JP6552353B2 - Information providing apparatus, information providing method, and information providing program - Google Patents

Description

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

  Conventionally, there is known a technique for searching or generating information related to input information based on an analysis result of the input information and outputting the searched or generated information as a response. As an example of such a technology, words, sentences, and contexts contained in the input text are converted into multidimensional vectors and analyzed, and based on the analysis result, text similar to the input text or input A technique of natural language processing is known which analogizes text following the text and outputs an analog result.

JP, 2006-127077, A

“Natural Language Processing with word2vec”, Yasukazu Nishio, May 2014, ISBN978-4-87311-683-9 "Study on Hypothesis-based Knowledge Generation Support for Creative Design", The Japan Society of Mechanical Engineers, No. 03-27, 13th Design Engineering and Systems Division Conference Proceedings

  However, in the above-mentioned prior art, there are cases where it is not possible to output information for assisting the creation of the user. For example, in the above-described prior art, only information that can be predicted by the user, such as text similar to the input text or text following the input text, is output, and is related to the input text. However, it is difficult for the user to output information that can not be predicted. For this reason, in the above-mentioned prior art, it is impossible to provide information that gives inspiration to the user.

  The present application is made in view of the above, and an object of the present invention is to provide an information providing apparatus, an information providing method, and an information providing program capable of outputting information for assisting a user's creation.

  The information providing apparatus according to the present application includes a receiving unit that receives input information, a pattern specifying unit that extracts a specific syntax and a group of words embedded in the specific syntax from the input information, and the specified pattern included in the specified pattern. The apparatus is characterized by including a similarity calculation unit that calculates a similarity using a distributed expression of word groups, and an output unit that outputs information extracted based on the similarity calculated by the similarity calculation unit.

  According to one aspect of the embodiment, it is possible to output information that assists the creation of the user.

FIG. 1 is a diagram illustrating an example of an information providing apparatus according to an embodiment. FIG. 2 is a diagram showing an equivalent conversion theory which is one of the invention finding methods. FIG. 3 is a diagram illustrating a first example of input and output variations of the information providing apparatus according to the embodiment. FIG. 4 is a view for explaining a second example of the variation of the input and the output of the information providing apparatus according to the embodiment. FIG. 5 is a diagram illustrating a third example of input and output variations of the information providing apparatus according to the embodiment. FIG. 6 is a diagram illustrating a fourth example of input and output variations of the information providing apparatus according to the embodiment. FIG. 7 is a diagram illustrating an example of a functional configuration of the information providing apparatus according to the embodiment. FIG. 8 is a diagram for explaining an example of a word extracted by the information providing apparatus according to the embodiment. FIG. 9 is a diagram for explaining an example of a word set extracted by the information providing apparatus according to the embodiment. FIG. 10 is a diagram for explaining an example of processing for extending the distributed representation space executed by the information providing apparatus according to the embodiment. FIG. 11 is a diagram for explaining the limitation and expansion of the distributed representation space executed by the information providing apparatus according to the embodiment. FIG. 12 is a diagram for explaining an example of processing for extracting similar words by the information providing apparatus according to the embodiment. FIG. 13 is a diagram for explaining an example of the relationship between word vectors. FIG. 14 is a diagram for explaining an example of the concept of a word proposed by the information providing apparatus. FIG. 15 is a flowchart illustrating the flow of the extraction process performed by the information providing apparatus according to the embodiment. FIG. 16 is a flowchart illustrating the flow of a specific process of the extraction process performed by the information providing apparatus according to the embodiment. FIG. 17 is a flowchart for explaining the flow of a specific process of the process in which the information providing apparatus according to the embodiment selects a word set which is likely to cause serendipity. FIG. 18 is a diagram illustrating an example of processing for outputting information related to information in an input field and information in a different field. FIG. 19 is a hardware configuration diagram showing an example of a computer that implements extraction processing.

  Hereinafter, an information providing apparatus, an information providing method, and a mode for implementing an information providing program according to the present application (hereinafter, referred to as “embodiment”) will be described in detail with reference to the drawings. Note that the information providing apparatus, the information providing method, and the information providing program according to the present application are not limited by the embodiment. Further, in the following embodiments, the same parts will be denoted by the same reference numerals, and overlapping descriptions will be omitted.

[1. Example of information providing device)
First, an example of a process performed by the information providing apparatus 10 will be described with reference to FIG. FIG. 1 is a diagram illustrating an example of an information providing apparatus according to an embodiment. For example, in the example illustrated in FIG. 1, the information providing apparatus 10 acquires a user's speech as input information when a plurality of users are holding a meeting such as brainstorming, and based on the acquired input information. An example of processing to support the user's thinking is described by generating an utterance that does not appear in ordinary thinking, converting the generated utterance into speech and outputting it to a robot or the like.

  The information providing device 10 illustrated in FIG. 1 is realized by an information processing device such as a server device. Note that the information providing device 10 may be realized by a single information processing device, and for example, a plurality of information processing devices existing on a cloud network may be realized in cooperation. Such an information providing apparatus 10 converts a user's speech into text data and analyzes the text data by natural language processing. Then, based on the analysis result, the information providing apparatus 10 generates a speech that supports the meeting or the user's thinking, and outputs the generated speech.

