JP2019057034A - Learning apparatus, generation device, learning method, generation method, learning program, generation program, and model - Google Patents

Abstract

To provide a learning apparatus capable of generating a precise text.SOLUTION: Disclosed a learning apparatus comprises an acquisition part and a learning part. The acquisition part is configured to acquire a piece of information relevant to a first text and a second text which is a text corresponding to the first text and satisfies predetermined conditions. The learning part is configured to learn a first model for generating an output text satisfying predetermined conditions from the information relevant to an input text on the basis of a piece of information relevant to a piece of first relationship information representing a relationship among multiple words included in the first text and the second text.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a learning device, a generation device, a learning method, a generation method, a learning program, a generation program, and a model.

  Techniques for generating text on a device are known. For example, a technique for causing a device to generate a summary sentence of a document is known. By causing the device to generate text, it is possible to save the person from generating text.

JP 2005-92616 A

  In recent years, machine learning has attracted attention. The generation of the text can be realized, for example, by using a learning model (hereinafter simply referred to as a model) obtained by learning a plurality of sets of text as learning data. However, text generated using a model may be less accurate (eg, accurate). For example, the text generated using the model is the document that means "the police arrested the criminal", but the output document is "the police criminal arrested" even though the original document was a document that meant "the police arrested the criminal" In some cases, it is clearly unnatural.

  The present application has been made in view of the above, and an object thereof is to enable generation of highly accurate text.

  The learning device according to the present application includes an acquisition unit that acquires information about the first text, and a second text that corresponds to the first text and satisfies a predetermined condition, and is included in the first text Learning a first model for generating an output text satisfying a predetermined condition from information on the input text based on the information on the first relation information indicating the relation between the words of the plurality of words and the second text. And a learning unit for performing.

  According to one aspect of the embodiment, it is possible to generate text with high accuracy.

FIG. 1 is a diagram illustrating an example of processing executed by the information processing apparatus according to the embodiment. FIG. 2 is a diagram illustrating a terminal device on which content is displayed. FIG. 3 is a diagram illustrating an example of learning data registered in the learning data database. FIG. 4 is a diagram illustrating a state in which the information processing apparatus generates related information. FIG. 5 is a diagram illustrating an example of a model. FIG. 6 is a diagram illustrating another example of processing executed by the information processing apparatus according to the embodiment. FIG. 7 is a diagram illustrating an example of a model. FIG. 8 is a diagram illustrating an example of a model. FIG. 9 is a diagram illustrating another example of processing executed by the information processing apparatus according to the embodiment. FIG. 10 is a diagram illustrating an example of a model. FIG. 11 is a diagram illustrating a configuration example of the information processing apparatus according to the embodiment. FIG. 12 is a diagram illustrating an example of information registered in the learning data database. FIG. 13 is a diagram illustrating an example of information registered in the model database. FIG. 14 is a diagram illustrating an example of information registered in the content information database. FIG. 15 is a flowchart illustrating an example of the learning process. FIG. 16 is a flowchart illustrating an example of the generation process. FIG. 17 is a hardware configuration diagram illustrating an example of a computer that implements the functions of the information processing apparatus.

  Hereinafter, a learning device, a generation device, a learning method, a generation method, a learning program, a generation program, and a form for implementing a model (hereinafter referred to as “embodiment”) according to the present application will be described in detail with reference to the drawings. explain. Note that the learning device, the generation device, the learning method, the generation method, the learning program, the generation program, and the model according to the present application are not limited to the following embodiments. In the following embodiments, the same portions are denoted by the same reference numerals, and redundant description is omitted.

[1-1. Example of information processing apparatus]
First, an example of a learning process and a generation process executed by an information processing apparatus that is an example of a generation apparatus and a learning apparatus will be described. FIG. 1 is a diagram illustrating an example of processing executed by the information processing apparatus according to the embodiment. The information processing apparatus 10 can communicate with a data server 20 and a terminal device 30 used by a predetermined client via a predetermined network such as the Internet.

  The information processing apparatus 10 is an apparatus that executes learning processing and generation processing described later. The information processing device 10 is realized by a server device, a cloud system, or the like.

  The data server 20 is an information processing apparatus that manages learning data used when the information processing apparatus 10 executes learning processing described later, and distribution content output when the information processing apparatus 10 executes generation processing described later. . The data server 20 is realized by a server device, a cloud system, or the like. For example, the data server 20 executes a distribution service that distributes news and various contents posted by the user to the terminal device 30. Such a distribution service is realized by, for example, various news distribution sites or SNS (Social Networking Service).

  The terminal device 30 is a smart device such as a smartphone or a tablet, and is a mobile terminal device that can communicate with an arbitrary server device via a wireless communication network such as 3G (3rd Generation) or LTE (Long Term Evolution). is there. The terminal device 30 may be not only a smart device but also an information processing device such as a desktop PC (Personal Computer) or a notebook PC.

[1-2. About distribution of summary content)
Here, when there are a plurality of distribution contents to be distributed, the data server 20 does not distribute all of the contents, but sends the summary contents including the text that is a summary of each content to the terminal device 30. May be delivered. The distribution content is, for example, a news article acquired from an information medium or an information distribution entity. As an example, the summary content is a top page of a portal site with a link to distribution content.

  The information medium is information indicating an information distribution medium or a posting medium, for example, information indicating “newspaper”, “magazine”, or the like. The distribution subject is an organization or an individual that distributes information such as news articles. For example, the distribution subject is a mass media such as a newspaper company, a publisher, a broadcasting station (a television station, a radio station), or the like. Of course, the distribution subject is not limited to mass media. The distribution subject may be a portal site operator or a mobile phone company. The distribution subject may be the operator of the information processing apparatus 10 itself.

  FIG. 2 is a diagram illustrating the terminal device 30 on which content is displayed. A state J1 in FIG. 2 is a diagram illustrating a state in which a top page of the portal site is displayed, and a state J2 in FIG. 2 is a diagram illustrating a state in which a page including the distribution content C11 is displayed. The distribution content C11 is, for example, a news article acquired by an operator of the information processing apparatus 10 (for example, an administrator of a portal site) from a predetermined distribution subject. The terminal device 30 shown in the state J1 displays a page on which a plurality of tiles with links to articles are arranged. In each tile, text indicating the content of the distribution content (hereinafter referred to as a headline) is displayed. For example, the heading U11 is displayed on the top tile of the plurality of tiles arranged in a vertical row. The heading U11 is text corresponding to the content of the distribution content C11. When the user taps the tile on which the heading U11 is displayed, a screen including the distribution content C11 is displayed on the terminal device 30. The distribution content C11 includes an article body K11 and an article title T11.

  Text associated with the distribution content (for example, a headline posted on the top page of the portal site) is created by a person based on the content of the distribution content. For example, the headline U11 is created by reading and summarizing the title T11 included in the distribution content C11 by an employee of the portal site operator. However, it is troublesome to manually generate a headline for each distribution content.

  Thus, it is conceivable to automatically generate a headline from the distribution content using a learning model (hereinafter simply referred to as a model) in which features of various information are learned. For example, it is conceivable to automatically generate a title as a headline from text (for example, a title) included in the distribution content. However, text generated using a model may be less accurate (eg, accurate). For example, text generated using a model may be text that is correct as a sentence but does not correctly represent the contents of the original document. For example, the text generated using the model is the document that means "the police arrested the criminal", but the output document is "the police criminal arrested" even though the original document was a document that meant "the police arrested the criminal" In some cases, it is clearly unnatural.

[1-3. Example of processing using relationship information)
[1-3-1. About the learning process)
Therefore, the information processing apparatus 10 learns a model for generating a headline text from text (for example, a title) included in the distribution content by executing the following learning process. Hereinafter, an example of the learning process executed by the information processing apparatus 10 will be described with reference to FIG.

  First, the information processing apparatus 10 acquires information serving as learning data from the data server 20 (step S11). For example, the information processing apparatus 10 acquires the distribution content distributed to the user in the past from the data server 20. Then, the information processing apparatus 10 registers the information acquired from the data server 20 in the learning data database.

  FIG. 3 is a diagram illustrating an example of learning data registered in the learning data database. In the learning data database, information associating the text, title, and heading is registered as learning data. A learning data ID (Identifier) is assigned to each set of a text, a title, and a heading. “Body” and “Title” are texts included in the distribution content.

