US5638543A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method and apparatus for automatic document summarization
Assignee: Xerox Corp
•
Filing: 1993-06-03
•
Granted: 1997-06-10
•
Expires: 2013-06-03
Abstract
Abstract Regions of a document such as sentences and blocks of sentences are scored and classified based upon their scores. An abstract of the document can be formed from the classified sentences. Sentences are classified by the use of words classified as stop words and vanish words. Sentences are scored based on the number of stop words and the number of strings of connected stop words, called stop-word runs, contained in the sentence. Passionate sentences, which usually contain information which the writer has strong feelings about, such as joy, admiration, or sadness, are identified. This method can also select sentences that are contrapassionate, which the writer may either have to strengthen or have inserted to complete the record and provide continuity or information.
Inventors
- 👤 Jan O. Pedersen
- 👤 John W. Tukey
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-28 15:48
CPC Classifications
G; G06; G06F; G06F16/00; G06F16/30; G06F16/34; G06F16/345
Claims Analysis
Total Claims: 64
Claim 1
Independent Claim
74 words
A method for automatically identifying regions of text from an electronic document using a digital computer, comprising the steps of: identifying stop words within the document; determining whether any of the identified stop words are vanish words; scoring regions of the document in response to the stop words and the vanish words in each of the regions; and identifying a predetermined number of regions of the document based on the score of the regions.
Claim 2
Independent Claim
74 words
A method for automatically identifying regions of text from an electronic document using a digital computer, comprising the steps of: identifying stop words within the document; determining whether any of the identified stop words are vanish words; scoring regions of the document in response to the stop words and the vanish words in each of the regions; and identifying a predetermined number of regions of the document based on the score of the regions.
Claim 3
Dependent Claim
15 words
The method according to claim 1, wherein the regions of the document correspond to sentences.
Claim 4
Dependent Claim
15 words
The method according to claim 1, wherein the regions of the document correspond to paragraphs.
Claim 5
Dependent Claim
25 words
The method according to claim 1, wherein the stop words are identified by determining whether words in a region are on a predetermined STOP list.
Claim 6
Dependent Claim
25 words
The method according to claim 4, wherein the vanish words are determined by determining whether the identified stop words are on a predetermined VANISH list.
Claim 7
Dependent Claim
67 words
The method according to claim 5, wherein the scoring step includes: reducing the length of each stop-word run by eliminating the vanish words from each stop-word run to form reduced stop-word runs, wherein each stop-word run is a maximal group of adjacent stop words having at least one stop word; and combining the reduced stop-word runs in a predetermined manner to generate the score of the region.
Claim 8
Dependent Claim
35 words
The method according to claim 6, wherein the combining in the predetermined manner is performed so that a further reduction of the reduced stop-word runs has an effect of non-decreasing a value of the score.
Claim 9
Dependent Claim
68 words
The method according to claim 5, wherein the score of each region of the document is determined by: subtracting a total number of vanish words from a total number of stop words in a region to determine a result; and dividing the result by a total number of stop-word runs, wherein a stop-word run is a maximal group of adjacent stop words having at least one stop word.
Claim 10
Dependent Claim
19 words
The method according to claim 8, wherein a predetermined numeral is added to the result before dividing the result.
Claim 11
Dependent Claim
50 words
The method according to claim 5, wherein the score of each region is determined by selecting long sequences of short stop-word runs, wherein a short stopword run is a maximal group of adjacent stop words having either one stop word or one stop word with at least one vanish word.
Claim 12
Dependent Claim
27 words
The method according to claim 1, wherein the stop words are identified by determining whether words in the region have a length within a predetermined length range.
Claim 13
Dependent Claim
18 words
The method according to claim 11, wherein the predetermined length range is between one letter and three letters.
Claim 14
Dependent Claim
29 words
The method according to claim 11, wherein the vanish words are determined by determining whether each identified stop word has a second predetermined length within said predetermined length range.
Claim 15
Dependent Claim
18 words
The method according to claim 13, wherein the second predetermined length is either one letter or three letters.