  Here, in the prior art, a distributed expression indicated by multi-dimensional word vectors constituting input text is used to analogize text similar to the input text and text following the input text. However, in the prior art, the input text and the distributed expression are similar texts, ie, texts that can be predicted by the user. For this reason, the conventional technology provides information that the user has never thought of, information that gives the user a new inspiration, that is, information that can cause the user's serendipity (inspiration, awareness, surprise). I could not Also, for example, if text that does not resemble a simple input text is output, text that is not related to the user's thought will be output, which may hinder the user's thought. is there.

  Therefore, the information providing device 10 executes the following process. First, the information processing apparatus 10 accepts a user's speech or the like as input information. Then, when receiving the input information, the information processing apparatus 10 specifies a specific syntax from the input information, and extracts a word group embedded in the specified specific syntax. In addition, the information processing apparatus 10 calculates the degree of similarity with the distributed expression of another word using the distributed expression of the word group included in the specific syntax. Then, the information providing device 10 outputs the information extracted based on the calculated degree of similarity.

  Specifically, the information providing apparatus 10 identifies the field to which the input information belongs. Further, the information providing apparatus 10 identifies a plurality of pieces of information having a predetermined relationship from the information belonging to the identified field. Then, the information providing apparatus 10 extracts information having the same concept as the concept generated by the plurality of identified information from the information belonging to the field different from the identified field.

  Here, the information providing apparatus 10 uses the distributed representation of a plurality of pieces of information having a predetermined degree of association to input information having an attribute in which the degree of similarity with the attribute generated by the plurality of pieces of information satisfies the predetermined condition. It extracts from the information which belongs to the field where the degree of similarity with the field to which it belongs falls below a predetermined threshold. For example, the information providing apparatus 10 identifies a plurality of words having a predetermined relationship from documents in a field to which input information belongs, and identifies a word having an attribute similar to the attribute generated by the identified plurality of words. The document is extracted from documents belonging to the field in which the degree of similarity with. Then, the information providing device 10 outputs the extracted information.

  Here, the concept generated by a plurality of pieces of information is, for example, an attribute of each piece of information. To explain a more specific example, when adopting words as information, an attribute having a plurality of words has a set of attributes such as what word each word is, what meaning each word has, etc. That is, the concept of a set of words. In the following description, the information providing apparatus 10 describes an attribute included in a word, biological information, content, and the like handled as information as a “concept” included in the information, and a plurality of words, a plurality of biometric information, and a plurality of contents. The attribute possessed by etc. is described as the "concept" possessed by the plurality of pieces of information.

  That is, the information providing apparatus 10 specifies the concept formed by the information having the predetermined relationship in the category to which the input information belongs, and from the information belonging to the category different from the input information, the information which can form the common concept Extract. As a result of such processing, the information providing apparatus 10 holds information that has an implicit relationship with the input information, but explicitly has information that has a discontinuous relationship, that is, as a result of different categories, the relationship is apparently irrelevant. Visible information can be provided.

  In this case, while retaining implicit connections with input information, information having explicit discontinuous relationships, that is, information that appears to be unrelated at first glance as a result of different categories is considered by the user. It can be information that was not present or that gives new inspiration to users.

  For example, the information providing apparatus 10 identifies, from information of the field to which the input information belongs, a plurality of pieces of information indicating points of the invention or idea in the field, and indicates a concept indicated by the identified information, that is, points of the invention or idea Identify the concept. Then, the information providing apparatus 10 extracts information similar to the identified concept from the information in the other fields. The information extracted by such processing indicates to the user information satisfying the concept indicating the point of the invention or idea in a field different from the field to which the input information belongs, that is, the point of the invention or idea in another field , It can be information that can cause serendipity of the user. As a result, the information providing apparatus 10 can assist the creation of the user.

For example, FIG. 2 is a figure which shows the equivalent conversion theory which is one of the invention excavation methods. For example, as shown in (A) in FIG. 2, Ao which is the original specific event (for example, the original invention) is a viewpoint vi that meets the development purpose as shown in (B) in FIG. It consists of. Further, as shown in FIG. 2C, Ao includes special conditions 特殊 a possessed by Ao, such as technical attributes to which Ao belongs. For this reason, as shown in FIG. 2D, the concept cε obtained by removing Σa from Ao can be a concept that is the core of Ao. Then, when a special condition) b different from aa is considered in the concept cε as shown in (E) in FIG. 2, a new invention B is conceptually obtained as shown in (F) in FIG. _It is considered that _τ can be derived.

Therefore, the information providing apparatus 10 extracts the concept of the idea in the field to which the input information belongs, that is, the concept cε that is the core of Ao from Ao, by the extraction process described later, and applies the extracted concept cε to the condition に よ り b of the other fields Thus, the information is provided to the user such that a new idea B _τ can be derived.

  Hereinafter, an example of a specific extraction process performed by the information providing apparatus 10 will be described using the example of FIG. 1. First, the information providing apparatus 10 receives the user's utterance A and utterance B in the conference as an input (step S1). For example, the information providing apparatus 10 converts the speech A uttered by the user into text data, and acquires the converted text data as input information.