  “Heading” is text for the user to easily grasp the contents of the distribution content. The headline is text corresponding to text (for example, a body or a title) included in the distribution content. A headline serving as learning data is created by a predetermined person (for example, an employee of a portal site operator) reading and summarizing text included in the distribution content. The headline is generated so as to satisfy a predetermined condition. For example, the heading is generated so as to be a predetermined number of characters or less. As an example, headings are generated to be 13.5 characters or less. At this time, 13.5 characters is the number of characters with one double-byte character as one character. One half-width character is 0.5 character. In the following description, the predetermined condition is referred to as “generation condition”. The generation conditions are not limited to 13.5 characters or less. The number of characters as a generation condition may be more or less than 13.5 characters. Of course, the generation condition is not limited to the number of characters.

  “Heading” can be rephrased as a summary, title, abbreviated title, topic, topic heading, or the like. Similarly, the “title” included in the distribution content can also be called a summary or the like. In the present embodiment, it is assumed that the “headline” is a shortened text created based on the “title” included in the distribution content. Of course, the “headline” may be a shortened text created based on the text. Note that even if the sentence is shortened syntactically, the number of characters of the “heading” may be larger than the number of characters of the “title”. For example, when an abbreviation included in the “title” is replaced with a formal name, the number of characters may increase as a result even if the sentence is shortened syntactically. In such a case, “heading” is a shortened text of “title”.

  Returning to FIG. 1, the information processing apparatus 10 acquires a plurality of sets of titles and headings from the data registered in the learning data database. Then, the information processing apparatus 10 learns the model M1 (first model) based on the title and the headline. The model M1 is a model that generates text satisfying a generation condition from data input to the model.

  As described above, text generated using a model may be less accurate (eg, accurate). For example, the text generated using the model is the document that means "the police arrested the criminal", but the output document is "the police criminal arrested" even though the original document was a document that meant "the police arrested the criminal" In some cases, it is clearly unnatural. Therefore, in this embodiment, the information processing apparatus 10 does not use the title as input data, but learns the model M using information related to the first text as input data. More specifically, the information processing apparatus 10 learns the model M using the relationship information of the first text as input data. The relationship information is information indicating a relationship between words of a plurality of words included in the text. For example, the relationship information is information indicating a dependency structure of a plurality of words included in the text (hereinafter referred to as dependency information). In the following description, a text for which relation information is generated during learning may be referred to as a “first text”. The first text is not limited to the title. The first text may be a body text.

  The information processing apparatus 10 generates relation information of the first text (Step S12). FIG. 4 is a diagram illustrating a state in which the information processing apparatus 10 generates related information. For example, assume that the title that is the first text is “Actor A has announced that his wife is pregnant”. In this case, the information processing apparatus 10 performs syntax analysis (for example, dependency analysis) on the first text, and generates dependency information as illustrated in the middle of FIG. 4, for example. The information processing apparatus 10 may generate dependency information by performing morphological analysis on the first text. Then, the information processing apparatus 10 acquires an analysis result (for example, dependency information) as related information. The relation information in the middle part of FIG. 4 is made visible by making the dependency structure of words and words into a tree structure.

  Note that the information processing apparatus 10 may acquire information indicating a predicate structure of a word included in the first text (hereinafter, referred to as predicate structure information) as related information. This is information including a case analysis result of a word included in the first text. For example, in the example of FIG. 4, the “announced” predicate structure information included in the first text is, for example, “<prescription: behavior> <case analysis result: ga / A; wo / thing; 2 / −; / −; De / −; time / −> ”. In this example, “announced” is a predicate (verb), indicating that “A” with the case particle “ga” and “thing” with the case particle “wo” are involved. ing.

  Further, the information processing apparatus 10 may use information obtained by shortening the relationship information of the first text (for example, a title) as the relationship information. For example, in the example of FIG. 4, the information processing apparatus 10 may use, as new relation information, the tree structure information illustrated in the lower part of FIG. 4 from which a part of the branch of the relation information represented by the tree structure is removed. . In the following description, the relationship information obtained by shortening the relationship information may be referred to as shortened relationship information. The information processing apparatus 10 may shorten the related information so that the shortened related information corresponds to a shortened text (for example, a headline) scheduled to be generated.

  The information processing apparatus 10 performs learning (for example, supervised learning) of the model M1 using the relationship information of the first text (in the case of the present embodiment, title) as input data and the headline as correct answer data (step S13). In the following description, a text (headline in the case of the present embodiment) that becomes correct answer data during learning may be referred to as a “second text”. In addition, when generating text, text output from the model (that is, text that satisfies the generation conditions) is referred to as “output text”, and text that should be the generation source of the second text is referred to as “input text”. Sometimes. In the following description, the correct answer data may be referred to as a teacher label. The input text is text that is input directly or indirectly to the model. The input text may not be entered directly into the model. For example, the input text may be converted into relation information and input to the model.

  FIG. 5 is a diagram illustrating an example of the model M1. The model M1 is, for example, a Sequence to Sequence Model (hereinafter also referred to as Seq2Seq) using LSTM (Long Short Term Memory) which is an example of RNN (Recurrent Neural Network). The model M1 may be Seq2Seq (Sequence to Sequence with Attention Model) to which an attention model is added. Seq2Seq is a kind of encoder-decoder model, and allows data strings (Sequence) of different lengths to be input and word strings (Sequence) of different lengths to be output.

The model M1 illustrated in FIG. 5 generates an encoder E1 that generates feature information based on relation information of an input text (for example, title), and an output text (for example, a headline) based on the feature information output by the encoder E1. And a decoder D1.

  The encoder E1 extracts features of the input text. The encoder E1 has an input layer and an intermediate layer (hidden layer). The encoder E1 receives the first text related information. The encoder E1 outputs feature information that is a multidimensional quantity (for example, a vector) indicating the feature of the input data. The feature information may be called an attention vector when the model M1 is an attention model.

  The decoder D1 outputs a text (that is, a headline) in which the input text is shortened. The decoder D1 has an intermediate layer (hidden layer) and an output layer. The feature information output from the encoder E1 is input to the decoder D1. When the feature information is input, the decoder D1 sequentially outputs words that become abbreviated text from the output layer.

  When the information related to the first text is input to the encoder E1, the information processing apparatus 10 learns the model M1 so that the second text corresponding to the first text is output from the decoder D1. For example, the information processing apparatus 10 uses a weight (that is, a value considered when a value is transmitted between the nodes by a method such as back propagation so that the shortened text output from the decoder D1 approaches the second text). , Correct the connection coefficient). As a result, the information processing apparatus 10 causes the model M1 to learn the characteristics of the first text related information. Note that the information processing apparatus 10 may correct the weight value based on the cosine similarity between a vector that is a distributed representation of the shortened text and a vector that is a distributed representation of the second learning text.

  The information processing apparatus 10 outputs the second text corresponding to the first text from the decoder D1 when the shortened relation information obtained by shortening the relation information of the first text is input to the encoder E1. The model M1 may be learned. Also in this case, the information processing apparatus 10 uses the weight that is considered when the value is transmitted between the nodes by a method such as backpropagation so that the shortened text output from the decoder D1 approaches the second text. The value of (that is, the connection coefficient) may be corrected.

[1-3-2. About generation processing)
Next, an example of a generation process that generates text that corresponds to the input text and satisfies the generation conditions using the model M1 that has been learned by the above-described learning process will be described. At this time, the output text is, for example, a shortened text of the input text.

  First, the information processing apparatus 10 acquires information for generating relation information input from the data server 20 to the model M1 (step S14). For example, the information processing apparatus 10 acquires distribution contents (text and title) scheduled to be distributed from the data server 20. Then, the information processing apparatus 10 acquires a title included in the acquired distribution content as input text for generating output text.

  Thereafter, the information processing apparatus 10 acquires related information on the input text as information on the input text (step S15). At this time, the information processing apparatus 10 may generate relation information (shortening relation information) corresponding to the output text (for example, a headline) by shortening the relation information of the input text.

  Then, the information processing apparatus 10 acquires the model M1 from the internal storage device. As described above, the model M1 is a model learned by using the relationship information of the first text as input data and the heading as correct data (teacher label). And the information processing apparatus 10 produces | generates an output text by inputting the relationship information of an input text into the model M1 (step S16). The information processing apparatus 10 may generate a text having a predetermined number of characters or less as the output text.