Claim 16
Dependent Claim
70 words
The method according to claim 13, wherein the score of each region is determined by: counting a total number of stop-word runs having a non-vanish word in the region to determine a result, wherein the stop-word run is a maximal group of adjacent stop words having at least one stop word; and dividing the result by a total number of stop-word runs in the region to form a passion score.
Claim 17
Dependent Claim
19 words
The method according to claim 15, wherein a predetermined numeral is added to the result before dividing the result.
Claim 18
Dependent Claim
66 words
The method according to claim 14, wherein the score of each region is determined by: counting a total number of stop-word runs having all vanish words in the region, wherein the stopword run is a maximal group of adjacent stop words containing at least one stop word; and dividing the result by a total number of stop-word runs in the region to form a contrapassion score.
Claim 19
Dependent Claim
19 words
The method according to claim 17, wherein a predetermined numeral is added to the result before dividing the result.
Claim 20
Dependent Claim
27 words
The method according to claim 1, wherein a combination of one stop word immediately adjacent to one or more vanish words is counted as one stop word.
Claim 21
Dependent Claim
15 words
The method according to claim 1, wherein each word is represented by a word shape.
Claim 22
Dependent Claim
24 words
The method according to claim 1, further comprising outputting the predetermined number of regions as an abstract of the document with an output device.
Claim 23
Dependent Claim
30 words
The method according to claim 1, further comprising querying a user of said digital computer as to whether regions of the document having a predetermined score are to be edited.
Claim 24
Independent Claim
51 words
A method for automatically scoring regions of text from an electronic document using a digital computer, comprising the steps of: identifying stop words within the document; and scoring regions of the document in response to a total number of stop words and relative locations of the stop words within the region.
Claim 25
Independent Claim
51 words
A method for automatically scoring regions of text from an electronic document using a digital computer, comprising the steps of: identifying stop words within the document; and scoring regions of the document in response to a total number of stop words and relative locations of the stop words within the region.
Claim 26
Dependent Claim
50 words
The method according to claim 23, wherein the score of each region of the document is determined by dividing the total number of stop words by the total number of stop-word runs, wherein a stop-word run is a maximal group of adjacent stop words having at least one stop word.
Claim 27
Dependent Claim
25 words
The method according to claim 23, wherein a vanish word is identified by determining if the identified stop words are on a predetermined VANISH list.
Claim 28
Dependent Claim
67 words
The method according to claim 25, wherein the scoring step includes: reducing the length of each stop-word run by eliminating the vanish words from each stop-word run to form reduced stop-word runs, wherein each stop-word run is a maximal group of adjacent stop words having at least one stop word; and combining the reduced stop-word runs in a predetermined manner to generate the score of the region.
Claim 29
Dependent Claim
35 words
The method according to claim 26, wherein the combining in the predetermined manner is performed so that a further reduction of the reduced stop-word runs has an effect of non-decreasing a value of the score.
Claim 30
Dependent Claim
68 words
The method according to claim 23, wherein the score of each region of the document is determined by: subtracting a total number of vanish words from a total number of stop words in a region to determine a result; and dividing the result by a total number of stop-word runs, wherein a stop-word run is a maximal group of adjacent stop words having at least one stop word.
Claim 31
Dependent Claim
19 words
The method according to claim 28, wherein a predetermined numeral is added to the result before dividing the result.
Claim 32
Dependent Claim
50 words
The method according to claim 27, wherein the score of each region is determined by identifying long sequences of short stop-word runs, wherein a short stopword run is a maximal group of adjacent stop words having either one stop word or one stop word with at least one vanish word.
Claim 33
Dependent Claim
46 words
The method according to claim 25, wherein the stop words are identified by determining whether words in the region have a length within a first predetermined length range and a vanish word is identified by determining whether each identified stop word has a second predetermined length.
Claim 34
Dependent Claim
32 words
The method according to claim 31, wherein the first predetermined length range is between one letter and three letters, and the second predetermined length range is either one letter or three letters.
