JP2019204445A - Information processing apparatus, information processing method, and program - Google Patents

Abstract

To make it possible to extract a more appropriate summarized word from a document.SOLUTION: An information processing apparatus comprises: an acquisition unit that acquires a document consisting of one or more sentences; a basic score imparting unit that assigns weights to the words included in the sentences according to their case elements; a score update unit that updates the weights of the words included in the sentences according to the predicates in the sentences; a summarized word selection unit that selects a summarized word on the basis of the weights of the words; and an output unit that outputs the selected summarized word.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an information processing apparatus, an information processing method, and a program.

  Patent Document 1 describes a technique for extracting a keyword (summary word) using various information in a document, particularly information at a parsing result level under a unified framework.

JP-A-8-44763

Here, in a sentence, whether a subject is important or an object is important differs depending on the sentence. For example, in the sentence “00 television has won three ratings”, the subject “00 television” is an important summary word. On the other hand, in the sentence “OO TV reported about the election of Mr. △△”, the subject “OO TV” is not an important summary word. In this way, even if the phrase includes the same particle, the importance of the phrase differs depending on the sentence.
However, in the technique described in Patent Document 1, the weight of a word included in a phrase is determined based on the particle of the phrase, but whether the subject is important or the object is important depends on the sentence. . For this reason, the technique described in Patent Document 1 has a problem that an appropriate summary word may not be extracted depending on the sentence.

  The present invention has been made in view of the above points, and an object thereof is to provide an information processing apparatus, an information processing method, and a program that can extract a more appropriate summary word from a sentence.

  (1) The present invention has been made to solve the above problems, and one aspect of the present invention is an acquisition unit that acquires a document composed of one or more sentences, and each word included in the sentence. , A basic score giving unit for weighting according to the case element, a score updating unit for updating the weight of each word included in the sentence according to the predicate in the sentence, and the document based on the weight of each word An information processing apparatus includes: a summary word selection unit that selects a summary word; and an output unit that outputs the selected summary word.

  (2) Moreover, one aspect of the present invention is the information processing apparatus according to (1), including a summary word integration unit that adds together the weights of words that are regarded as the same.

  (3) One embodiment of the present invention is the information processing apparatus according to (1) or (2), in which the score update unit updates weights according to different rules according to the category of the document.

  (4) Moreover, one aspect of the present invention is the information processing apparatus according to any one of (1) to (3), in which the basic score giving unit is a subject of the sentence when the sentence is active If the sentence is passive, the object weight of the sentence is increased.

  (5) Moreover, one aspect of the present invention is the information processing apparatus according to (4), in which the score update unit, when a specific predicate is included in a sentence, the weight and purpose of the subject of the sentence Reverse the weight of the word.

  (6) Moreover, 1 aspect of this invention is an information processing apparatus as described in (5), Comprising: The said specific predicate is a verb meaning the act which provides information, Comprising: The said score update part, When the specific predicate is included in a sentence, the weight of information provided in the sentence is increased.

  (7) One aspect of the present invention is the information processing apparatus according to (6), in which the specific predicate differs according to a category of the document.

  (8) In addition, the present invention has been made to solve the above-described problems, and one aspect of the present invention is a process in which an acquisition unit acquires a document composed of one or more sentences and grants a basic score. A process of weighting each word included in the document according to a case element, and a process of updating a weight of each word included in the sentence according to a predicate in the sentence The information processing method includes: a summary word selection unit selecting a summary word of the document based on the weight of each word; and an output unit outputting a selected summary word.

  (9) Further, the present invention has been made to solve the above problems, and one aspect of the present invention includes a step in which a computer acquires a document composed of one or more sentences, and the sentence includes Weighting each word according to its case element, updating the weight of each word included in the sentence according to the predicate in the sentence, and summarizing the document based on the weight of each word A program for executing a step of selecting a word and a step of outputting a selected summary word.

  According to the present invention, a more appropriate summary word can be extracted from a document.

It is a schematic block diagram which shows an example of a function structure of the summary word production | generation apparatus which concerns on the 1st Embodiment of this invention. It is the schematic which shows the data structure and example of a data of the basic score rule table which the basic score rule memory | storage part which concerns on the 1st Embodiment of this invention memorize | stores. It is the schematic which shows the data structure and data example of a specific predicate list table which the specific predicate list memory | storage part which concerns on the 1st Embodiment of this invention memorize | stores. It is the schematic which shows the data structure and data example of the score update rule table which the score update rule memory | storage part which concerns on the 1st Embodiment of this invention memorize | stores. It is a schematic block diagram which shows an example of the hardware constitutions of the summary word production | generation apparatus which concerns on the 1st Embodiment of this invention. It is a figure for demonstrating the process which the summary word production | generation apparatus which concerns on the 1st Embodiment of this invention analyzes the morpheme and syntax of a sentence. It is a figure for demonstrating the process which the summary word production | generation apparatus which concerns on the 1st Embodiment of this invention provides a score. It is a figure for demonstrating the process which the summary word production | generation apparatus which concerns on the 1st Embodiment of this invention provides a score. It is a figure for demonstrating the process which the summary word production | generation apparatus which concerns on the 1st Embodiment of this invention provides a score. It is a flowchart which shows an example of the summary word production | generation process which the summary word production | generation apparatus which concerns on the 1st Embodiment of this invention performs. It is a schematic block diagram which shows an example of a function structure of the summary word production | generation apparatus which concerns on the 2nd Embodiment of this invention. It is the schematic which shows the data structure and data example of the category specific predicate list table which the category specific predicate list memory | storage part which concerns on the 2nd Embodiment of this invention memorize | stores. It is a figure for demonstrating the process which the summary word production | generation apparatus which concerns on the 2nd Embodiment of this invention provides a score. It is a figure for demonstrating the process which the summary word production | generation apparatus which concerns on the 2nd Embodiment of this invention provides a score. It is a flowchart which shows an example of the summary word production | generation process which the summary word production | generation apparatus which concerns on the 2nd Embodiment of this invention performs.

