Text comparison apparatus

• US 20030028566A1
• Filed: 07/12/2002
• Published: 02/06/2003
• Est. Priority Date: 07/12/2001
• Status: Active Grant

• Current Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Inventors: Nakano, Hiraku
• Granted Patent: US 7,010,515 B2
• Time in Patent Office: Days
• US Class Current: 715/538
• CPC Class Codes: G06F 16/90344 by using string matching techniques
• CPC Class Codes: Y10S 707/931 Patent comparison
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A text comparison apparatus computes the occurrence count of text elements, stores those text elements that have an occurrence count of at least a occurrence count threshold for storage in a text element storage unit, uses those text elements that have an occurrence count of at least a occurrence count threshold for similarity calculation to calculate similarity, and calculates discrepancy for those text elements for which the difference of occurrence counts is at least a occurrence count threshold for discrepancy calculation.

Claims

1. A text comparison apparatus, comprising:

a text storage means for storing a plurality of texts;

a text input means for inputting a text from the text storage means;

a text element extraction means for extracting elements from the text obtained with the text input means;

a text element counting means for counting how many of each of the text elements are included the text;

a text element storage means for storing text elements and occurrence counts as sets;

a text element input means for inputting text elements and their occurrence counts with regard to two texts from the text element storage means; and

a similarity calculation means for calculating a similarity of the texts by dividing the sum of the occurrence counts of the text elements included in both of the two texts by the sum of the occurrence counts of all text elements in each of the texts.

2. The text comparison apparatus according to claim 1,

further comprising a storage means for storing occurrence count threshold settings for storage;

wherein text elements having an occurrence count of at least the occurrence count threshold for storage given out by the storage means for storing occurrence count threshold settings for storage, as well as their occurrence counts, are stored in the text element storage means.

3. The text comparison apparatus according to claim 1,

further comprising a storage means for storing occurrence count threshold settings for similarity calculation, wherein, when inputting text elements and occurrence counts from the text element input means and calculating similarity with the similarity calculation means, text elements having an occurrence count that is less than the occurrence count threshold for similarity calculation are not used for the similarity calculation.

4. The text comparison apparatus according to claim 1,

further comprising an uncharacteristic text element storage means;

wherein text elements and their occurrence counts are not stored in the text element storage means when those text elements are stored in the uncharacteristic text element storage means.

5. The text comparison apparatus according to claim 1,

further comprising an uncharacteristic text element storage means;

wherein text elements and their occurrence counts are not used for the similarity calculation with the similarity calculation means when those text elements are stored in the uncharacteristic text element storage means.

6. The text comparison apparatus according to claim 1, further comprising:

a storage means for storing occurrence count threshold settings for discrepancy calculation, which gives an occurrence count threshold of text elements to be used when calculating the discrepancy of two texts; and

a discrepancy calculation means for calculating the discrepancy of texts by summing up the differences between the occurrence counts of text elements included in the two texts for those text elements with a value of at least that given by the storage means for storing occurrence count threshold settings for discrepancy calculation while assigning them to the text containing more of those text elements, and dividing this sum by the sum of the occurrence counts of all text elements in each of the texts.

7. The text comparison apparatus according to claim 6,

further comprising an uncharacteristic text element storage means;

wherein text elements and their occurrence counts are not used for the calculation of discrepancy with the discrepancy calculation means when those text elements are stored in the uncharacteristic text element storage means.

8. The text comparison apparatus according to claim 1,

wherein the two compared texts are patent specifications;

and further comprising;

a means for detecting an international patent classification (IPC) code from patent specifications that have been read out via the text input means; and

an IPC code analysis means for detecting whether the two texts contain the same IPC code.

9. A text comparison apparatus, comprising:

a text storage means for storing a plurality of texts;

a text input means for inputting a text from the text storage means;

a text element extraction means for extracting elements from the text obtained with the text input means;

a text element counting means for counting how many of each of the text elements are included the text;

a text element storage means for storing text elements and occurrence counts as sets;

a text element input means for inputting text elements and their occurrence counts with regard to two texts from the text element storage means;

a storage means for storing occurrence count threshold settings for discrepancy calculation, which gives an occurrence count threshold of text elements to be used when calculating the discrepancy of two texts; and

a discrepancy calculation means for calculating the discrepancy of two texts by summing up the differences between the occurrence counts of text elements included in the two texts for those text elements with a value of at least that given by the storage means for storing occurrence count threshold settings for discrepancy calculation while assigning them to the text containing more of those text elements, and dividing this sum by the sum of the occurrence counts of all text elements in each of the texts.