  Note that the model M1 may be a model learned by using the shortening relation information of the first text as input data and the second text as correct answer data (teacher label). In this case, the information processing apparatus 10 generates output text by inputting the shortening relation information of the input text to the first model.

  When the output text is generated, the information processing apparatus 10 generates summary content using the output text. Then, the information processing device 10 delivers the summary content to the terminal device 30 (step S17).

  According to this embodiment, the model M1 is configured not to use the title itself as input data, but to use title relation information as input data. The relationship information is not data representing a sentence as a simple character string, but data indicating the syntax of the sentence. For this reason, the information processing apparatus 10 can reduce the generation of a headline with an unnatural syntax that has a different meaning from the original document, such as “the police criminal is arrested”. As a result, the information processing apparatus 10 can generate a headline with high accuracy.

[1-4. Other examples of processing using relationship information]
[1-4-1. About the learning process)
Note that the model used for generating the output text may be configured so that the input text can be input in addition to the related information. FIG. 6 is a diagram illustrating another example of the process executed by the information processing apparatus 10 according to the embodiment. First, the learning process will be described.

  First, the information processing apparatus 10 acquires information serving as learning data from the data server 20 (step S21). For example, the information processing apparatus 10 acquires the distribution content distributed to the user in the past from the data server 20. Then, the information processing apparatus 10 acquires the title included in the learning data as the first text and generates relation information of the first text (step S22).

  The information processing apparatus 10 learns the model M2 (second model) using the generated relation information as correct answer data (teacher label) (step S23). The model M2 is a model that generates input text relationship information from the input text.

  FIG. 7 is a diagram illustrating an example of the model M2. The model M2 is, for example, Seq2Seq using LSTM which is an example of RNN. The model M2 may be Seq2Seq to which an attention model is added. The model M2 shown in FIG. 7 includes an encoder E2 that generates feature information based on input text (for example, a title), and a decoder D2 that generates relation information of input text based on the feature information output from the encoder E2. Prepare. The encoder E2 extracts features included in the input text.

  The encoder E2 extracts features included in the input text. The encoder E2 has an input layer and an intermediate layer (hidden layer). The first text is input to the encoder E2. The encoder E2 outputs feature information that is a multidimensional quantity (for example, a vector) indicating the feature of the input text.

  The decoder D2 outputs the relation information of the input text. The decoder D2 has an intermediate layer (hidden layer) and an output layer. The feature information output from the encoder E2 is input to the decoder D2. When the feature information is input, the decoder D2 outputs data as relation information from the output layer.

  The information processing apparatus 10 learns the model M2 so that when the first text is input to the encoder E2, the relationship information of the first text is output from the decoder D2. For example, the information processing apparatus 10 uses a weight that is considered when a value is transmitted between nodes by a method such as backpropagation so that the data output from the decoder D2 approaches the relation information of the first text. (Ie, the connection coefficient) value is corrected. As a result, the information processing apparatus 10 causes the model M2 to learn the characteristics of the first text.

  Subsequently, the information processing apparatus 10 inputs the relation information between the first text (in the case of this embodiment, title) and the first text as input data, and the second text (in the case of this embodiment) as correct answer data. As a result, the model M3 (first model) is learned (step S24).

  FIG. 8 is a diagram illustrating an example of the model M3. The model M3 is, for example, Seq2Seq using LSTM which is an example of RNN. The model M3 may be Seq2Seq to which an attention model is added. A model M3 shown in FIG. 8 includes an encoder E31 that generates first feature information based on input text (for example, a title), an encoder E32 that generates second feature information based on relationship information of the input text, A synthesis model SM1 that synthesizes the first feature information and the second feature information to generate synthesis information, and a decoder D3 that generates an output text (heading) based on the synthesis information.

  The encoder E31 extracts features of the input text. The encoder E31 has an input layer and an intermediate layer (hidden layer). In the encoder E31, words included in the first text (in this embodiment, a title) are sequentially input. The encoder E31 outputs first feature information which is a multidimensional quantity (for example, a vector) indicating the feature of the input text.

  The encoder E32 extracts a feature included in the related information. The encoder E32 has an input layer and an intermediate layer (hidden layer). The encoder E32 receives the first text related information. The encoder E32 outputs second feature information that is a multidimensional quantity (for example, a vector) indicating the feature of the input relation information.

  The synthesis model SM1 outputs a linear combination of the first feature information output from the encoder E31 and the second feature information output from the encoder E32 as synthesis information. Note that the synthesis model SM1 may generate synthesis information in which a predetermined weight is applied to each feature information. Note that the model M3 may be configured to convolve synthesis information output from the synthesis model SM1 up to the number of dimensions corresponding to the input layer of the decoder D3.

  The decoder D3 outputs a text (that is, a headline) obtained by shortening the input text. The decoder D3 has an intermediate layer (hidden layer) and an output layer. The decoder D3 receives synthesis information (that is, first feature information and second feature information) output from the synthesis model SM1. When the synthesis information is input, the decoder D3 sequentially outputs words that become abbreviated text from the output layer.

  When the information processing apparatus 10 inputs the first text to the encoder E31 and the first text relation information to the encoder E32, the decoder D3 outputs the second text corresponding to the first text. The model M3 is learned. For example, the information processing apparatus 10 uses a weight (that is, a weight that is taken into consideration when a value is transmitted between nodes by a method such as back propagation so that the shortened text output from the decoder D3 approaches the second text). , Correct the connection coefficient). As a result, the information processing apparatus 10 causes the model M3 to learn the characteristics of the first text.

[1-4-2. About generation processing)
Next, an example of a generation process that generates an output text using the model M2 and the model M3 learned by the learning process described above will be described.

  First, the information processing apparatus 10 acquires information input from the data server 20 to the model M2 and the model M3 (step S25). For example, the information processing apparatus 10 acquires distribution contents (text and title) scheduled to be distributed from the data server 20. Then, the information processing apparatus 10 acquires a title included in the acquired distribution content as an input text.

  Thereafter, the information processing apparatus 10 acquires the model M2 from the internal storage device. As described above, the model M1 is a model learned using the first text as input data and the relation information of the first text as correct data (teacher label). And the information processing apparatus 10 produces | generates the relationship information of an input text by inputting an input text into the model M2 (step S26).

  Then, the information processing apparatus 10 acquires the model M3 from the internal storage device. As described above, the model M3 is a model learned by using the relation information between the first text and the first text as input data and the second text as correct data (teacher label). And the information processing apparatus 10 produces | generates an output text by inputting the relationship information of an input text into the model M3 (step S27).

  When the output text is generated, the information processing apparatus 10 generates summary content using the output text. Then, the information processing device 10 distributes the summary content to the terminal device 30 (step S28).

  According to the present embodiment, the information processing apparatus 10 is configured to generate a headline based on title relation information in addition to a title. The relationship information is not data representing a sentence as a simple character string, but data indicating the syntax of the sentence. For this reason, the information processing apparatus 10 can reduce the generation of a headline whose syntax is unnatural, such as “the police criminal arrested”, which has a different meaning from the original document. As a result, the information processing apparatus 10 can generate a headline with high accuracy.

[1-5. (Comparison using relationship information)
[1-5-1. About the learning process)
In the above-described embodiment, the relationship information is used as input data to the model. However, the relationship information does not necessarily have to be used as an input to the model. FIG. 9 is a diagram illustrating another example of processing executed by the information processing apparatus 10 according to the embodiment. First, the learning process will be described.

  First, the information processing apparatus 10 acquires information serving as learning data from the data server 20 (step S31). For example, the information processing apparatus 10 acquires the distribution content distributed to the user in the past from the data server 20. Then, the information processing apparatus 10 acquires the title included in the learning data as the first text.

  The information processing apparatus 10 learns the model M4 using the first text (in the case of the present embodiment) as input data and the second text (in the case of the present embodiment) as correct data (teacher label). (Step S32).

  FIG. 10 is a diagram illustrating an example of the model M4. The model M4 is, for example, Seq2Seq using LSTM which is an example of RNN. The model M4 may be Seq2Seq to which an attention model is added. A model M4 shown in FIG. 10 includes an encoder E4 that generates feature information based on input text (for example, a title), a decoder D4 that generates output text (heading) based on the feature information output from the encoder E4, Is provided.