  In such a case, the information providing apparatus 10 is a word belonging to a field different from the field to which the input information belongs, and shows a concept similar to the concept formed by a plurality of words having predetermined relationships in the field to which the input information belongs. An extraction process for extracting words is executed (step S2). Hereinafter, the flow of the generation process performed by the information providing apparatus 10 will be described by dividing it into steps S3 to S8.

  First, the information providing apparatus 10 identifies the field to which the input information belongs (step S3). For example, the information providing apparatus 10 analyzes text data of a user's utterance and identifies in which field the word included in the text data is used. For example, when the information providing apparatus 10 includes words such as “glasses type”, “watch type”, “display”, “smart device” and the like, the field to which the input information belongs is “wearable device”. The information providing apparatus 10 stores, for example, a word that will be included in the input information and a field to which the input information including the word belongs in advance in association with each other, and is associated with the word included in the input information The field may be specified as the field to which the input information belongs. For example, a search service on the web is used to search the field associated with the word included in the input information, and the field obtained as the search result is input It may be a field to which information belongs.

  Subsequently, the information providing device 10 extracts a word set including a plurality of words having a predetermined relationship from the document data in the identified field (step S4). For example, the information providing apparatus 10 acquires document data of a specified field from a document database describing various technologies, such as disclosed patent publications and patent publications. Subsequently, the information providing apparatus 10 performs morphological analysis on the acquired document data, and extracts a plurality of word pairs that appear in a predetermined order from the document data. More specifically, the information providing apparatus 10 specifies a sentence having a structure used when explaining ideas and features of the invention, and extracts a word set included in the specified sentence.

  To explain a specific example, for example, a sentence that will explain an idea or a feature of an invention, that is, a feature of an idea, is a noun indicating the object of processing, a state expression word indicating the state of the object indicated by the noun, or an adverbial phrase And a verb indicating the content of the process. Also, in such a sentence, a part of speech that can exist between words such as particles is enclosed in curly brackets, and "(noun) {"} (state expression word, adverb phrase) {, by}} (verb It is predicted to have a structure such as {}, can}. Hereinafter, the structure of such a syntax may be described as the cε dictionary method.

  Therefore, the information providing apparatus 10 extracts the sentence having the structure described above from the literature data of the specified field, and three words included in the extracted sentence, that is, a noun, a state expression word or an adverb phrase, and a verb, Extracted as a word set indicating an idea. Thus, the information providing apparatus 10 extracts a specific syntax from input information, and extracts a word group embedded in the extracted specific syntax. Here, the specific syntax is a syntax having a predetermined pattern, that is, the structure of the cε dictionary method. Since the word set extracted in this way can reproduce a sentence indicating an idea such as an invention or an idea only by adding a predetermined case particle, a concept indicated by a sentence serving as an extraction source of the word set, That is, the concept of an idea such as an invention or an idea can be shown.

  Here, it is predicted that document data belonging to a certain field contains many word pairs indicating concepts that are commonly used in the field. For example, it is predicted that the literature data in the program field will contain many sentences with the concept of “realizing” an arbitrary idea by “creating” a “program”. Therefore, among word pairs extracted from document data in the program field, word pairs including words such as "program, create, realize", etc. may contain a concept that can cause serendipity of the user. Absent.

  Therefore, the information providing apparatus 10 selects a word group that can cause serendipity from the extracted word group (step S5). That is, the information providing apparatus 10 selects, from among the word pairs, a word pair that is a concept that is more difficult for the user to conceive, and that indicates a concept that is easy to understand when presented. Specifically, the information providing apparatus 10 can generate serendipity based on the number of word pairs indicating the same concept among the other extracted word sets and the relationship between words included in the word set. Choose a set.

  For example, the information providing apparatus 10 selects a word set whose number of word sets showing the same concept is equal to or less than a predetermined number, in order to select a word set showing a concept which is more difficult for the user to think. Further, the information providing apparatus 10 selects a word set indicating a concept that is easy to understand when presented, so that, among the selected word sets, a word set in which words included in the word set are related words, When a related word is traced from a word contained in a word set, a word set is selected in which the number of related words traced until reaching another word contained in the same word set is equal to or less than a predetermined threshold.

  Here, even if it is a word set that can cause serendipity, a word set that is used across a plurality of documents is a word set that is easy for a user to predict, and therefore, the probability of causing serendipity is low. Therefore, the information providing apparatus 10 removes the word group used across the selected word group (step S6).

  For example, the information providing apparatus 10 calculates a sentence frequency (DF: Document Frequency) which is a frequency at which the selected word set is included in each document. To explain a more specific example, the information providing apparatus 10 calculates a value obtained by dividing the number of documents in which the word group appears for each word group by the number of all documents. Then, the information providing apparatus 10 selects a word set whose calculated value is equal to or less than a predetermined threshold. The documents used when calculating the sentence frequency may be limited to the documents of the field specified in step S3, and may not be limited.