Claim 35
Dependent Claim
70 words
The method according to claim 31, wherein the score of each region is determined by: counting a total number of stop-word runs having a non-vanish word in the region to determine a result, wherein the stop-word run is a maximal group of adjacent stop words having at least one stop word; and dividing the result by a total number of stop-word runs in the region to form a passion score.
Claim 36
Dependent Claim
66 words
The method according to claim 31, wherein the score of each region is determined by: counting a total number of stop-word runs having all vanish words in the region, wherein the stopword run is a maximal group of adjacent stop words containing at least one stop word; and dividing the result by a total number of stop-word runs in the region to form a contrapassion score.
Claim 37
Independent Claim
51 words
A method for automatically scoring sentences of text from an electronic document using a digital computer, comprising the steps of: identifying stop words within the document; and scoring sentences of the document in response to a total number of stop words and relative locations of the stop words within the sentences.
Claim 38
Independent Claim
51 words
A method for automatically scoring sentences of text from an electronic document using a digital computer, comprising the steps of: identifying stop words within the document; and scoring sentences of the document in response to a total number of stop words and relative locations of the stop words within the sentences.
Claim 39
Dependent Claim
50 words
The method according to claim 35, wherein the score of each sentence of the document is determined by dividing the total number of stop words by the total number of stop-word runs, wherein a stop-word run is a maximal group of adjacent stop words having at least one stop word.
Claim 40
Dependent Claim
24 words
The method according to claim 35, wherein a vanish word is identified by determining whether each stop word is on a predetermined VANISH list.
Claim 41
Dependent Claim
68 words
The method according to claim 37, wherein the score of each sentence of the document is determined by: subtracting a total number of vanish words from a total number of stop words in a sentence to determine a result; and dividing the result by a total number of stop-word runs, wherein a stop-word run is a maximal group of adjacent stop words having at least one stop word.
Claim 42
Dependent Claim
50 words
The method according to claim 37, wherein the score of each sentence is determined by selecting long sequences of short stop-word runs, wherein a short stopword run is a maximal group of adjacent stop words having either one stop word or one stop word with at least one vanish word.
Claim 43
Dependent Claim
46 words
The method according to claim 35, wherein the stop words are identified by determining whether words in the sentence have a length within a first predetermined length range and a vanish word is identified by determining whether each identified stop word has a second predetermined length.
Claim 44
Dependent Claim
33 words
The method according to claim 40, wherein the first predetermined length range is between one letter and three letters, and the second predetermined length is to have either one letter or three letters.
Claim 45
Dependent Claim
70 words
The method according to claim 40, wherein the score of each sentence is determined by: counting a total number of stop-word runs having a non-vanish word in the sentence to determine a result, wherein the stop-word run is a maximal group of adjacent stop words having at least one stop word; and dividing the result by a total number of stop-word runs in the sentence to form a passion score.
Claim 46
Dependent Claim
66 words
The method according to claim 40, wherein the score of each sentence is determined by: counting a total number of stop-word runs having all vanish words in the sentence, wherein the stopword run is a maximal group of adjacent stop words containing at least one stop word; and dividing the result by a total number of stop-word runs in the sentence to form a contrapassion score.
Claim 47
Independent Claim
55 words
An apparatus for automatically identifying regions of text from an electronic document, comprising: means for identifying stop words within the document; means for selecting regions of the document in response to a total number of stop words and relative locations of stop words within the region; and an output device which utilizes the selected regions.
Claim 48
Independent Claim
55 words
An apparatus for automatically identifying regions of text from an electronic document, comprising: means for identifying stop words within the document; means for selecting regions of the document in response to a total number of stop words and relative locations of stop words within the region; and an output device which utilizes the selected regions.
Claim 49
Dependent Claim
15 words
The apparatus according to claim 44, further comprising an input means for inputting the document.
Claim 50
Dependent Claim
20 words
The apparatus according to claim 44, wherein the means for identifying and the means for selecting comprise a digital computer.