  The encoder E4 extracts features included in the input text. The encoder E4 has an input layer and an intermediate layer (hidden layer). In the encoder E4, words included in the first text (in the present embodiment, a title) are sequentially input. The encoder E4 outputs feature information that is a multidimensional quantity (for example, a vector) indicating the feature of the input text.

  The decoder D4 outputs a text (that is, a heading) obtained by shortening the input text. The decoder D4 has an intermediate layer (hidden layer) and an output layer. The feature information output from the encoder E4 is input to the decoder D4. When the feature information is input, the decoder D4 sequentially outputs words that become abbreviated text from the output layer.

  The information processing apparatus 10 learns the model M4 so that when the first text is input to the encoder E4, the second text corresponding to the first text is output from the decoder D4. For example, the information processing apparatus 10 uses a weight (that is, a value that is considered when a value is transmitted between nodes by a method such as backpropagation so that the shortened text output from the decoder D4 approaches the second text). , Correct the connection coefficient). As a result, the information processing apparatus 10 causes the model M4 to learn the characteristics of the first text.

  Note that the information processing apparatus 10 can also use the relationship information to correct the weight value. For example, the information processing apparatus 10 creates relation information of the first text in advance. The information processing apparatus 10 creates the short text related information each time the short text (the heading text) is output from the decoder D4. Then, the information processing apparatus 10 calculates the similarity between the first text relation information and the shortened text relation information. At this time, the information processing apparatus 10 may calculate a cosine similarity between a vector that is a distributed representation of the relationship information of the first text and a vector that is a distributed representation of the relationship information of the shortened text. Of course, the similarity is not limited to the cosine similarity.

  The relationship information is not data representing a sentence as a simple character string, but data indicating the syntax of the sentence. For this reason, if the syntax of a sentence differs, the similarity will become low. For example, suppose the first text is “The police arrested the criminal on 1 October”. Then, it is assumed that “Police arrested by criminal” is output as a shortened text from model M4. In this example, the subject relating to “Arrest” is “Police” in the first text, whereas the subject relating to “Arrest” is “criminal” in the short text. Both have greatly different dependency structures. Therefore, in this example, the similarity between the relationship information of the first text and the relationship information of the shortened text is low. The information processing apparatus 10 corrects the weight value of the model M4 so that the degree of similarity increases.

  Further, the information processing apparatus 10 may correct the weight value of the model M4 based on the similarity between the relationship information of the second text and the relationship information of the shortened text. For example, the information processing apparatus 10 creates the related information of the second text in advance. The information processing apparatus 10 creates the short text related information each time the short text (the heading text) is output from the decoder D4. Then, the information processing apparatus 10 calculates the similarity between the relationship information of the second text and the relationship information of the shortened text. For example, suppose that the second text is “Police arrests criminal”. Then, it is assumed that “Police arrested by criminal” is output as a shortened text from model M4. In this example as well, in the second text, the subject relating to “Arrest” is “Police”, whereas in the short text, the subject relating to “Arrest” is “criminal”. Since the dependency structure is greatly different between the two, the similarity between the relationship information of the second text and the relationship information of the shortened text is low. The information processing apparatus 10 corrects the weight value of the model M4 so that the degree of similarity increases.

[1-5-2. About generation processing)
Next, an example of a generation process that generates an output text using the model M4 learned by the above-described learning process will be described.

  First, the information processing apparatus 10 acquires information input from the data server 20 to the model M4 (step S33). For example, the information processing apparatus 10 acquires distribution contents (text and title) scheduled to be distributed from the data server 20. Then, the information processing apparatus 10 acquires a title included in the acquired distribution content as an input text.

  Then, the information processing apparatus 10 acquires the model M4 from the internal storage device. As described above, the model M4 is a model learned using the first text as input data and the second text as correct data (teacher label). And the information processing apparatus 10 produces | generates an output text by inputting the relationship information of an input text into the model M4 (step S34).

  Thereafter, the information processing apparatus 10 creates relationship information about the input text and relationship information about the output text. Then, the information processing apparatus 10 compares the relation information of the input text with the relation information of the output text (step S35).

  For example, it is assumed that the related information is dependency information. The information processing apparatus 10 shortens the input text relationship information by removing words that do not appear in the output text from the input text relationship information. Then, the information processing apparatus 10 determines whether or not the dependency structure (tree structure) of the shortened relationship information of the input text and the relationship information of the output text is similar. At this time, the information processing apparatus 10 calculates the similarity between the shortened relationship information of the input text and the relationship information of the output text, and determines whether the similarity is similar based on whether the similarity is equal to or greater than a predetermined threshold. May be. If the relationship information is predicate structure information, the information processing apparatus 10 may determine whether the case analysis results of the words that are predicates are similar.

  The information processing apparatus 10 stores the comparison result (for example, the similarity or the determination result based on the similarity) between the relationship information of the input text and the relationship information of the output text, or a storage device or an external device (for example, the data server 20 or the terminal device 30).

  When the relation information of the input text is similar to the relation information of the output text, the information processing apparatus 10 generates summary content using the output text. Then, the information processing device 10 distributes the summary content to the terminal device 30 (step S36).

  According to the present embodiment, the information processing apparatus 10 is configured to compare input text relation information and output text relation information. Therefore, the user of the information processing apparatus 10 can immediately find an unnatural output text.

[2. Configuration example of information processing apparatus]
The operation of the information processing apparatus 10 according to the present embodiment has been described above. Hereinafter, the configuration of the information processing apparatus 10 will be described.

  The information processing apparatus 10 is a server host computer (hereinafter simply referred to as a “server”) that processes a request from a client computer such as the terminal apparatus 30. The information processing apparatus 10 may be a PC server, a midrange server, or a mainframe server. Further, the information processing apparatus 10 may be configured by a single server, or may be configured by a plurality of servers that cooperate to execute processing. When the information processing apparatus 10 includes a plurality of servers, the installation locations of these servers may be separated. Even if the installation locations are separated, these servers can be regarded as a single information processing device as long as the processing is executed in cooperation. The information processing device 10 functions as a generation device and a learning device.

  As described above, the information processing apparatus 10 is connected to the data server 20 and the terminal apparatus 30 via the network. The network is a communication network such as a LAN (Local Area Network), a WAN (Wide Area Network), a telephone network (a mobile phone network, a fixed telephone network, etc.), a regional IP (Internet Protocol) network, and the Internet. The network may include a wired network or a wireless network.

  FIG. 11 is a diagram illustrating a configuration example of the information processing apparatus 10 according to the embodiment. The information processing apparatus 10 includes a communication unit 11, a storage unit 12, and a control unit 13. Note that the configuration shown in FIG. 11 is a functional configuration, and the hardware configuration may be different.

  The communication unit 11 is a communication interface that communicates with an external device. The communication unit 11 may be a network interface or a device connection interface. For example, the communication unit 11 may be a LAN interface such as a NIC (Network Interface Card) or a USB interface configured by a USB (Universal Serial Bus) host controller, a USB port, or the like. The communication unit 11 may be a wired interface or a wireless interface. The communication unit 11 functions as a communication unit of the information processing apparatus 10. The communication unit 11 communicates with the data server 20 and the terminal device 30 according to the control of the control unit 13.

  The storage unit 12 is a storage device capable of reading and writing data, such as a dynamic random access memory (DRAM), a static random access memory (SRAM), a flash memory, and a hard disk. The storage unit 12 functions as a storage unit of the information processing apparatus 10. The storage unit 12 stores a learning data database 121, a model database 122, and a content information database 123.

  Learning data is registered in the learning data database 121. FIG. 12 is a diagram illustrating an example of information registered in the learning data database 121. Information having items such as “learning data ID (Identifier)”, “text”, “title”, and “heading” is registered in the learning data database 121.

  Of the information shown in FIG. 12, “text”, “title”, and “headline” correspond to “text”, “title”, and “headline” shown in FIG. In addition to the information shown in FIG. 12, various information related to the user who has viewed the learning data and the summary data may be registered in the learning data database 121. In the example shown in FIG. 12, conceptual information such as “K11 to K13”, “T11 to T13”, and “U11 to U13” is shown as information registered in the learning data database 121. Text data and binary data will be registered.

  Here, the “learning data ID” is an identifier for identifying learning data. Further, the “body” is a text that becomes a body included in content (for example, an article) distributed to the user. The “title” is a title attached to the content or the text included in the content. “Heading” is a heading attached to content (text or title).