  Here, when the sentence frequency is set to a value lower than the predetermined threshold, the number of selected word sets is reduced. Therefore, the information providing apparatus 10 extends the selected word pair using a verb (hereinafter referred to as an operational verb) indicating various operations performed on the object indicated by the noun (step S7). . That is, the information providing apparatus 10 generates a new set in which a plurality of pieces of information indicating a predetermined operation is combined with respect to a set including a plurality of pieces of information having a predetermined relevance, thereby providing a plurality of pieces of predetermined relevance Extends the concept of a set containing information on

  For example, the information providing device 10 stores, in advance, Kobark & Bakunar's operation verbs. Here, Koverk & Bakunar operational verbs are, for example, “increase”, “divide”, “remove”, “relieve”, “reverse”, “disconnect”, “replace”, “integrate” , “Twist”, “rotate”, “smooth out”, “squeeze”, “supplement”, “submerge”, “freeze”, “soften”, “inflate”, “detour” "Do", "add", "subtract", "lighten", "repeat", "thicken", "stretch", "push out", "splash", "defend", "pull away", These verbs are "Integrate", "Symbolize", "Abstract", and "Disconnect".

  Subsequently, the information providing apparatus 10 newly generates a word set in which the verb included in the selected word set is replaced with each of the above-described operational verbs. For example, the information providing device 10 newly generates a word set such as “glasses, wear, increase”, “glasses, wear, divide” and the like from the word set such as “glasses, wear, viewing”. As a result, the information providing apparatus 10 can generate a variation of a concept that is highly likely to cause serendipity to the user.

  Subsequently, the information providing apparatus 10 maps, for each of the word sets, a vector indicating the word set on a space in which the words in various fields are indicated by a vector, that is, on the distributed representation space including the distributedly represented words. . Then, the information providing apparatus 10 extracts, from the distributed representation space, words that are words belonging to a field different from the field to which the word set belongs and that show the same concept as the concept indicated by the word set (step S8).

  For example, the information providing apparatus 10 uses W2V (word 2vec) to store in advance a distributed representation space including distributed representations of words belonging to various fields. In such a distributed expression space, distributed expressions of words included in documents belonging to the field specified in step S3 are also included. Subsequently, using W2V, the information providing device 10 converts each word included in the word set into a distributed expression, and calculates the sum of the distributed expression of each word. That is, the information providing device 10 calculates the sum of vectors indicating each word included in the word set.

  Here, in the distributed representation space, it is considered that the direction indicated by the sum of vectors indicating each word included in the word set (hereinafter referred to as a vector of the word set) suggests the concept indicated by the word set. Be For this reason, it is considered that the word indicated by the vector whose direction is similar to that of the word set vector has a concept similar to the concept indicated by the word set. However, even if the word has a concept similar to the concept indicated by the word set, the word in the same field as the field to which the word set belongs is a word that can be predicted by the user and is less likely to cause serendipity.

  Therefore, the information providing apparatus 10 extracts words that have a concept similar to the concept indicated by the word set and are included in documents in a field different from the field to which the word set belongs. More specifically, the information providing apparatus 10 is a vector whose direction is similar to the vector of the word set, and which is indicated by the vector at a distance from the vector of the word set in the distributed representation space, that is, the word set It is a word showing a concept similar to the concept shown by, and a word belonging to a different field from the word set is extracted.

  For example, as the directions of two vectors are aligned, the value of the cosine distance between the two vectors increases. Also, as the cosine distance between a certain composite vector and another vector increases, the value of the cosine distance between each word vector constituting the composite vector and the other vector decreases. Therefore, the information providing apparatus 10 calculates the cosine distance between the word set vector and each vector in the distributed expression space. Also, the information providing device 10 calculates the cosine distance between the vector of each word included in the word set and each vector in the distributed representation space.

  Then, in the information providing apparatus 10, the value of the cosine distance with the vector of the word set is larger than the predetermined threshold, and the sum of the cosine distances with the vector of each word included in the word set is smaller than the predetermined threshold. The vector is extracted, and the word indicated by the extracted vector is identified. That is, the information providing apparatus 10 extracts words having a concept similar to the concept indicated by the word set, and that is a word in a different field from the field to which the word set belongs. In other words, based on the equivalence conversion theory, the information providing device 10 extracts words having an analogy similar to the word group extracted from the specific syntax and belonging to a field different from the extracted word group.

  Then, the information providing device 10 outputs the extracted word as a proposal (step S9). For example, when the word “rotation” is extracted by the process shown in step S8 for the word set “graph, collective, add”, the information providing apparatus 10 reads “Rotate?” A sentence is generated that includes the extracted word and that suggests the concept indicated by the extracted word, and the user makes a proposal by causing the robot or the like to read out the generated sentence as the utterance C.

  When such a proposal is made, the user is highly likely to generate new inspiration based on the concept of the proposed word. For this reason, the information providing apparatus 10 can cause the serendipity of the user.