(First embodiment)
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a schematic block diagram showing an example of a functional configuration of a summary word generation device 1 according to the first embodiment of the present invention.
The summary word generation device 1 is an information processing device that generates and outputs a summary word of an input document from an input document composed of one or a plurality of sentences. The input document is, for example, a news sentence released on the Internet or the like. The summary word is an important word in the input document such as a keyword, and is a word that summarizes the input document. That is, the summary word generation device 1 generates a summary word that expresses the subject matter of a sentence in a single word from a document composed of one or more input sentences. Thereby, for example, a message such as “News about OO” can be generated from a news article. The summary word may be a single word, a plurality of words, or a combination of a plurality of words.

  The summary word generation device 1 is, for example, a server device to which a plurality of terminal devices can be connected, and outputs summary words of an input document to the connected terminal devices. Or the summary word production | generation apparatus 1 may be integrated in each terminal devices, such as a mobile telephone, a smart phone, a tablet terminal, and a personal computer. For example, when introducing a news published on the Internet, the terminal device uses the summary word “OO” of the news acquired from the abstract word generation device 1 and reads “News about OO”. Etc. can be provided.

The summary word generation device 1 includes an input unit 11, a control unit 12, a storage unit 13, and an output unit 14.
The input unit 11 is an interface for inputting an input document, such as a keyboard, a microphone, or a network interface. The input unit 11 outputs the input document that has been input to the control unit 12. Note that the input unit 11 may acquire an input sentence from a network or another device.

  The control unit 12 controls the summary word generation device 1 in an integrated manner. The control unit 12 includes a morpheme / syntax analysis unit 121, a basic score assignment unit 122, a score update unit 123, a summary word integration unit 124, and a summary word selection unit 125.

The morpheme / syntax analysis unit 121 analyzes the morpheme and syntax of each sentence in the input document based on the analysis dictionary stored in the analysis dictionary storage unit 131, and displays the part of speech and dependency information of each clause in each sentence. Assigned to the input document and output to the basic score assigning unit 122.
The basic score assigning unit 122 assigns a weight (hereinafter also referred to as “score”) to each phrase word included in the input document according to the case element. The phrase word is, for example, a phrase obtained by removing a case particle from a case element. Specifically, the basic score assigning unit 122 assigns a score to each phrase word based on the basic score rule table stored in the basic score rule storage unit 132. For example, the basic score assigning unit 122 may increase the weight of the subject of the sentence when the sentence is active and increase the weight of the object of the sentence when the sentence is passive. The basic score assignment unit 122 outputs the score assigned to each word and the input document to the score update unit 123.

  The score updating unit 123 updates the word weight of each clause in each sentence according to the predicate of each sentence included in the input document. Specifically, the score update unit 123 stores the score update rule table stored in the score update rule storage unit 134 when the specific predicate in the specific predicate list stored in the specific predicate list storage unit 133 is a statement predicate. Based on this, the score of the word in each clause of the sentence is updated. The specific predicate is a verb meaning an act of providing information, for example. The specific predicate is not limited to a verb meaning an act of providing information, and may be another verb. When the specific predicate is included in the sentence, the score update unit 123 reverses the subject weight and the object weight of the sentence. That is, when the specific predicate is included in the sentence, the score update unit 123 increases the weight of the information provided in the sentence. In other words, when the specific predicate is included in the sentence, the score update unit 123 updates the weighting of the information provided in the sentence. The score update unit 123 outputs the updated score and the input document to the summary word integration unit 124.

When there are a plurality of sentences in the input document, the summary word integration unit 124 compares the sentences and adds scores for similar words that can be regarded as the same, and the summed word and the input document are summed up. Output to.
The summary word selection unit 125 selects a summary word of the input document based on the score of each word, and outputs the selected summary word to the output unit 14. Specifically, the summary word selection unit 125 sorts all words in descending order of score, and selects the word with the highest score as the summary word. In addition, when there are a plurality of words having the highest score, the summary word selection unit 125 may select the one having the highest basic score assigned by the basic score assignment unit 122 from among the words having the highest score. You can choose a word or choose it randomly.

  The storage unit 13 includes an analysis dictionary storage unit 131, a basic score rule storage unit 132, a specific predicate list storage unit 133, and a score update rule storage unit 134, and stores various types of information.