  For example, in the example illustrated in FIG. 12, information such as a learning data ID “1001”, a text “K11”, a title “T11”, and a heading “U11” is registered in association with each other. Such information indicates, for example, that the learning data indicated by the learning data ID “1001” includes the text indicated by “K11”, the title indicated by “T11”, and the heading indicated by “U11”. .

  Returning to FIG. 11, model data of the information processing apparatus 10 is registered in the model database 122. FIG. 13 is a diagram illustrating an example of information registered in the model database 122. In the example illustrated in FIG. 13, information such as “model ID” and “model data” is registered in the model database 122.

  Here, the “model ID” is information for identifying each model. The “model data” is data of the model indicated by the associated “model ID”. For example, nodes in each layer, functions adopted by each node, connection relationships between nodes, and connections between nodes Information including a connection coefficient set for.

  For example, in the example illustrated in FIG. 13, information such as model ID “2001” and model data “M1” is registered in association with each other. Such information indicates, for example, that the model data indicated by “2001” is “M1”. In the example shown in FIG. 13, conceptual information such as “M1 to M4” is described as information registered in the model database 122, but actually, a character string or a numerical value indicating the model structure or connection coefficient Etc. will be registered. Hereinafter, M1 to M4 and the like are collectively referred to as a model M.

  The model M includes information related to first relationship information indicating a relationship between words of a plurality of words included in the first text, and text corresponding to the first text and second text satisfying a generation condition. It is a model learned based on this. The model M generates text corresponding to the input text from information on the input text and satisfying the generation condition. Such a model M includes an input layer for inputting information related to the input text, an output layer for outputting text corresponding to the input text and satisfying the generation condition, and any layer from the input layer to the output layer. A first element belonging to a layer other than the output layer, and a second element whose value is calculated based on the first element and the weight of the first element, and for information input to the input layer Information regarding the input text input to the input layer by performing an operation based on the first element and the weight of the first element (that is, the connection coefficient) with each element belonging to each layer other than the output layer as the first element This is a model for causing the computer to function so that text corresponding to the input text and satisfying the generation condition is output from the output layer.

  The model M has the first text as input data, the second text as correct data, the second text relation information, the text relation information obtained by inputting the first text into the model, The model may be learned based on the degree of similarity. At this time, the model M generates text corresponding to the input text from the input text and satisfying the generation condition. Such a model M includes an input layer for inputting input text, an output layer for outputting text corresponding to the input text and satisfying a generation condition, and any layer from the input layer to the output layer. A first element belonging to a layer other than the output layer, and a second element whose value is calculated based on the first element and the weight of the first element, and outputs the information input to the input layer By performing an operation based on the first element and the weight of the first element (that is, the connection coefficient) with each element belonging to each layer other than the layer as the first element, according to the input text input to the input layer, This is a model for causing a computer to function so that text corresponding to an input text and satisfying a generation condition is output from an output layer.

  Further, the model M may be a model in which the first text is learned as input data and the first relationship information is learned as correct answer data. At this time, the model M generates input text relationship information from the input text. Such a model M includes an input layer for inputting input text, an output layer for outputting relation information of the input text, and a layer belonging to any layer other than the output layer, from the input layer to the output layer. Each element belonging to each layer other than the output layer with respect to the information input to the input layer. By performing an operation based on the first element and the weight of the first element (that is, the connection coefficient) as the first element, the relation information of the input text is obtained from the output layer according to the input text input to the input layer. It is a model for making a computer function so as to output.

  Here, it is assumed that the model M is realized by a regression model represented by “y = a1 * x1 + a2 * x2 +... + Ai * xi”. In this case, the first element included in the model M corresponds to input data (xi) such as x1 and x2. The weight of the first element corresponds to the coefficient ai corresponding to xi. Here, the regression model can be regarded as a simple perceptron having an input layer and an output layer. When each model is regarded as a simple perceptron, the first element can correspond to any node of the input layer, and the second element can be regarded as a node of the output layer.

  Further, it is assumed that the model M is realized by a neural network having one or a plurality of intermediate layers such as DNN (Deep Neural Network). In this case, the first element included in the model M corresponds to any node of the input layer or the intermediate layer. The second element corresponds to the next node, which is a node to which a value is transmitted from the node corresponding to the first element. The weight of the first element corresponds to a connection coefficient that is a weight considered for a value transmitted from a node corresponding to the first element to a node corresponding to the second element.

  The information processing apparatus 10 calculates the output text using a model having an arbitrary structure such as the above-described regression model or neural network. Specifically, in the model M, when information related to the input text is input, a coefficient is set so that the text corresponding to the input text and satisfying the generation condition is output. Alternatively, in the model M, an input text is input, but when input, the coefficient is set so that text corresponding to the input text and satisfying a predetermined condition is output. For example, the information processing apparatus 10 sets the coefficient based on the similarity between the second text relation information and the text relation information obtained by inputting the first text into the model M. The information processing apparatus 10 generates text (for example, a headline) that satisfies the generation condition using such a model M. In the model M, a coefficient may be set so as to output relation information of the input text when the input text is input.

  In the above example, when the model M receives input text or information related to the input text, the model M outputs a text (for example, a headline) that corresponds to the input text and satisfies the generation conditions (for example, headings). An example of model V) is shown. However, the model M according to the embodiment may be a model generated based on a result obtained by repeatedly inputting / outputting data to / from the model V. For example, the model V is a model learned to receive “input text or information about the input text” as input and output “text corresponding to the input text and satisfying the generation conditions” output from the model V. There may be.

  When the information processing apparatus 10 performs a learning process or a generation process using GAN (Generative Adversarial Networks), the model M may be a model that constitutes a part of the GAN.

  Returning to FIG. 11, the content information database 123 registers information about the content to be distributed to the user. For example, articles acquired from content partners are registered in the content information database 123. FIG. 14 is a diagram illustrating an example of information registered in the content information database 123. In the content information database 123, information having items such as “content ID”, “text”, and “title” is registered. The “content ID” is an identifier for identifying content data. The “text” and “title” are the same as the “text” and “title” shown in FIG.

  Returning to FIG. 11, the control unit 13 is a controller and is stored in a storage device inside the information processing apparatus 10 by a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). Various programs are implemented by executing the RAM or the like as a work area. The control unit 13 is a controller, and may be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

  In addition, the control unit 13 performs data processing (for example, relationship information, title) input to the input layer of the model M by information processing according to the model M (models M1 to M4, etc.) stored in the storage unit 12. The calculation based on the coefficients of the model M (that is, the coefficients corresponding to the various features learned by the model M) is performed, and output text (for example, headings) is output from the output layer of the model M.

  As shown in FIG. 11, the control unit 13 includes a learning data acquisition unit 131, a relationship information acquisition unit 132, a learning unit 133, an output information acquisition unit 134, a generation unit 135, a comparison unit 136, and output control. Part 137. Blocks constituting the control unit 13 (learning data acquisition unit 131 to output control unit 137) are functional blocks indicating functions of the control unit 13, respectively. These functional blocks may be software blocks or hardware blocks. For example, each functional block described above may be one software module realized by software (including a microprogram), or may be one circuit block on a semiconductor chip (die). Of course, each functional block may be one processor or one integrated circuit. The method of configuring the functional block is arbitrary. In addition, the control part 13 may be comprised by the functional unit different from the above-mentioned functional block.

  The learning data acquisition unit 131 acquires, as learning data, information related to the first text and a second text that corresponds to the first text and satisfies the generation condition. For example, the learning data acquisition unit 131 acquires information related to first relationship information indicating a relationship between words of a plurality of words included in the first text as information related to the first text. For example, the learning data acquisition unit 131 acquires first relation information as information related to the first text. Or the learning data acquisition part 131 acquires the 2nd relationship information which shortened the 1st relationship information as information regarding a 1st text. Or the learning data acquisition part 131 acquires a 1st text and 1st relationship information as information regarding a 1st text. The learning data acquisition unit 131 acquires the first text and the second text that corresponds to the first text and satisfies the generation condition.

  The relationship information acquisition unit 132 acquires first relationship information indicating a relationship between words of a plurality of words included in the first text. Moreover, the relationship information acquisition part 132 acquires the 2nd relationship information which shortened the 1st relationship information.