[2. Variations on concepts provided by information providing devices)
Here, in the example illustrated in FIG. 1, the process of proposing a word indicating a concept that causes the user to generate serendipity from the words included in the user's utterance has been described. For example, FIG. 3 is a diagram for explaining a first example of variations of input and output of the information providing apparatus according to the embodiment. For example, as shown in (A) in FIG. 3, the information providing apparatus 10 uses the word W2V to map words included in patent specifications such as published gazettes in a distributed representation space as a document, thereby enabling old technology and different fields. Create a distributed representation space that contains a common analogy of technology.

  Subsequently, as shown in FIG. 3B, the information providing apparatus 10 specifies a service and a technical field from the input information. And the information provision apparatus 10 is close to the concept which the analogy in the specified field | area has among the analogies of the field | area which is different from the specified field | area by performing the extraction process mentioned above as shown to (C) in FIG. Extract analogies with concepts. As a result, the information providing apparatus 10 can output a hint of a new idea as shown in (D) of FIG.

  The hint output in this way is a concept similar to the concept possessed by the idea in the field to which the input information belongs, and it is predicted that the user can recall the concept in a different field. For this reason, the information providing apparatus 10 utilizes an idea obtained by developing the idea of the old technology in a new environment as shown in (E) of FIG. 3 or a technology in a different field as shown in (F) of FIG. Help create new technologies.

  However, the embodiment is not limited to this. For example, the information providing apparatus 10 uses a distributed expression space in which the contents of the business model and the contents of the contract in various types of business are replaced with the distributed expression, using the business model and the field of business as input instead of words. It may also output information that gives the user a hint of serendipity to create a new business model.

  For example, FIG. 4 is a diagram illustrating a second example of input and output variations of the information providing apparatus according to the embodiment. For example, as shown in FIG. 4 (A), the information providing apparatus 10 maps, as a document, features of past business to a distributed representation space by W2V, so that a common analogy of old business and different field technologies can be obtained. Create a distributed representation space that contains

  Subsequently, as shown in FIG. 4B, the information providing apparatus 10 specifies the type of business to which the service or business belongs from the input information. Then, the information providing apparatus 10 extracts the features of the service or business of the identified field, and converts the extracted plurality of features into a distributed representation. In addition, the information providing apparatus 10 extracts the business feature having a feature similar to the sum of the features converted to the distributed representation from the business in the other field, as shown in (C) in FIG. And an analogy having a concept similar to the concept possessed by the analogy in the specified type of business, among the analogies of different types of business. As a result, the information providing apparatus 10 can output a new business hint as shown in (D) of FIG.

  The hint output in this way is a concept similar to the business concept in the industry to which the input information belongs, and it is predicted that the user can recall concepts in different industry. Therefore, the information providing apparatus 10, as shown in (E) in FIG. 4, is a business in which the idea of the old business is developed in a new environment, and as shown in (F) in FIG. Help create new business models that you use.

  Further, for example, the information providing apparatus 10 uses serendipity for creating new content by using arbitrary content such as music, images, advertisements, and the like instead of words as input and using a distributed expression space of the content. You may output content that causes people to wake you up.

  For example, FIG. 5 is a diagram illustrating a third example of input and output variations of the information providing apparatus according to the embodiment. For example, as shown in (A) in FIG. 5, the information providing apparatus 10 maps the features of the past content to the distributed representation space instead of the literature, thereby making the common analogy of old content and content in different fields. Create a distributed representation space that contains

  Subsequently, as shown in FIG. 5B, when the information providing apparatus 10 receives content such as sound or image as input information, the information providing apparatus 10 identifies the field to which the content belongs. Further, the information providing device 10 extracts a plurality of features included in the content of the specified field, and converts the plurality of extracted features into a distributed representation. In addition, as shown in (C) in FIG. 5, the information providing apparatus 10 extracts content similar to the sum of the features converted to the distributed representation from content in other fields, so that the information providing apparatus 10 is in a different field from the identified field. Among the analogies possessed by the content, an analogy having a concept close to the concept possessed by the content analogy in the specified field is extracted. As a result, the information providing apparatus 10 can output a hint of new content as shown in (D) in FIG.

  The hint output in this way is a concept similar to the concept possessed by the content of the field to which the input information belongs, and it is predicted that the user can be reminded of the concept of content of a different field. For this reason, the information providing apparatus 10 utilizes the content in which the idea of old content is developed in a new environment as shown in (E) in FIG. 5 and the content in different fields as shown in (F) in FIG. Help create new content. In addition, as shown in (G) in FIG. 5, for example, when the output is a sound, the information providing device 10 can help, for example, creation of a phrase that is a source of composition.

  Further, for example, the information providing apparatus 10 generates serendipity to the user by using, as a input, biological information such as five senses acquired from the user instead of the word and using a distributed expression space in which various biological information is decentralized and expressed. It may also output information to give hints.