The analysis dictionary storage unit 131 stores an analysis dictionary for analyzing document morphemes and syntax.
The basic score rule storage unit 132 stores a basic score rule table indicating basic score rules when a score is given.

The specific predicate list storage unit 133 stores a specific predicate list indicating the specific predicate.
The score update rule storage unit 134 stores a score update rule table indicating score update rules when updating the score.

  The output unit 14 is an interface that outputs summary words such as a display, a speaker, and a network interface. The output unit 14 outputs the summary word input from the summary word selection unit 125.

FIG. 2 is a schematic diagram illustrating a data configuration and a data example of a basic score rule table stored in the basic score rule storage unit 132 according to the first embodiment of the present invention.
As shown in the figure, the basic score rule table includes items of sentence state, phrase case, and score. The sentence state indicates whether the sentence is active or passive. The phrase case indicates the phrase case particle. The score indicates the weight attached to the phrase word.

  In the illustrated example, the score corresponding to the sentence state “active” and the phrase case “ga” is “1.0”. The score corresponding to the sentence state “active” and the phrase case “wo” is “0.8”. The score corresponding to the sentence state “active” and the phrase case “second case” is “0.5”. The score corresponding to the sentence state “active” and the phrase case “predicate” is “0.0”. The score corresponding to the sentence state “passive” and the phrase case “ga” is “0.8”. The score corresponding to the sentence state “passive” and the phrase case “second case” is “1.0”.

FIG. 3 is a schematic diagram illustrating a data configuration and a data example of the specific predicate list table stored in the specific predicate list storage unit 133 according to the first embodiment of the present invention.
As shown in the figure, the specific predicate list table has items of a category and a specific predicate list. The category is a document category. The specific predicate list is a list of specific predicates. The specific predicate is a verb meaning an act of providing information by a media such as a newspaper, a television, or the Internet. In the illustrated example, the specific predicate list corresponding to the category “all” includes predicates such as “publish, introduce, report,...”.

FIG. 4 is a schematic diagram illustrating a data configuration and a data example of the score update rule table stored in the score update rule storage unit 134 according to the first embodiment of the present invention.
As shown in the figure, the score update rule table has items of sentence state, phrase case, and score. The sentence state indicates whether the sentence is active or passive. The phrase case indicates the phrase case particle. The score indicates a change value of the score, “+” indicates an added point, and “−” indicates a deducted point.

  In the illustrated example, the score corresponding to the sentence state “active” and the phrase case “ga” is “−0.5”. The score corresponding to the sentence state “active” and the phrase case “wo” is “+0.5”. The score corresponding to the sentence state “active” and the phrase case “second case” is “−0.5”. The score corresponding to the sentence state “active” and the phrase case “predicate” is “0.0”. The score corresponding to the sentence state “passive” and the phrase case “ga” is “+0.5”. The score corresponding to the sentence state “passive” and the phrase case “second case” is “−0.5”.

  That is, in the score update rule of the example shown in the figure, when the predicate of the sentence is a specific predicate, the case element that is assumed to include the media name (ga case for active and two cases for passive) Is deducted, and a case element (wo case for active and ga case for passive) that is assumed to contain information conveyed by the media is added.

FIG. 5 is a schematic block diagram illustrating an example of a hardware configuration of the summary word generation device 1 according to the first embodiment of the present invention.
The summary word generation device 1 includes a CPU 101, a storage medium interface unit 102, a storage medium 103, an input device 104, an output device 105, a ROM 106 (Read Only Memory), a RAM 107 (Random Access Memory), and an auxiliary storage. Unit 108 and a network interface unit 109. The CPU 101, the storage medium interface unit 102, the input device 104, the output device 105, the ROM 106, the RAM 107, the auxiliary storage unit 108, and the network interface unit 109 are connected to each other via a bus.
The CPU 101 here indicates a general processor, and includes not only a device called a CPU in a narrow sense but also a GPU, a DSP, and the like. The CPU 101 referred to here is not limited to being realized by a single processor, and may be realized by combining a plurality of processors of the same or different types.

The CPU 101 reads out and executes programs stored in the auxiliary storage unit 108, the ROM 106, and the RAM 107, reads various data stored in the auxiliary storage unit 108, the ROM 106, and the RAM 107, and stores various data in the auxiliary storage unit 108, the RAM 107. The summary word generation device 1 is controlled by writing. In addition, the CPU 101 reads various data stored in the storage medium 103 via the storage medium interface unit 102 and writes various data to the storage medium 103. The storage medium 103 is a portable storage medium such as a magneto-optical disk, a flexible disk, or a flash memory, and stores various data.
The storage medium interface unit 102 is an interface for reading from and writing to the storage medium 103.

The input device 104 is an input device such as a mouse, a keyboard, a touch panel, a volume adjustment button, a power button, a setting button, and an infrared receiver.
The output device 105 is an output device such as a display unit or a speaker.
The ROM 106 and RAM 107 store programs and various data for operating the functional units of the summary word generation device 1.
The auxiliary storage unit 108 is a hard disk drive, a flash memory, or the like, and stores a program for operating each functional unit of the summary word generation device 1 and various data.
The network interface unit 109 has a communication interface and is connected to the network NW by wireless communication.