10. The text comparison apparatus according to claim 9,

further comprising a storage means for storing occurrence count threshold settings for storage;

wherein text elements having an occurrence count of at least the occurrence count threshold for storage given out by the storage means for storing occurrence count threshold settings for storage, as well as their occurrence counts, are stored in the text element storage means.

11. The text comparison apparatus according to claim 9,

further comprising an uncharacteristic text element storage means;

wherein text elements and their occurrence counts are not stored in the text element storage means when those text elements are stored in the uncharacteristic text element storage means.

12. The text comparison apparatus according to claim 9,

further comprising an uncharacteristic text element storage means;

wherein text elements and their occurrence counts are not used for the calculation of discrepancy with the discrepancy calculation means when those text elements are stored in the uncharacteristic text element storage means.

13. The text comparison apparatus according to claim 9,

wherein the two compared texts are patent specifications;

and further comprising;

a means for detecting an international patent classification (IPC) code from patent specifications that have been read out via the text input means; and

an IPC code analysis means for detecting whether the two texts contain the same IPC code.

14. A text comparison apparatus, comprising:

a text storage means for storing a plurality of texts;

a text input means for inputting a text from the text storage means;

a word analysis means for analyzing words and their part of speech from the text obtained with the text input means;

a word counting means for counting how many of each of the respective words are included the text, the counting being carried out for each of the part-of-speech data in case of words that have a plurality of part-of-speech data;

a word storage means for storing words, part-of-speech data and occurrence counts as sets;

a word input means for inputting words, part-of-speech data and occurrence counts with regard to two texts from the word storage means; and

a similarity calculation means for calculating a similarity of the texts by dividing the sum of the occurrence counts of the words that are included in both of the two texts and have matching part-of-speech data by the sum of the occurrence counts of all words in each of the texts.

15. The text comparison apparatus according to claim 14,

further comprising a storage means for storing occurrence count threshold settings for storage;

wherein words having an occurrence count of at least the occurrence count threshold for storage given out by the storage means for storing occurrence count threshold settings for storage, as well as their part-of-speech data and occurrence counts, are stored in the word storage means.

16. The text comparison apparatus according to claim 14;

further comprising a storage means for storing occurrence count threshold settings for similarity calculation;

wherein, when inputting words and occurrence counts from the word input means and calculating similarity with the similarity calculation means, words having an occurrence count that is less than the occurrence count threshold for similarity calculation are not used for the similarity calculation.

17. The text comparison apparatus according to claim 14;

further comprising an uncharacteristic word storage means;

wherein words and their occurrence counts are not stored in the word storage means when those words are stored in the uncharacteristic word storage means.

18. The text comparison apparatus according to claim 14;

further comprising an uncharacteristic word storage means;

wherein words and their occurrence counts are not used for the similarity calculation with the similarity calculation means when those words are stored in the uncharacteristic word storage means.

19. The text comparison apparatus according to claim 14,

wherein the two compared texts are patent specifications;

and further comprising;

a means for detecting an international patent classification (IPC) code from patent specifications that have been read out via the text input means; and

an IPC code analysis means for detecting whether the two texts contain the same IPC code.

20. The text comparison apparatus according to claim 14,

further comprising a means for converting nouns between different languages, wherein, when the two compared texts are in different languages, based on the part-of-speech data associated with the words given out from the word storage means, parts of speech other than nouns and verbs are removed from the scope of the similarity calculation before calculating the similarity with the similarity calculation means;

and

further comprising a verb-noun conversion means, which converts verbs into corresponding nouns, and increases the occurrence counts of those nouns accordingly, whereafter the words included in one text are converted into the language of the other text, and their similarity is calculated.

21. The text comparison apparatus according to claim 14,

further comprising a verb-noun conversion means, which, only for nouns and verbs, converts verbs into corresponding nouns, and increases the occurrence counts of those nouns accordingly, whereafter they are stored in the word storage means.

22. The text comparison apparatus according to claim 21,

further comprising a means for converting nouns between different languages;

wherein, when the two compared texts are in different languages, words included in one text given out from the word storage means are converted into the language of the other text before calculating the similarity with the similarity calculation means.

23. The text comparison apparatus according to claim 14, further comprising:

a storage means for storing occurrence count threshold settings for discrepancy calculation, which gives an occurrence count threshold of words to be used when calculating the discrepancy of two texts; and

a discrepancy calculation means for calculating the discrepancy of two texts by summing up the differences between the occurrence counts of words that are included in the two texts and that have matching part-of-speech data, for those words with a value of at least that given by the storage means for storing occurrence count threshold settings for discrepancy calculation while assigning them to the text containing more of those words, and dividing this sum by the sum of the occurrence counts of all words in each of the texts.