  For example, FIG. 6 is a diagram illustrating a fourth example of input and output variations of the information providing apparatus according to the embodiment. For example, as shown in (A) in FIG. 6, the information providing device 10 maps the feature of the biological information acquired in the past to the distributed representation space in place of the document, thereby separating the old biological information and the different types of information. A distributed representation space including a common analogy (for example, a pattern) possessed by the biological information of

  Subsequently, as shown in (B) of FIG. 6, the information providing apparatus 10 specifies the type of biometric information of the user from the input information. In addition, the information providing apparatus 10 extracts features of a plurality of biological information of the specified type, and converts the extracted features into a distributed representation. In addition, the information providing apparatus 10 extracts the biological information having a feature similar to the sum of the features converted to the distributed representation from the biological information of the other type, thereby identifying the type specified as shown in (C) in FIG. The analogy having a concept close to the concept of the analogy of the specified type of biological information is extracted from the analogies of the different types of biological information. As a result, the information providing apparatus 10 can output a hint for causing new biological information as shown in (D) in FIG.

  The hint output in this way can be used, for example, to identify a comfortable sensation for the user who is the source of the biological information, as shown in (F) in FIG. For example, when the biological information acquired from the user is a taste that the information providing device 10 prefers, the information providing device 10 can output a color, a pitch, and the like that the user likes.

  In addition, in order to execute the process described above, it is necessary to map content other than words on the distributed representation space. Therefore, the information providing apparatus 10 extracts features of contract documents, business books, various contents, and biological information using, for example, a method such as neural network or deep learning, and distributedly represents the vectors indicating the extracted features. It may be mapped on space. That is, the information providing apparatus 10 applies the above-described extraction process to arbitrary information as long as the information is information that can map features on the distributed representation space and can be classified into a plurality of types. be able to.

[3. Configuration of Information Providing Device]
Next, the configuration of the information providing apparatus 10 that executes the extraction process shown in FIG. 1 will be described using FIG. 7. In the following description, the words contained in a plurality of documents are mapped on the distributed expression space, and the words in the field different from the field of technology to which the user's speech belongs, and in the field to which the user's speech belongs An example of the information providing apparatus 10 that outputs a word having a concept similar to the concept of an idea as a hint will be described.

  FIG. 7 is a diagram illustrating an example of a functional configuration of the information providing apparatus according to the embodiment. As shown in FIG. 7, the information providing device 10 is connected to the input device 30 and the output device 31. The information providing device 10 further includes a communication unit 11, a storage unit 12, and a control unit 16.

  The communication unit 11 is realized by, for example, a network interface card (NIC). The communication unit 11 is connected to an input device 30 such as a microphone and a keyboard, and an output device 31 such as a monitor and a printer, and a robot that can utter voice, and transmits and receives various information.

  The storage unit 12 is realized by, for example, a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 12 also includes a document database 13, a distributed expression space database 14, and an expanded word database 15.

  Documents belonging to various fields are registered in the document database 13. For example, in the document database 13, documents including concepts of ideas made in the past, such as published patent publications, published patent publications, patent publications, and papers, are registered.

  In the distributed expression space database 14, distributed expressions of words belonging to various fields are registered. For example, in the distributed expression space database 14, vectors indicating words included in documents registered in the document database 13 are registered. In addition, since the distributed expression of each word is generated based on the relationship between each word, the direction or distance between each word corresponds to the similarity of the concept or co-occurrence of each word. Become.

  In the expanded word database 15, operation verbs used when expanding a word set are registered. For example, in the extended word database 15, an operational verb list including the Kobark & Baknar operational verbs is registered in advance.

  The control unit 16 is realized by, for example, a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like executing various programs stored in the storage device inside the information providing apparatus 10 using the RAM as a work area. Be done. Moreover, the control part 16 is implement | achieved by integrated circuits, such as ASIC (Application Specific Integrated Circuit) and FPGA (Field Programmable Gate Array), for example.

  As shown in FIG. 7, the control unit 16 includes a reception unit 17, a field specification unit 18, a pattern extraction unit 19, a related word extraction unit 20, a word space limitation unit 21, a word space expansion unit 22, and a suggested word extraction unit 23. An output unit 24 and a learning unit 25 are included.

  The receiving unit 17 receives the user's speech as input information. For example, the reception unit 17 acquires the user's utterance from the input device 30 realized by a microphone, a keyboard, or the like. In such a case, the reception unit 17 converts the received statement of the user into text data. Then, the reception unit 17 outputs the converted text data to the field identification unit 18.

  The field identification unit 18 identifies the field to which the input information belongs. For example, the field identification unit 18 performs morphological analysis of the text data received from the reception unit 17 and identifies in which field the word included in the text data is a word used. Then, the field identification unit 18 notifies the pattern extraction unit 19 of the identified field.

  The pattern extraction unit 19 extracts a word set including a plurality of words having a predetermined relationship in the identified field. For example, when receiving the notification of the field from the field identification unit 18, the pattern extraction unit 19 extracts documents belonging to the notified field from the document database 13. Then, the pattern extraction unit 19 specifies a sentence having a predetermined structure from the extracted document, extracts a plurality of words contained in a predetermined order in the specified sentence, and selects a word set including the extracted plurality of words. Output to the related word extraction unit 20.