  For example, the control unit 12 in the functional configuration of the summary word generation device 1 corresponds to the CPU 101, the storage unit 13 corresponds to the ROM 106, the RAM 107, the auxiliary storage unit 108, or any combination thereof, and the input unit 11 corresponds to the input device 104 or the network interface unit 109, and the output unit 14 corresponds to the output device 105 or the network interface unit 109.

  Next, the operation of the summary word generation device 1 will be described in detail using a specific example. First, the morpheme / syntax analyzer 121 analyzes the morpheme and syntax of each sentence included in the input document.

  FIG. 6 is a diagram for explaining processing in which the summary word generation device 1 according to the first exemplary embodiment of the present invention analyzes sentence morphemes and syntaxes. In the example shown in the figure, the input sentence is “00 television reported about the winning of △△”. First, the morpheme / syntax analyzing unit 121 divides the input sentence into the phrases “00 TV is”, “△△ Ms.”, “Winning” and “reported”. Then, the morpheme / syntax analyzing unit 121 classifies each clause into case elements “00 TV is”, “△△ Ms.”, “Winning” and a predicate “reported”. The morpheme / syntax analyzer 121 gives information on the word type (proprietary noun, sub-variant, etc.) and clause case to the case element, and gives information on the word type and sentence state to the predicate. . In the example shown in the figure, “00 TV is” is “ga / proper noun (company name)”, “△△ 's” is “no case / proprietary noun (person name)”, and “wins” Is “wo case / general noun”, and “reported” is “predicate / sa-variable / active”. In addition, the morpheme / syntax analyzer 121 analyzes dependency information between clauses and represents the information in a tree structure. In the example shown in the figure, “OO TV is” related to “reported”, “△△ Ms.” Is related to “winning”, and “winning” is related to “reported”.

  Next, the basic score assigning unit 122 and the score updating unit 123 assign a score to each phrase word. 7-9 is a figure for demonstrating the process which the summary word production | generation apparatus 1 which concerns on the 1st Embodiment of this invention provides a score.

  In the example shown in FIG. 7, the input document is “00 television has won three ratings”. That is, the input document does not indicate that the media provides information. First, as shown in FIG. 7 (A), the morpheme / syntax analyzing unit 121 determines that “00 TV is” in the input document is “ga rating / proper noun (company name)”, "Is a" wo case / common noun "and" acquired "is a" predicate / sa-variant / active ".

  Subsequently, as shown in FIG. 7B, the basic score assigning unit 122 assigns the basic score “1. ..” to the word “00 TV” of the clause “00 TV is” based on the basic score rule table. "0", a basic score of "0.8" is given to the word "view rating triple crown" of the sentence "O rating triple", and the word "acquisition" of the predicate phrase "acquired" A basic score of “0.0” is given.

  Subsequently, the score update unit 123 determines whether or not the predicate “acquired” is in the specific predicate list. Since the predicate “acquired” is not in the specific predicate list, the score update unit 123 sets all score updates that are values to be updated from the basic score to “0.0”. That is, the score update unit 123 does not update the score. As a result, the total score for each word remains the basic score. Therefore, the total score of the word “00 television” is “1.0”, the total score of the word “view rating triple crown” is “0.8”, and the total score of the word “acquired” is “0. 0 ". Finally, the summary word selection unit 125 selects the word “00 television” having the highest total score as the summary word of the input document “00 television has obtained three ratings”. That is, the summary word generation device 1 outputs “OO television” as a subject word as a summary word when the media does not indicate that information is provided.

  On the other hand, in the example shown in FIG. 8, the input document is “OO TV reported that △△ was elected”. That is, the input document indicates that the media provides information. First, as shown in FIG. 8 (A), the morpheme / syntax analyzing unit 121 determines that “OO TV is” in the input document is “ga rating / proper noun (company name)” and “ Analyzes that “is” is “wo case / common noun (personal name) + common noun”, and “reported” is “predicate / sa variable / active”.

  Subsequently, as shown in FIG. 8 (B), the basic score assigning unit 122 sets the basic score “1. "0", a basic score of "0.8" is given to the word "△△ Mr.'s winning" of the sentence of wo, and the word "reporting" of the predicate clause "reported" Is given a basic score of “0.0”.

  Subsequently, the score update unit 123 determines whether or not the predicate “reported” is in the specific predicate list. Since the predicate “reported” is in the specific predicate list, the score update unit 123 adds the score update “−0.5” to the basic score “1.0” of “00 television” based on the score update rule table. And update the total score to “0.5”, add the score update “+0.5” to the basic score “0.8” of the word “△△ Mr. The score is updated by adding “0.0” to the basic score “0.0” of the word “report” and the total score is updated to “0.0”.

  As a result, “OO TV”, which is the subject before the update, has the highest basic score, whereas “Winning of Δ △”, which is the object, has the highest total score after the update. Therefore, the summary word selection unit 125 selects the word “△△ Mr. Winning” with the highest total score as the summary word of the input document “OO TV has reported the winning of Mr. Δ △”. That is, the summary word generating device 1 outputs “winning of Mr. ΔΔ”, which is information reported by “00 television” as the media, as a summary word. Therefore, the summary word generation device 1 can generate and output a more appropriate summary word for the input document.