Claim 51
Dependent Claim
56 words
The apparatus according to claim 44, wherein the means for selecting each region of the document by determining a score of each region by dividing the total number of stop words by the total number of stop-word runs, wherein a stop-word run is a maximal group of adjacent stop words having at least one stop word.
Claim 52
Dependent Claim
53 words
The apparatus according to claim 44, wherein the means for selecting each region of the document is by selecting long sequences of short stop-word runs, wherein a short stop-word run is a maximal group of adjacent stop words having either one stop word or one stop word with at least one vanish word.
Claim 53
Dependent Claim
45 words
The apparatus according to claim 44, wherein the means for identifying stop words determines whether words in the region have a length within a first predetermined length range and identifies a vanish word by determining whether each identified stop word has a second predetermined length.
Claim 54
Dependent Claim
76 words
The apparatus according to claim 49, wherein the means for selecting each region by determining a score of each region by: counting a total number of stop-word runs having a non-vanish word in the region to determine a result, wherein the stop-word run is a maximal group of adjacent stop words having at least one stop word; and dividing the result by a total number of stop-word runs in the region to form a passion score.
Claim 55
Dependent Claim
72 words
The apparatus according to claim 49, wherein the means for selecting each region by determining a score of each region by: counting a total number of stop-word runs having all vanish words in the region, wherein the stopword run is a maximal group of adjacent stop words containing at least one stop word; and dividing the result by a total number of stop-word runs in the region to form a contrapassion score.
Claim 56
Independent Claim
69 words
An apparatus for automatically identifying regions of text from an electronic document, comprising: means for identifying stop words within the document; means for scoring regions of the document for selecting a predetermined number of regions of the document, the score being based on a total number of stop words and relative locations of stop words within the region; and an output device which utilizes the scoring of the regions.
Claim 57
Independent Claim
69 words
An apparatus for automatically identifying regions of text from an electronic document, comprising: means for identifying stop words within the document; means for scoring regions of the document for selecting a predetermined number of regions of the document, the score being based on a total number of stop words and relative locations of stop words within the region; and an output device which utilizes the scoring of the regions.
Claim 58
Dependent Claim
15 words
The apparatus according to claim 52, further comprising an input means for inputting the document.
Claim 59
Dependent Claim
20 words
The apparatus according to claim 52, wherein the means for identifying and the means for scoring comprise a digital computer.
Claim 60
Dependent Claim
52 words
The apparatus according to claim 52, wherein the means for scoring each region of the document determines the score by dividing the total number of stop words by the total number of stop-word runs, wherein a stop-word run is a maximal group of adjacent stop words having at least one stop word.
Claim 61
Dependent Claim
53 words
The apparatus according to claim 52, wherein the means for scoring each region of the document is by selecting long sequences of short stop-word runs, wherein a short stop-word run is a maximal group of adjacent stop words having either one stop word or one stop word with at least one vanish word.
Claim 62
Dependent Claim
50 words
The apparatus according to claim 52, wherein the means for identifying stop words determines whether words in the region have a length within a first predetermined length range and identifies a vanish word is identified by determining whether each identified stop word has one of a set of predetermined lengths.
Claim 63
Dependent Claim
72 words
The apparatus according to claim 57, wherein the means for scoring each region determines the score by: counting a total number of stop-word runs having a non-vanish word in the region to determine a result, wherein the stop-word run is a maximal group of adjacent stop words having at least one stop word; and dividing the result by a total number of stop-word runs in the region to form a passion score.
Claim 64
Dependent Claim
68 words
The apparatus according to claim 57, wherein the means for scoring each region determines the score by: counting a total number of stop-word runs having all vanish words in the region, wherein the stopword run is a maximal group of adjacent stop words containing at least one stop word; and dividing the result by a total number of stop-word runs in the region to form a contrapassion score.
References Cited (Backward Citations)
US5077668A
Cited
US5297027A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2000-06-09 | 2000-12-09 | 2001-06-10 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2004-06-09 | 2004-12-09 | 2005-06-10 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2008-06-09 | 2008-12-09 | 2009-06-10 | $7700.00 | Window Closed / Lapsed |