24. The text comparison apparatus according to claim 23;

further comprising an uncharacteristic word storage means;

wherein words and their occurrence counts are not used in the discrepancy calculation with the discrepancy calculation means when those words are stored in the uncharacteristic word storage means.

25. The text comparison apparatus according to claim 23,

further comprising a means for converting nouns between different languages, wherein, when the two compared texts are in different languages, based on the part-of-speech data associated with the words given out from the word storage means, parts of speech other than nouns and verbs are removed from the scope of the similarity calculation before calculating the similarity with the similarity calculation means;

and

further comprising a verb-noun conversion means, which converts verbs into corresponding nouns, and increases the occurrence counts of those nouns accordingly, whereafter the words included in one text are converted into the language of the other text, and their similarity is calculated.

26. The text comparison apparatus according to claim 23,

further comprising a verb-noun conversion means, which, only for nouns and verbs, converts verbs into corresponding nouns, and increases the occurrence counts of those nouns accordingly, whereafter they are stored in the word storage means.

27. The text comparison apparatus according to claim 26,

further comprising a means for converting nouns between different languages;

wherein, when the two compared texts are in different languages, words included in one text given out from the word storage means are converted into the language of the other text before calculating the similarity with the similarity calculation means.

28. A text comparison apparatus, comprising:

a text storage means for storing a plurality of texts;

a text input means for inputting a text from the text storage means;

a word analysis means for analyzing words and their part of speech from the text obtained with the text input means;

a word counting means for counting how many of each of the respective words are included the text, the counting being carried out for each of the part-of-speech data in case of words that have a plurality of part-of-speech data;

a word storage means for storing words, part-of-speech data and occurrence counts as sets;

a word input means for inputting words, part-of-speech data and occurrence counts with regard to two texts from the word storage means; and

a storage means for storing occurrence count threshold settings for discrepancy calculation, which gives an occurrence count threshold of words to be used when calculating the discrepancy of two texts; and

a discrepancy calculation means for calculating the discrepancy of two texts by summing up the differences between the occurrence counts of words that are included in the two texts and that have matching part-of-speech data, for those words with a value of at least that given by the storage means for storing occurrence count threshold settings for discrepancy calculation while assigning them to the text containing more of those words, and dividing this sum by the sum of the occurrence counts of all words in each of the texts.

29. The text comparison apparatus according to claim 28,

further comprising a storage means for storing occurrence count threshold settings for storage;

wherein words having an occurrence count of at least the occurrence count threshold for storage given out by the storage means for storing occurrence count threshold settings for storage, as well as their part-of-speech data and occurrence counts, are stored in the word storage means.

30. The text comparison apparatus according to claim 28;

further comprising an uncharacteristic word storage means;

wherein words and their occurrence counts are not stored in the word storage means when those words are stored in the uncharacteristic word storage means.

31. The text comparison apparatus according to claim 28;

further comprising an uncharacteristic word storage means;

wherein words and their occurrence counts are not used for the discrepancy calculation with the discrepancy calculation means when those words are stored in the uncharacteristic word storage means.

32. The text comparison apparatus according to claim 28,

wherein the two compared texts are patent specifications;

and further comprising;

a means for detecting an international patent classification (IPC) code from patent specifications that have been read out via the text input means; and

an IPC code analysis means for detecting whether the two texts contain the same IPC code.

33. The text comparison apparatus according to claim 28,

further comprising a means for converting nouns between different languages, wherein, when the two compared texts are in different languages, based on the part-of-speech data associated with the words given out from the word storage means, parts of speech other than nouns and verbs are removed from the scope of the discrepancy calculation before calculating the discrepancy with the discrepancy calculation means;

and

further comprising a verb-noun conversion means, which converts verbs into corresponding nouns, and increases the occurrence counts of those nouns accordingly, whereafter the words included in one text are converted into the language of the other text, and their similarity is calculated.

34. The text comparison apparatus according to claim 28,

further comprising a verb-noun conversion means, which, only for nouns and verbs, converts verbs into corresponding nouns, and increases the occurrence counts of those nouns accordingly, whereafter they are stored in the word storage means.

35. The text comparison apparatus according to claim 34,

further comprising a means for converting nouns between different languages;

wherein, when the two compared texts are in different languages, words included in one text given out from the word storage means are converted into the language of the other text before calculating the discrepancy with the discrepancy calculation means.