  In the example shown in FIG. 9, the input document is “Mr. A handed a letter to Mr. B. Mr. B introduced the letter”. First, as shown in FIG. 9A, the morpheme / syntax analyzing unit 121 sets “Mr. A” in the first sentence “Mr. A gave a letter to Mr. B” in the input document as “ "Proper noun (person name)", "Mr. B" is "second case / proper noun (person name)", "letter" is "wo case / common noun", "handed" is "predicate" / Verb / active ”is analyzed. Further, as shown in FIG. 9B, the morpheme / syntax analyzer 121 sets “Mr. B” in the second sentence “Mr. B introduced the letter” in the input document to “ (Person name) ”,“ the letter ”is“ wo case / indicative pronoun + common noun ”, and“ introduced ”is“ predicate / sa variable / active ”.

  Subsequently, as illustrated in FIG. 9C, the basic score assigning unit 122 assigns a basic score “to the word“ Mr. A ”of the phrase“ Mr. A ”in the first sentence based on the basic score rule table. 1.0 ”, the basic score“ 0.5 ”is assigned to the word“ Mr. B ”of the second sentence“ Mr. B ”, and the word“ letter ”of the sentence“ L A score “0.8” is assigned, and a basic score “0.0” is assigned to the word “pass” in the clause “passed” of the predicate. Moreover, since the predicate “passed” of the first sentence is not in the specific predicate list, the score update unit 123 sets all score updates of the first sentence to “0.0” and does not update the score of the first sentence. As a result, the total score for each word remains the basic score.

  Further, as shown in FIG. 9D, the basic score assigning unit 122 adds a basic score “to the word“ Mr. B ”of the phrase“ Mr. B ”in the second sentence based on the basic score rule table. 1.0 ”, a basic score“ 0.8 ”is assigned to the word“ that letter ”in the verse“ the letter ”, and a basic score is assigned to the word“ introduction ”in the clause“ introduced ”in the predicate. “0.0” is assigned.

  Subsequently, since the predicate “introduced” of the second sentence is in the specific predicate list, the score update unit 123 updates the score to the basic score “1.0” of “Mr. B” based on the score update rule table “ -0.5 "is added to update the total score to" 0.5 ", and the score update" +0.5 "is added to the basic score" 0.8 "of the word" that letter "to add the total score to" 1. 3 ”and the score update“ 0.0 ”is added to the basic score“ 0.0 ”of the word“ introduction ”to update the total score to“ 0.0 ”.

  Subsequently, as shown in FIG. 9E, the summary word integration unit 124 compares the first sentence and the second sentence, adds up the scores of the word “Mr. Is set to “1.0”, and the score of the word “letter” is added together to set the total score to “2.1”. Thereby, the score of the word which appears frequently in an input document becomes high. Finally, the summary word selection unit 125 selects the word “letter” with the highest total score as the summary word of the input document “Mr. A handed the letter to Mr. B. Mr. B introduced the letter”. As described above, when there are a plurality of sentences in the input document, the summary word generation device 1 summarizes important words that frequently appear in the input document in order to select a summary word by adding the scores of words regarded as the same. Can be selected as a word.

FIG. 10 is a flowchart showing an example of summary word generation processing executed by the summary word generation device 1 according to the first embodiment of the present invention.
In step S <b> 101, an input document is input to the input unit 11.
In step S103, the morpheme / syntax analyzer 121 analyzes the input document based on the analysis dictionary.

In step S105, the basic score giving unit 122 gives a score to each sentence element included in the input document based on the basic score rule table. A sentence element is a word of each clause.
In step S107, the score update unit 123 determines whether a specific predicate in the specific predicate list is used in a sentence included in the input document. When the specific predicate is used (step S107; YES), the summary word generation device 1 proceeds to the process of step S109. On the other hand, when the specific predicate is not used (step S107; NO), the summary word generation device 1 proceeds to the process of step S111.

  In step S109, the score update unit 123 updates the score of the sentence element of the sentence in which the specific predicate is used, based on the score update rule table.

In step S111, when there are a plurality of sentences in the input document, the summary word integration unit 124 compares the sentences and adds up the scores of similar words that can be regarded as the same.
In step S113, the summary word selection unit 125 selects the word with the highest score as the summary word.
In step S115, the output unit 14 outputs a summary word. Thereafter, the process ends.

  As described above, the summary word generation device 1 according to the first embodiment includes an input unit 11 (acquisition unit) that acquires a sentence including one or more sentences, and each word included in the document as a case element. In accordance with a predicate in the sentence, a score updating unit 123 that updates the weight of each word included in the sentence according to the predicate in the sentence, and a summary word of the document is selected based on the weight of each word. The summary word selection part 125 and the output part 14 which outputs the selected summary word are provided.

  With such a configuration, it is possible to extract an appropriate summary word even when the weight of a clause including the same particle changes depending on the predicate. That is, a more appropriate summary word can be extracted from the document.

  The summary word generation device 1 also includes a summary word integration unit 124 that adds up the weights of words that are regarded as the same. With such a configuration, it is possible to increase the weight of frequently appearing words in a document.