  For example, FIG. 8 is a diagram for explaining an example of a word extracted by the information providing apparatus according to the embodiment. As shown in (A) of FIG. 8, the sentence indicating the feature of the idea is “(noun) {}} (state expression word, adverb phrase) {with, with} (verb) {perform, can} It is predicted to have a structure of “ Therefore, the pattern extraction unit 19 extracts documents belonging to the notified field from the document database 13, and using a technique such as morphological analysis, “(noun) {}} (state expression word, adverb phrase) { Extract sentences with the structure {}, {can, do}} according to. Then, as shown in (B) in FIG. 8, the pattern extraction unit 19 sets the noun included in the extracted sentence as word # 1, sets the state expression word or adverbial phrase as word # 2, and sets the verb as word # 3. A word set including extracted and extracted words # 1 to # 3 is generated. The word set generated in this manner includes the feature of the idea indicated by the extracted sentence, that is, the concept cε, as shown in (C) in FIG.

  Returning to FIG. 7, the description will be continued. The related word extraction unit 20 selects, from among the word pairs extracted by the pattern extraction unit 19, a word combination that is a concept that is more difficult for the user to conceive and that is easy to understand when presented. For example, FIG. 9 is a diagram for explaining an example of a word set extracted by the information providing apparatus according to the embodiment. In the example shown in FIG. 9, each document is subjected to morphological analysis and a word set extracted from a sentence having a predetermined structure, that is, a word set including a plurality of words having a predetermined relationship, each document Listed in the order of appearance.

  For example, as shown in FIG. 9A, the document contains many word pairs such as “program creation and realization” and “trellis structure degeneration calculation”. However, since such a word set is a word set which is often used in a specified field or a word set having a sense of stability, the concept of the idea which such a word set has is commonly used in the field. Concept. As a result, the frequently appearing word set has low serendipity.

  On the other hand, as shown in (B) in FIG. 9, the word set such as "wearing glasses and watching" is not very frequently used in the specified field, and thus it is a word set showing an uncommon concept in the specified field, It is predicted to be a word set that is likely to cause serendipity. In addition, as shown in FIG. 9C, a word set with an appearance frequency that is too low, that is, a word set with an appearance frequency lower than a predetermined threshold, is difficult to understand, and causes serendipity to the user. The possibility of making it low.

  Therefore, the related word extraction unit 20 is configured such that, among the word pairs extracted by the pattern extraction unit 19, the number of word pairs showing the same concept is equal to or less than the first threshold, and the second threshold in the documents belonging to the identified field. Extract word pairs that are above. As a result, the related word extraction unit 20 can extract a word set including a concept that is likely to cause serendipity in the specified field, that is, the field to which the user's utterance belongs.

  Note that the related word extraction unit 20 uses a related word dictionary that indicates a related relation between words, and indicates a relation existing between any word included in the word set and another word included in the same word set. Count the number of words. Then, the related word extraction unit 20 may extract a word set including a concept that is likely to cause serendipity by extracting a word set in which the number of counted related words falls within a predetermined range.

  Returning to FIG. 7, the description will be continued. The word space limiting unit 21 further limits the distributed expression space formed by each word included in the word set extracted by the related word extracting unit 20. Specifically, the word space limitation unit 21 counts, among all the documents stored in the document database 13, the number of documents including sentences including the word set extracted by the related word extraction unit 20. Then, the word space limiting unit 21 calculates a value obtained by dividing the counted value by the number of documents, that is, a DF of the word set, and extracts a word set whose calculated DF value is equal to or less than a predetermined threshold. As a result, the information providing apparatus 10 can extract a word set including a concept that is more likely to cause serendipity based on all fields.

  The word space expansion unit 22 generates a plurality of word sets in which verbs included in the word set are replaced with predetermined operational verbs. As a result, the word space expanding unit 22 expands the distributed expression space limited by the word space limiting unit 21 without largely changing the concept included in the word set.

  For example, FIG. 10 is a diagram for explaining an example of a process of extending the distributed presentation space performed by the information providing apparatus according to the embodiment. For example, the word space expanding unit 22 acquires a word set such as “icon click display” as the word set extracted by the word space limiting unit 21. In such a case, as shown in (A) of FIG. 10, the word space expansion unit 22 registers the verb “display” included in the word set in the expanded word database 15 as shown in (B) of FIG. It generates word pairs replaced with each word in the operation verb list.

  As a result, for example, as shown in (C) in FIG. 10, the word space expanding unit 22 newly generates word pairs such as “Icon click increase”, “Icon click division”, and “Icon click removal”. The word space expansion unit 22 also outputs the word pair “icon click display” as the original word to the proposal word extraction unit 23 as a newly generated word pair.

  As described above, the word space restriction unit 21 and the word space expansion unit 22 calculate a value obtained by dividing the number of sentences including the word pair extracted by the related word extraction unit 20 by the number of sentences in all fields. A new word pair is generated by combining a word pair whose calculated value is smaller than a predetermined threshold value and an operation verb. Since the newly generated word set is a word set obtained by converting the verb of the original word set, when mapped on the distributed expression space, the vector will indicate a direction similar to the original word set. Although predicted, such vectors are predicted to indicate similar concepts. As a result, even when the number of extracted word pairs is small, the information providing apparatus 10 includes a word pair including a concept similar to the word pair extracted by the word space limiting unit 21, that is, a concept that easily causes serendipity. Can be generated.