  The learning unit 133 learns the model M and stores the learned model M in the model database 122. More specifically, when the learning unit 133 inputs information related to the first text to the model M, the model M outputs the second text (for example, a headline corresponding to the input title). Set the connection coefficient of model M. That is, the learning unit 133 learns the model M so that when the information related to the input text is input, the model M outputs the output text satisfying the generation condition that is the text corresponding to the input text. In addition, the learning unit 133 sets the connection coefficient of the model M so that when the first text is input to the model M, the model M outputs the second text. For example, the model M sets the connection coefficient of the model M based on the similarity between the relationship information of the second text and the relationship information of the text obtained by inputting the first text into the model.

  For example, the learning unit 133 is a node in the input layer of the model M, and the first text or the first text is input to the node corresponding to the input layer of the encoder that has learned the feature corresponding to the information input to the input layer. By inputting information related to the text and propagating data through the intermediate layers to the output layer of the model M, the text corresponding to the first text and satisfying the generation condition is output. Then, the learning unit 133 corrects the connection coefficient of the model M based on the difference between the text actually output by the model M and the second text (for example, heading) included in the learning data. For example, the learning unit 133 may correct the connection coefficient using a technique such as backpropagation. The model M may correct the connection coefficient based on the similarity between the relationship information of the second text and the relationship information of the text obtained by inputting the first text into the model. Further, the learning unit 133 may correct the connection coefficient based on the cosine similarity between the vector that is the distributed representation of the second text and the vector that is the distributed representation of the text actually output by the model M. Good.

  Note that the learning unit 133 may learn the model M using any learning algorithm. For example, the learning unit 133 may learn the model M using a learning algorithm such as a neural network, a support vector machine, clustering, or reinforcement learning.

  Here, the learning unit 133 determines a predetermined condition from the information regarding the input text based on the information regarding the first relationship information indicating the relationship between the words of the plurality of words included in the first text and the second text. A first model that generates an output text that satisfies is learned. At this time, the first model may be a model that generates text having a predetermined number of characters or less as output text.

  The learning unit 133 learns the first model using the first relation information as input data and the second text as correct data.

  The learning unit 133 learns the first model using the second relation information obtained by shortening the first relation information as input data and the second text as correct data.

  The learning unit 133 learns the first model using the first text and the first relation information as input data and the second text as correct answer data.

  In addition, the learning unit 133 learns the second model that generates the relationship information of the input text from the input text using the first text as input data and the first relationship information as correct data.

  The learning unit 133 also learns a model that generates an output text that satisfies a predetermined condition from the input text by using the first text as input data and the second text as correct data. At this time, the learning unit 133 is based on the similarity between the relationship information of the first text or the relationship information of the second text and the relationship information of the text obtained by inputting the first text into the model. Train the model.

  The output information acquisition unit 134 is information related to first relation information indicating a relation between words of a plurality of words included in the first text, and text corresponding to the first text, and a second condition that satisfies a predetermined condition. The information about the input text input to the first model learned based on the text is acquired. The output information acquisition unit 134 acquires input text that is input to a model learned using the first text as input data and the second text as correct answer data.

  The generation unit 135 inputs information related to the input text to the first model, thereby generating an output text that corresponds to the input text and satisfies a predetermined condition.

  The first model is a model obtained by learning the first relation information as input data and the second text as correct answer data. The generation unit 135 generates output text by inputting the relationship information of the input text to the first model.

  The first model is a model in which the second relation information obtained by shortening the first relation information is learned as input data and the second text is learned as correct data. The generation unit 135 generates output text by inputting the relationship information obtained by shortening the relationship information of the input text to the first model.

  The first model is a model in which the first text and the first relationship information are learned as input data, and the second text is learned as correct data. The generation unit 135 generates an output text by inputting the input text and the relationship information of the input text to the first model.

  The first model is a model in which the first text and the first relationship information are learned as input data, and the second text is learned as correct data. The generation unit 135 generates the relationship information of the input text by inputting the input text to the second model learned using the first text as input data and the first relationship information as correct data.

  The generation unit 135 generates a text having a predetermined number of characters or less as the output text.

  In addition, the generation unit 135 generates an output text that satisfies a predetermined condition that is a text corresponding to the input text by inputting the input text into the model.

  The comparison unit 136 compares the relationship information of the input text with the relationship information of the output text.

  The output control unit 137 outputs the text generated by the generation unit 135 to the user. For example, the output control unit 137 distributes content including the text generated by the generation unit 135 in response to a request from the terminal device 30. Note that the output control unit 137 may provide content including the text generated by the generation unit 135 to the data server 20 and distribute the content from the data server 20.

[3. Processing flow of information processing apparatus]
Next, a procedure of processing executed by the information processing apparatus 10 will be described. The information processing apparatus 10 performs learning processing and generation processing. The information processing apparatus 10 includes a multitask OS and can execute these processes in parallel.

[3-1. Learning process)
First, the learning process will be described. FIG. 15 is a flowchart illustrating an example of the learning process. The learning process is a process of learning a model M that generates output text from information related to input text. When the information processing apparatus 10 receives a process start command from the user, the information processing apparatus 10 executes a learning process.

  First, the information processing apparatus 10 acquires learning data (step S101). For example, the information processing apparatus 10 acquires information on the distribution content (text and title) from the data server 20 and acquires a headline corresponding to the distribution content. At this time, the information processing apparatus 10 acquires the title as the first text and the heading as the second text.

  Subsequently, the information processing apparatus 10 acquires the relationship information of the first text (Step S102).

  Subsequently, the information processing apparatus 10 performs learning of a model for generating relationship information (for example, the model M2 shown in FIG. 6) using the first text as input data and the relationship information of the first text as correct data. (Step S103).

  Subsequently, the information processing apparatus 10 uses the relation information of the first text as input data and the second text as correct answer data, for example, a model for generating the second text (heading) (for example, the model M1 shown in FIG. 1). ) Is executed (step S104). The information processing apparatus 10 uses the relationship information between the first text and the first text as input data and the second text as correct data, and generates a model (for example, shown in FIG. 6) for generating the second text (heading). Learning of model M3) may be performed.

  Alternatively, the information processing apparatus 10 learns a model for generating the second text (heading) (for example, the model M4 shown in FIG. 9) using the first text and input data and the second text as correct answer data. May be executed. At this time, the information processing apparatus 10 is based on the similarity between the relationship information of the first text or the relationship information of the second text and the relationship information of the text obtained by inputting the first text into the model. The model may be learned.

  When the execution of learning ends, the information processing apparatus 10 ends the learning process.

[3-2. Generation process)
Next, the generation process will be described. FIG. 16 is a flowchart illustrating an example of the generation process. The generation process is a process for generating an output text (short text of the input text) from information related to the input text. “Input text” is text input directly or indirectly to the model, and output text is text output from the model. Further, “information related to input text” may be input text related information, or input text and input text related information. The “information about the input text” may be the input text itself. When the information processing apparatus 10 receives a process start command from the user, the information processing apparatus 10 executes a generation process.

  First, the information processing apparatus 10 acquires content information (step S201). For example, the information processing apparatus 10 acquires information on distribution contents (text and title) from the data server 20. At this time, the information processing apparatus 10 may acquire a title as input text.

  Subsequently, the information processing apparatus 10 acquires related information of the input text (step S202). At this time, the information processing apparatus 10 may acquire the relationship information by inputting the input text to the model M2 illustrated in FIG.

  Subsequently, the information processing apparatus 10 generates the output text by inputting the information acquired in step S201 and / or step S202 into the model (step S203). For example, the information processing apparatus 10 generates output text by inputting the relationship information acquired in step S202 to the model M1 illustrated in FIG. Alternatively, the information processing apparatus 10 generates the output text by inputting the input text acquired in step S201 and the relationship information acquired in step S202 to the model M3 illustrated in FIG. Alternatively, the information processing apparatus 10 generates the output text by inputting the input text acquired in step S201 to the model M4 illustrated in FIG.

  Subsequently, the information processing apparatus 10 compares the relation information of the input text with the relation information of the output text (step S204). As a result of the comparison, when the two are not similar (for example, when the similarity is a predetermined threshold value or less), the information processing apparatus 10 may output to the user that the two are not similar.

  Subsequently, the information processing apparatus 10 outputs the output text (heading) generated in step S203 to the user (step S205). When the output is completed, the information processing apparatus 10 ends the generation process.