  The basic score assigning unit 122 increases the subject weight of the sentence when the sentence is active, increases the object weight of the sentence when the sentence is passive, and the score update unit 123 When a specific predicate is included in a sentence, the subject weight and object weight of the sentence are reversed. More specifically, the specific predicate is a verb meaning an act of providing information, and the score updating unit 123 weights the information provided in the sentence when the specific predicate is included in the sentence. Make it heavy. With such a configuration, in the case where the media or the like provides information, the provided information can be extracted as a summary word.

(Second Embodiment)
The second embodiment of the present invention will be described below with reference to the drawings.
Even in the case of sentences using the same predicate, case elements that are appropriate summary words may differ depending on the category of the input document. The category indicates the genre to which the sentence belongs (for example, performing arts, sports, politics, etc.). For example, in the sentence “Mr. OO announced marriage” whose category is “Entertainment”, the subject “Mr. OO” is appropriate as a summary word, whereas the sentence “〇 〇” whose category is “Politics” is suitable. In the "Prime Minister Announced Tax Reduction", the object "△△ Tax Reduction" is appropriate as a summary word. Therefore, this embodiment is different from the first embodiment in that the specific predicate differs according to the category of the input document.

  FIG. 11 is a schematic block diagram showing an example of a functional configuration of the summary word generation device 1A according to the second exemplary embodiment of the present invention. The summary word generation device 1A is an information processing device that generates and outputs a summary word of a document from an input document including one or a plurality of sentences. The summary word generation device 1A is a server device to which a plurality of terminal devices can be connected, for example, and outputs summary words of the input document to the connected terminal devices. Alternatively, the summary word generation device 1A may be incorporated in each terminal device such as a mobile phone, a smartphone, a tablet terminal, or a personal computer. Since the hardware configuration of the summary word generation device 1A is the same as that of the first embodiment, the description thereof is omitted. Hereinafter, the same components as those in the first embodiment are denoted by the same reference numerals, and the description thereof is omitted.

  The summary word generation device 1 </ b> A includes an input unit 11 </ b> A, a control unit 12 </ b> A, a storage unit 13 </ b> A, and an output unit 14. The input unit 11A is an interface for inputting an input document together with its category, such as a keyboard, a microphone, and a network interface. The input unit 11A outputs the inputted input document and its category to the control unit 12A.

  The control unit 12A controls the summary word generating device 1A. The control unit 12A includes a morpheme / syntactic analysis unit 121, a basic score assignment unit 122, a score update unit 123A, a summary word integration unit 124, and a summary word selection unit 125.

  The score update unit 123 </ b> A updates the word weight of each clause in each sentence according to the predicate of each sentence included in the input document with different rules according to the category of the input document. Specifically, the score updating unit 123A reads the specific predicate list corresponding to the category of the input document from the category specific predicate list storage unit 133A, and the specific predicate included in the read specific predicate list is a statement predicate. In this case, the score of the word of each clause of the sentence is updated based on the score update rule table stored in the score update rule storage unit 134. The score update unit 123A outputs the updated score and the input document to the summary word integration unit 124.

  The storage unit 13A includes an analysis dictionary storage unit 131, a basic score rule storage unit 132, a category specific predicate list storage unit 133A, and a score update rule storage unit 134, and stores various types of information. The category specific predicate list storage unit 133A stores a specific predicate list indicating a specific predicate for each category.

FIG. 12 is a schematic diagram illustrating a data configuration and a data example of the category specific predicate list table stored in the category specific predicate list storage unit 133A according to the second embodiment of the present invention.
As shown in the drawing, the category specific predicate list table includes items of a category and a specific predicate list. The category is a document category. The specific predicate list is a list of specific predicates.
In the illustrated example, the specific predicate list corresponding to the category “entertainment” includes predicates such as “report, post, etc.”. The specific predicate list corresponding to the category “sports” includes predicates such as “achieve, acquire,...”. The specific predicate list corresponding to the category “politics” includes predicates such as “publish, announce,...”. The specific predicate list of each category may include the same predicate in a plurality of categories.

  Next, the operation of the summary word generating device 1A will be described in detail using a specific example. FIG. 13 and FIG. 14 are diagrams for explaining processing in which the summary word generating device 1A according to the second embodiment of the present invention assigns a score.

  In the example shown in FIG. 13, the input document is “Mr. OO announced marriage”, and the category is entertainment. First, as shown in FIG. 13A, the morpheme / syntax analyzing unit 121 determines that “Mr. OO” in the input document is “ga case / proper noun (person name)” and “marriage” is “wo”. It is analyzed that it is “case / common noun” and “announced” is “predicate / sa variation / active”.

  Subsequently, as shown in FIG. 13B, the basic score assigning unit 122 assigns the basic score “1. 0 is assigned, the basic score “0.8” is assigned to the word “marriage” of the sentence “marriage”, and the basic score “0.0” is assigned to the word “announcement” of the clause “announced” of the predicate. Is given.

  Subsequently, the score update unit 123A determines whether or not the predicate “announced” is in the specific predicate list of the category “entertainment” of the input document. Since the predicate “announced” is not in the specific predicate list of the category “entertainment”, the score update unit 123A sets the score update of all words to “0.0” and does not update the score of each word. As a result, the total score for each word remains the basic score. Therefore, the summary word selection unit 125 selects the word “Mr. OO” having the highest total score as a summary word of the input document “Mr. OO announced marriage”.

