collection.

5. The apparatus according to claim 1, wherein the means for determining the number of documents containing the constituent words constituting the compound word determines the number of documents containing the constituent words constituting the compound word from a word list derived from the document collection.

6. The apparatus according to claim 1, wherein the threshold value is dependent on the document collection.

7. The apparatus according to claim 1, wherein the threshold value is 3.

8. The apparatus according to claim 1, the compound word and constituent words belong to a grammatical category of nouns.

9. The apparatus according to claim 3, wherein the word list includes single and plural forms of each noun.

10. An information retrieval system comprising:

means for entering at least one search term comprising a compound word;

means for determining, from a document collection, a number of documents containing the compound word;

means for determining, from the document collection, a number of documents containing constituent words constituting the compound word;

means for determining a ratio between the number of documents containing the compound word and he number of documents containing the constituent words constituting the compound word;

means for splitting the compound word into the constituent words when the ratio is smaller than a threshold value; and

means for carrying out a search with the delivered constituent words as search terms.

11. A method for determining constituent words of a compound word, the method comprising the steps of:

determining, from a document collection, a number of documents containing a compound word;

determining, from the document collection, a number of documents containing constituent words constituting the compound word;

determining a ratio between the number of documents containing the compound word and the number of documents containing the constituent words constituting the compound word; and

splitting the compound word into the constituent words when the ratio is smaller than a threshold value.

12. The method according to claim 11, further comprising:

detecting a linking morpheme in the compound word.

13. The method according to claim 11, further comprising:

finding the constituent words by using a word list derived from the document collection.

14. The method according to claim 11, wherein the step of determining the number of documents containing the compound word determines the number of documents containing the compound word from a word list derived from the document collection.

15. The method according to claim 11, wherein the means step for determining the number of documents containing the constituent words constituting the compound word determines the number of documents containing the constituent words constituting the compound word from a word list derived from the document collection.

16. The method according to claim 11, wherein the threshold value depends on the document collection.

17. The method according to claim 11, wherein the threshold value is 3.

18. The method according to claim 11, wherein the compound word and constituent words belong to a grammatical category of nouns.

19. The method according to claim 13, wherein the word list includes single and plural forms of each noun.

20. A method for information retrieval, the method comprising the steps of:

entering at least one search term comprising a compound word;

determining, from a document collection, a number of documents containing the compound word;

determining, from the document collection, a number of documents containing constituent words constituting the compound word;

determining a ratio between the number of documents containing the compound word and the number of documents containing the constituent words constituting the compound word;

splitting the compound word into the constituent words when the ratio is smaller than a threshold value;

delivering the constituent words; and

carrying out a search with the delivered constituent words as search terms.

21. A computer program product embodied on at least one computer-readable medium, for determining constituent words of a compound word, the product comprising computer-executable instructions for:

determining, from a document collection, a number of documents containing a compound word;

determining, from the document collection, a number of documents containing constituent words constituting the compound word;

determining a ratio between the number of documents containing the compound word and the number of documents containing the constituent words constituting the compound word; and

splitting the compound word into the constituent words when the ratio is smaller than a threshold value.

22. A computer program product embodied on at least one computer-readable medium, for retrieving information, the product comprising computer-executable instructions for:

entering at least one search term comprising a compound word;

determining, from a document collection, a number of documents containing the compound word;

determining, from the document collection, a number of documents containing constituent words constituting the compound word;

determining a ratio between the number of documents containing the compound word and the number of documents containing the constituent words constituting the compound word;

splitting the compound word into the constituent words when the ratio is smaller than a threshold value;

delivering the constituent words; and

carrying out a search with the delivered constituent words as search terms.