[4. (Modification)
The above-described embodiment shows an example, and various changes and applications are possible.

  For example, in the above-described embodiment, the information processing apparatus 10 learns the models M1 to M4 using the first text as the title and the second text as the headline. However, the first text and the second text are not limited to the above example. For example, the information processing apparatus 10 may learn the models M1 to M4 using the first text as the body and the second text as the headline. In addition, the text used as the first text and the second text can be arbitrarily changed. Similarly, when generating text, the text used as input text can be arbitrarily changed.

  Further, in the above-described embodiment, the information processing apparatus 10 learns the model M2 using the first text as input data and the relationship information of the first text as correct data. However, the information processing apparatus 10 may learn the model M2 using the first text as input data and the shortened relationship information obtained by shortening the relationship information of the first text as correct data. Then, the information processing apparatus 10 may generate the shortening relation information by inputting the first text to the model M2 learned in this way.

  Further, in the above-described embodiment, the encoder and the decoder configuring the model M are configured by LSTM, but may be configured by RNN other than LSTM. The encoder and decoder that constitute the model M may be a neural network other than the RNN, such as a CNN (Convolution Neural Network). In addition, the encoder and decoder constituting the model M may be a neural network that simply changes the number of dimensions of the input information amount. At this time, the encoders constituting the model M may be configured to extract features of the input information by compressing the dimensionality of the input information. In addition, the decoder constituting the model M may be configured to increase the dimension amount of the feature extracted by the encoder and output information having a smaller number of dimensions than the information input to the encoder.

  The control device that controls the information processing apparatus 10 according to the present embodiment may be realized by a dedicated computer system or an ordinary computer system. For example, a program or data (for example, model M) for executing the above-described operation is stored and distributed in a computer-readable recording medium such as an optical disk, a semiconductor memory, a magnetic tape, or a flexible disk, and the program is distributed to the computer. The control device may be configured by installing the software and executing the above-described processing. The control device may be an external device (for example, a personal computer) of the information processing device 10 or an internal device (for example, the control unit 13). Further, the program may be stored in a disk device provided in a server device on a network such as the Internet so that it can be downloaded to a computer. Further, the above-described functions may be realized by cooperation between an OS (Operating System) and application software. In this case, a part other than the OS may be stored and distributed in a medium, or a part other than the OS may be stored in a server device and downloaded to a computer.

  In addition, among the processes described in the above embodiments, all or a part of the processes described as being automatically performed can be manually performed, or the processes described as being manually performed All or a part of the above can be automatically performed by a known method. In addition, the processing procedures, specific names, and information including various data and parameters shown in the document and drawings can be arbitrarily changed unless otherwise specified. For example, the various types of information illustrated in each drawing is not limited to the illustrated information.

  Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured.

  Moreover, each embodiment mentioned above can be combined suitably in the range which does not contradict a process content.

[5. Hardware configuration)
The information processing apparatus 10 according to the embodiment and the modification may be realized by a computer 1000 having a configuration as illustrated in FIG. FIG. 17 is a hardware configuration diagram illustrating an example of a computer that implements the functions of the information processing apparatus 10. The computer 1000 includes a CPU (Central Processing Unit) 1100, a RAM 1200, a ROM 1300, an HDD (Hard Disk Drive) 1400, a communication interface (I / F) 1500, an input / output interface (I / F) 1600, and a media interface (I / F). ) 1700.

  The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400 and controls each unit. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started up, a program depending on the hardware of the computer 1000, and the like.

  The HDD 1400 stores programs executed by the CPU 1100, data used by the programs, and the like. The communication interface 1500 receives data from other devices via the network N and sends the data to the CPU 1100, and transmits data generated by the CPU 1100 to other devices via the network N.

  The CPU 1100 controls an output device such as a display and a printer and an input device such as a keyboard and a mouse via an input / output interface 1600. The CPU 1100 acquires data from the input device via the input / output interface 1600. In addition, the CPU 1100 outputs the generated data to the output device via the input / output interface 1600.

  The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads such a program from the recording medium 1800 onto the RAM 1200 via the media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase change rewritable disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. Etc.

  For example, when the computer 1000 functions as the information processing apparatus 10 according to the embodiment, the CPU 1100 of the computer 1000 implements the control unit 13 by executing a program or data (for example, model M) loaded on the RAM 1200. To do. The CPU 1100 of the computer 1000 reads these programs or data (for example, model M) from the recording medium 1800 and executes them, but as another example, these programs or data (for example, for example, from other devices via the network N). Model M) may be obtained.

  As described above, some of the embodiments of the present application have been described in detail with reference to the drawings. It is possible to implement the present invention in other forms with improvements.

[6. effect〕
According to the present embodiment, the information processing apparatus 10 acquires information related to the first text and a second text that corresponds to the first text and satisfies the generation condition. Then, the information processing apparatus 10 determines the generation condition from the information related to the input text based on the information related to the first relationship information indicating the relationship between the words of the plurality of words included in the first text and the second text. A first model that generates output text that satisfies is learned. At this time, the first model may be a model that generates text having a predetermined number of characters or less as output text. Since the first model learns based on the relationship information, the generation of text with an unnatural syntax is small. By using the first model, the information processing apparatus 10 can generate highly accurate text.

  The information processing apparatus 10 learns the first model using the first relation information as input data and the second text as correct data. Thereby, the information processing apparatus 10 can generate a model with high accuracy. As a result, the information processing apparatus 10 can generate text with high accuracy.

  The information processing apparatus 10 learns the first model using the second relation information obtained by shortening the first relation information as input data and the second text as correct data. The first model learns based on second relation information (shortening relation information) closer to the second text. Since the accuracy of the first model is high, the information processing apparatus 10 can generate text with high accuracy.

  The information processing apparatus 10 learns the first model using the first text and the first relation information as input data and the second text as correct answer data. Thereby, the information processing apparatus 10 can generate a model with high accuracy. As a result, the information processing apparatus 10 can generate text with high accuracy.

  The information processing apparatus 10 learns the second model that generates the relationship information of the input text from the input text using the first text as input data and the first relationship information as correct data. Thereby, the information processing apparatus 10 can generate a highly accurate model for generating the relationship information.

  The information processing apparatus 10 acquires the first text and the second text that corresponds to the first text and satisfies the generation condition. In addition, the information processing apparatus 10 learns a model that generates an output text satisfying a generation condition from the input text by using the first text as input data and the second text as correct data. And the information processing apparatus 10 is based on the similarity between the relationship information of the first text or the relationship information of the second text and the relationship information of the text obtained by inputting the first text into the model. Train the model. Since the model generated by the information processing apparatus 10 learns based on the similarity of the relationship information, the generation of text with an unnatural syntax is small. By using the first model, the information processing apparatus 10 can generate highly accurate text.

  The information processing apparatus 10 includes information related to first relationship information indicating a relationship between words of a plurality of words included in the first text, and text corresponding to the first text and satisfying a generation condition. The information regarding the input text input into the 1st model learned based on these is acquired. And the information processing apparatus 10 produces | generates the output text which is a text corresponding to an input text, and satisfy | fills a production | generation condition by inputting the information regarding an input text into a 1st model. At this time, the information processing apparatus 10 may generate a text having a predetermined number of characters or less as the output text. Since the first model learns based on the relationship information, the generation of text with an unnatural syntax is small. As a result, the information processing apparatus 10 can generate text with high accuracy.

  In the information processing apparatus 10, the first model is a model learned using the first relation information as input data and the second text as correct data. And the information processing apparatus 10 produces | generates an output text by inputting the relationship information of an input text into a 1st model. Thereby, the information processing apparatus 10 can generate text with high accuracy.

  The first model is a model in which the second relation information obtained by shortening the first relation information is learned as input data and the second text is learned as correct data. The information processing apparatus 10 generates output text by inputting the relationship information obtained by shortening the relationship information of the input text to the first model. Thereby, the information processing apparatus 10 can generate text with high accuracy. The first model learns based on second relation information (shortening relation information) closer to the second text. Since the accuracy of the first model is high, the information processing apparatus 10 can generate text with high accuracy.

  The first model is a model in which the first text and the first relationship information are learned as input data, and the second text is learned as correct data. The information processing apparatus 10 generates the output text by inputting the input text and the relationship information of the input text to the first model. Thereby, the information processing apparatus 10 can generate text with high accuracy.