  On the other hand, in the example shown in FIG. 14, the input document is “0.00 Prime Minister has announced a △△ tax cut”, and the category is politics. First, as shown in FIG. 14 (A), the morpheme / syntax analyzing unit 121 determines that “0.00 Prime Minister” in the input document is “Ga ’s case / proprietary noun (person name)”, and “ Is “wo case / common noun” and “announced” is “predicate / sa-variable / active”.

  Next, as shown in FIG. 14 (B), the basic score assigning unit 122 assigns the basic score “1. "0", a basic score of "0.8" is given to the word "△△ tax reduction" in the sentence "△△ tax reduction", and the word "announcement" of the predicate clause "announced" A basic score of “0.0” is given.

  Subsequently, the score updating unit 123A determines whether or not the predicate “announced” is in the specific predicate list of the category “politics” of the input document. Since the predicate “announced” is in the specific predicate list of the category “politics”, the score update unit 123A updates the score “−” to the basic score “1.0” of “00 prime” based on the score update rule table. 0.5 ”is added to update the total score to“ 0.5 ”, and the score“ +0.5 ”is added to the basic score“ 0.8 ”of the word“ △△ tax reduction ”to add the total score to“ 1 ” .3 ”, the score update“ 0.0 ”is added to the basic score“ 0.0 ”of the word“ announcement ”, and the total score is updated to“ 0.0 ”.

  As a result, the basic score “OO Prime Minister” was the highest before the update, whereas the target “△△ Tax Reduction” was the highest total score after the update. Therefore, the summary word selection unit 125 selects the word “ΔΔ tax reduction” having the highest total score as the summary word of the input document “0.00 Prime Minister has announced ΔΔ tax reduction”.

  As described above, since the summary word generation device 1A uses different specific predicate lists depending on the category of the input document, even if the same predicate “present” is used, a summary word suitable for each input document is selected. Can be generated and output.

FIG. 15 is a flowchart showing an example of summary word generation processing executed by the summary word generation device 1A according to the second embodiment of the present invention.
In step S201, the input document and its category are input to the input unit 11A.
Since the process of step S203-step S205 is the same as the process of step S103-step S105 in 1st Embodiment, the description is abbreviate | omitted.

  In step S207, the score update unit 123A determines whether or not a specific predicate in the specific predicate list corresponding to the category of the input document is used in a sentence included in the input document. When the specific predicate is used (step S207; YES), the summary word generation device 1A proceeds to the process of step S209. On the other hand, when the specific predicate is not used (step S207; NO), the summary word generation device 1A proceeds to the process of step S211.

  Since the process of step S209-step S215 is the same as the process of step S109-step S115 in 1st Embodiment, the description is abbreviate | omitted.

  In this embodiment, the specific predicate list is made different depending on the category of the input document. However, the present invention is not limited to this, and the score may be updated using different score update rules depending on the category of the input document.

  As described above, the summary word generation device 1A according to the second embodiment includes an input unit 11 (acquisition unit) that acquires a document including one or a plurality of sentences, and each word included in the sentence as a case element. In accordance with the predicate in the sentence, the score updating unit 123A for updating the weight of each word included in the sentence according to the predicate in the sentence, and the summary word of the document are selected based on the weight of each word. The summary word selection part 125 and the output part 14 which outputs the selected summary word are provided. The score update unit 123A updates the weight with different rules according to the document category. More specifically, the specific predicate for updating the weight differs depending on the document category.

  With such a configuration, it is possible to extract an appropriate summary word even when the weight of a clause including the same particle changes depending on the predicate. Moreover, since the weight is updated according to the category of the document even if the sentence uses the same predicate, summary words suitable for the document can be extracted. That is, a more appropriate summary word can be extracted from the document.

  The embodiments of the present invention have been described in detail above with reference to the drawings. However, the specific configuration is not limited to the above-described one, and various design changes and the like can be made without departing from the scope of the present invention. Is possible.

  For example, in the above-described embodiment, the summary word generation device 1 or 1A updates the score with the same score update rule when the sentence predicate is a specific predicate. You may update a score by a different score update rule.

  In the above-described embodiment, the summary word generation devices 1 and 1A output one word as a summary word. However, the present invention is not limited to this, and a predetermined plurality of words may be output in descending order of score.

  In the above-described embodiment, the case where the input document is in Japanese is shown as an example. However, the present invention is not limited to this, and another language may be used.

  The summary word generation devices 1 and 1A may receive an input of evaluation for the output summary word and change the score update rule by machine learning based on the evaluation.

  In addition, the summary word generation devices 1 and 1A are often more important as the previous sentence, so if there are multiple sentences in the input document, the score is given so that the score of the previous sentence is higher. May be.

  Further, in the above-described embodiment, in the score update rule, in the case of active, the ga rating and the second rating are deducted and the wo rating is added, but not limited to this, other de ratings, Any case element may be used as long as it deducts a case element that is likely to include a media name such as a color case and adds a case element that is likely to include information conveyed by the media such as a case rating or a nickel case. Similarly, in the score update rule, in the case of passive voice, ga rating is added and 2 rating is deducted, but not limited to this, media names such as de-rated, nyol and kara are also included. As long as the case elements that are likely to include information conveyed by the media, such as the case rating and the Nikku case, are added.

  Note that one or more of the above-described embodiments and modifications, and a part or all of the above-described embodiments and modifications may be combined to realize one aspect of the present invention.

  Note that one program that operates in the summary word generation devices 1 and 1A according to one aspect of the present invention is provided so as to realize the functions described in the above embodiments and modifications according to one aspect of the present invention, or A plurality of programs (programs that cause a computer to function) that control a processor such as a CPU (Central Processing Unit) may be used. Information handled by each of these devices is temporarily stored in a RAM (Random Access Memory) at the time of processing, and then stored in various storages such as a flash memory and an HDD (Hard Disk Drive). Then, it may be read out, corrected and written by a CPU or the like.

  Note that a part or all of the summary word generation devices 1 and 1A in the above-described embodiments and modifications may be realized by a computer including one or a plurality of processors. In that case, the program for realizing the control function may be recorded on a computer-readable recording medium, and the program recorded on the recording medium may be read by the computer system and executed.

  Here, the “computer system” is a computer system built in the summary word generation devices 1 and 1A and includes an OS and hardware such as peripheral devices. The “computer-readable recording medium” refers to a storage device such as a flexible medium, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system.

  Further, the “computer-readable recording medium” is a medium that dynamically holds a program for a short time, such as a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line, In such a case, a volatile memory inside a computer system serving as a server or a client may be included and a program that holds a program for a certain period of time. The program may be a program for realizing a part of the functions described above, and may be a program capable of realizing the functions described above in combination with a program already recorded in a computer system.

  In addition, a part or all of the summary word generation devices 1 and 1A in the above-described embodiments and modifications may be realized as an LSI that is typically an integrated circuit, or may be realized as a chip set. . In addition, each functional block of the summary word generation device 1 or 1A in each of the embodiments and modifications described above may be individually chipped, or a part or all of them may be integrated into a chip. Further, the method of circuit integration is not limited to LSI's, and implementation using dedicated circuitry and / or general purpose processors is also possible. In addition, when an integrated circuit technology that replaces LSI appears due to progress in semiconductor technology, an integrated circuit based on the technology can also be used.

  As described above, the embodiments and modifications as one aspect of the present invention have been described in detail with reference to the drawings. However, specific configurations are not limited to the embodiments and modifications, and depart from the gist of the present invention. This includes design changes that do not occur. In addition, one aspect of the present invention can be modified in various ways within the scope of the claims, and the technical aspects of the present invention also relate to embodiments obtained by appropriately combining technical means disclosed in different embodiments. Included in the range. Moreover, it is the element described in said each embodiment and modification, and the structure which substituted the element which has the same effect is also contained.

1, 1A summary word generation device 11, 11A input unit 12, 12A control unit 121 morpheme / syntax analysis unit 122 basic score assignment unit 123, 123A score update unit 124 summary word integration unit 125 summary word selection unit 13, 13A storage unit 131 Analysis dictionary storage unit 132 Basic score rule storage unit 133 Specific predicate list storage unit 133A Category specific predicate list storage unit 134 Score update rule storage unit 14 Output unit 101 CPU
102 Storage medium interface unit 103 Storage medium 104 Input device 105 Output device 106 ROM
107 RAM
108 Auxiliary storage unit 109 Network interface unit

Claims (9)

An acquisition unit for acquiring a document composed of one or more sentences;
A basic score giving unit that weights each word included in the document according to its case element;
A score updating unit that updates the weight of each word included in the sentence according to the predicate in the sentence;
A summary word selector for selecting summary words of the document based on the weight of each word;
An output unit for outputting the selected summary word;
An information processing apparatus comprising: A summary word integration unit that adds together the weights of words that are considered identical;
The information processing apparatus according to claim 1. The score update unit updates weights according to different rules according to the category of the document.
The information processing apparatus according to claim 1 or 2. The basic score assigning unit increases the weight of the subject of the sentence when the sentence is active, and increases the weight of the object of the sentence when the sentence is passive.
The information processing apparatus according to any one of claims 1 to 3. The score update unit, when a specific predicate is included in a sentence, reverses the subject weight and object weight of the sentence,
The information processing apparatus according to claim 4. The specific predicate is a verb meaning an act of providing information,
The score update unit increases the weight of information provided in the sentence when the specific predicate is included in the sentence.
The information processing apparatus according to claim 5. The information processing apparatus according to claim 5, wherein the specific predicate differs according to a category of the document. A process in which the acquisition unit acquires a document including one or more sentences;
A process in which the basic score assigning unit weights each word included in the document according to the case element;
A process in which the score updating unit updates the weight of each word included in the sentence according to the predicate in the sentence;
A summary word selection unit selecting a summary word of the document based on the weight of each word;
A process in which the output unit outputs the selected summary word;
An information processing method comprising: Computer
Obtaining a document comprising one or more sentences;
Weighting each word contained in the document according to its case element;
Updating the weight of each word included in the sentence according to the predicate in the sentence;
Selecting summary words of the document based on the weight of each word;
Outputting the selected summary word;
A program for running.

Images