  The information processing apparatus 10 generates the relationship information of the input text by inputting the input text to the second model learned using the first text as input data and the first relationship information as correct data. Thereby, the information processing apparatus 10 can generate text with high accuracy.

  The information processing apparatus 10 acquires input text input to a model learned using the first text as input data and the second text as correct data. Then, the information processing apparatus 10 generates the output text that satisfies the generation condition and is the text corresponding to the input text by inputting the input text into the model. Then, the information processing apparatus 10 compares the relationship information of the input text with the relationship information of the output text. Thereby, the user of the information processing apparatus 10 can immediately find an unnatural output text.

  As described above, some of the embodiments of the present application have been described in detail with reference to the drawings. However, these are merely examples, and various modifications, including the aspects described in the disclosure section of the invention, based on the knowledge of those skilled in the art, It is possible to implement the present invention in other forms with improvements.

  In addition, the “section (module, unit)” described above can be read as “means” or “circuit”. For example, the learning unit can be read as learning means or a learning circuit.

DESCRIPTION OF SYMBOLS 10 ... Information processing apparatus 11 ... Communication part 12 ... Storage part 121 ... Learning data database 122 ... Model database 123 ... Content information database 13 ... Control part 131 ... Learning data acquisition part 132 ... Relation information acquisition part 133 ... Learning part 134 ... Output Information acquisition unit 135... Generation unit 136... Comparison unit 137... Output control unit 20... Data server 30.. D2, D3, D4 ... Decoder

Claims (24)

An acquisition unit that acquires information about the first text, and a second text that corresponds to the first text and satisfies a predetermined condition;
An output text satisfying the predetermined condition is generated from the information related to the input text based on the information related to the first relation information indicating the relationship between the words of the plurality of words included in the first text and the second text. A learning unit for learning the first model to be
A learning apparatus comprising: The learning unit learns the first model using the first relation information as input data and the second text as correct data.
The learning apparatus according to claim 1. The learning unit learns the first model by using the second relation information obtained by shortening the first relation information as input data and the second text as correct data.
The learning apparatus according to claim 1. The learning unit learns the first model using the first text and the first relation information as input data, and the second text as correct answer data.
The learning apparatus according to claim 1. The learning unit learns a second model that generates relation information of the input text from the input text using the first text as input data and the first relation information as correct answer data.
The learning apparatus according to claim 4. The first model is a model that generates a text having a predetermined number of characters or less as the output text.
The learning apparatus according to claim 1, wherein the learning apparatus is a learning apparatus. An acquisition unit configured to acquire a first text and a second text that corresponds to the first text and satisfies a predetermined condition;
A learning unit that learns a model that generates the output text satisfying the predetermined condition from the input text using the first text as input data and the second text as correct answer data,
The learning unit is based on a degree of similarity between the relationship information of the first text or the relationship information of the second text and the relationship information of the text obtained by inputting the first text into the model. Learn the model,
A learning apparatus characterized by that. Based on information related to first relation information indicating a relation between words of a plurality of words included in the first text, and a second text that corresponds to the first text and satisfies a predetermined condition. An acquisition unit for acquiring information related to the input text input to the learned first model;
A generation unit that generates information corresponding to the input text and that satisfies the predetermined condition by inputting information related to the input text to the first model;
A generating apparatus comprising: The first model is a model obtained by learning the first relation information as input data and the second text as correct answer data,
The generation unit generates the output text by inputting relation information of the input text to the first model.
The generating apparatus according to claim 8. The first model is a model in which the second relation information obtained by shortening the first relation information is learned as input data, and the second text is learned as correct data,
The generation unit generates the output text by inputting relationship information obtained by shortening the relationship information of the input text to the first model.
The generating apparatus according to claim 8. The first model is a model in which the first text and the first relation information are learned as input data and the second text is learned as correct data,
The generation unit generates the output text by inputting the input text and relation information of the input text to the first model.
The generating apparatus according to claim 8. The generating unit generates the relation information of the input text by inputting the input text to a second model learned using the first text as input data and the first relation information as correct data.
The generating apparatus according to claim 11. The generation unit generates a text having a predetermined number of characters or less as the output text.
The generation device according to claim 8, wherein the generation device is a device. An acquisition unit that acquires an input text input to a model that has been learned using the first text as input data and the second text as correct data;
Generating a text corresponding to the input text and generating an output text satisfying a predetermined condition by inputting the input text into the model;
A comparison unit that compares the relationship information of the input text and the relationship information of the output text;
A generating apparatus comprising: A learning method executed by a learning device,
An acquisition step of acquiring information relating to the first text, and a second text corresponding to the first text and satisfying a predetermined condition;
An output text satisfying the predetermined condition is generated from the information related to the input text based on the information related to the first relation information indicating the relationship between the words of the plurality of words included in the first text and the second text. A learning process for learning the first model to be performed;
The learning method characterized by including. A learning method executed by a learning device,
An acquisition step of acquiring a first text and a second text corresponding to the first text and satisfying a predetermined condition;
A learning step of learning a model that generates the output text satisfying the predetermined condition from the input text, using the first text as input data and the second text as correct answer data,
In the learning step, the model is learned based on the similarity between the relationship information of the second text and the relationship information of the text obtained by inputting the first text to the model.
A learning method characterized by that. A generation method executed by a generation device,
Based on information related to first relation information indicating a relation between words of a plurality of words included in the first text, and a second text that corresponds to the first text and satisfies a predetermined condition. An acquisition step of acquiring information relating to the input text input to the learned first model;
Generating a text corresponding to the input text and satisfying the predetermined condition by inputting information on the input text into the first model;
A generation method comprising: A generation method executed by a generation device,
An acquisition step of acquiring an input text input to a model learned using the first text as input data and the second text as correct data;
Generating a text corresponding to the input text and satisfying a predetermined condition by inputting the input text into the model;
A comparison step of comparing the relationship information of the input text with the relationship information of the output text;
A generation method comprising: An acquisition procedure for acquiring information related to the first text, and a second text corresponding to the first text and satisfying a predetermined condition;
An output text satisfying the predetermined condition is generated from the information related to the input text based on the information related to the first relation information indicating the relationship between the words of the plurality of words included in the first text and the second text. Learning procedure to learn the model to be
A learning program to make a computer execute. An acquisition procedure for acquiring a first text and a second text that corresponds to the first text and satisfies a predetermined condition;
A learning procedure for learning a model for generating an output text satisfying the predetermined condition from the input text, using the first text as input data and the second text as correct answer data;
In the learning procedure, the model is learned based on the similarity between the relationship information of the second text and the relationship information of the text obtained by inputting the first text to the model.
A learning program characterized by that. Based on information related to first relation information indicating a relation between words of a plurality of words included in the first text, and a second text that corresponds to the first text and satisfies a predetermined condition. An acquisition procedure for acquiring information about the input text input to the learned model;
A generation procedure for generating output text satisfying the predetermined condition, which is text corresponding to the input text by inputting information on the input text to the model;
A generation program that causes a computer to execute. An acquisition procedure for acquiring an input text input to a model learned using the first text as input data and the second text as correct data;
Generating a text corresponding to the input text and generating an output text satisfying a predetermined condition by inputting the input text into the model;
A comparison procedure for comparing the relationship information of the input text with the relationship information of the output text;
A generation program that causes a computer to execute. Based on information related to first relation information indicating a relation between words of a plurality of words included in the first text, and a second text that corresponds to the first text and satisfies a predetermined condition. A learned model,
An input layer for entering information about the input text;
An output layer that outputs text corresponding to the input text and satisfying the predetermined condition;
A first element belonging to any layer from the input layer to the output layer other than the output layer;
A second element whose value is calculated based on the first element and a weight of the first element;
According to the information on the input text input to the input layer, the text corresponding to the input text and satisfying the predetermined condition is output from the output layer,
A model for making a computer function. The first text is input data, the second text is correct data, and the degree of similarity between the relation information of the second text and the relation information of the text obtained by inputting the first text into the model A model learned based on
An input layer for entering input text,
An output layer that outputs text corresponding to the input text and satisfying a predetermined condition;
A first element belonging to any layer from the input layer to the output layer other than the output layer;
A second element whose value is calculated based on the first element and a weight of the first element;
In response to the input text input to the input layer, text corresponding to the input text and satisfying the predetermined condition is output from the output layer.
A model for making a computer function.

Images