  In addition, when the information providing apparatus 10 targets information other than words, the word space expansion unit 22 indicates a predetermined operation for a set of information having predetermined relevancy similarly to the operation verb. Information may be combined to generate a plurality of new information sets. In addition, the word space expansion unit 22 may generate, for example, a set obtained by changing information of a predetermined type into information indicating a predetermined operation among pieces of information included in a set of information having predetermined relevance.

  Returning to FIG. 7, the description will be continued. The proposed word extraction unit 23 extracts words having the same concept as the concept generated by the extracted word set from the words belonging to the field different from the identified field. For example, the suggested word extraction unit 23 inputs each word included in the word set generated by the word space expansion unit 22 to W2V, and acquires a distributed representation of each word. Subsequently, the proposal word extraction unit 23 calculates the sum of the acquired distributed expressions, that is, the distributed expression of the word set. Then, the proposal word extraction unit 23 is a distributed expression of the direction similar to the distributed expression of the word pair among the distributed expressions stored in the distributed expression space database 14, and the distance from the distributed expression of the word pair is separated Identify the distributed expressions that are present. That is, the proposed word extraction unit 23 uses the distributed representation of the words included in the word set to set a word having a concept whose similarity with the concept generated by the word set satisfies a predetermined condition with the field to which the input information belongs. Extract from words that belong to different fields. More specifically, the proposal word extraction unit 23 extracts a word corresponding to a distributed expression in which the degree of similarity with the distributed expression of the word set is equal to or less than a predetermined threshold.

  For example, FIG. 11 is a diagram for explaining the limitation and expansion of the distributed representation space executed by the information providing apparatus according to the embodiment. For example, as shown in (A) of FIG. 11, the related word extraction unit 20 extracts word pairs # 1 to #N as a word pair including a concept that is likely to cause serendipity. In such a case, as shown in (B) of FIG. 11, the word space limiting unit 21 calculates and calculates the DF value of each of the word pairs # 1 to #N for all the documents stored in the document database 13. Perform word set filtering with the value of DF. As a result, as shown in (C) in FIG. 11, the word space limiting unit 21 extracts, for example, word pairs # 1, # 3, #N, etc., as word pairs whose DF value is less than or equal to a predetermined threshold. Do.

  In addition, as shown in (D) in FIG. 11, the word space expansion unit 22 expands the word pair using the Kovark & Baknar check list. For example, the word space expansion unit 22 generates word pairs # 1-1 to # 1-m in which verbs included in the word pair # 1 are replaced with operation verbs included in the Kovark & Baknar check list. Similarly, the word space expansion unit 22 generates word pairs # 3-1 to # 3-m in which the verb included in word pair # 3 is replaced with an operation verb, and operates the verbs included in word pair #N. Word pairs # N-1 to # N-m replaced with the target verb are generated.

  Subsequently, as shown in (E) of FIG. 11, the suggested word extraction unit 23 selects each word pair # 1-1 to # 1-m, # 3-1 to # 3-m, and # N-1 to ##. The following processing is executed for each N−m. That is, the proposal word extraction unit 23 inputs the word included in the word set into W2V to acquire a vector, and is a word similar to the sum of the acquired vectors and is a field to which the user's utterance belongs, that is, identification The words of the field different from the field which has been selected are acquired from the distributed representation space.

  Hereinafter, an example of a process in which the proposed word extraction unit 23 acquires a word in a field that is similar to the sum of vectors of word pairs and is different from the identified field from the distributed representation space will be described using the drawings. For example, FIG. 12 is a diagram for explaining an example of processing of extracting similar words by the information providing apparatus according to the embodiment. For example, as shown in FIG. 12A, the suggested word extraction unit 23 acquires the word set “view wearing glasses”. In such a case, as shown in (B) in FIG. 12, the suggested word extraction unit 23 inputs each word included in the word set, that is, “glasses”, “wear”, and “view” into W2V. Compute a distributed representation of each word, ie a vector.

  Subsequently, as shown in (C) in FIG. 12, the suggested word extraction unit 23 extracts a vector similar to the sum of the vectors of the respective words of the word set from the distributed representation space database 14, and the extracted vector indicates Get the word. For example, as shown in FIG. 12D, the suggested word extraction unit 23 acquires a word indicated by a vector having a large cosine distance value from the word set vector.

  For example, in the example shown in FIG. 12D, “age”, “spouse”, “appearance”, “acting”, “optimization”, “moderator”, “real estate_owned”, etc., in descending order of the cosine distance value. Words such as are obtained. Here, “real estate_owned” is a word extracted as one word by morphological analysis when the word “real estate” and the word “owned” are used as one idiom by being continuous.

  Here, it can be said that a word having a large cosine distance value is a word having a concept similar to the concept indicated by the word set. However, a word having a large cosine distance simply does not provide a concept that the user did not predict, and therefore, the serendipity is predicted to be low. On the other hand, a word with a small value of cosine distance includes a word having a concept different from the concept indicated by the word set, but a word providing a concept that the user does not think at first glance, ie, a high serendipity It is predicted that words will also be included.

  For example, FIG. 13 is a diagram for explaining an example of the relationship between word vectors. In the example shown in FIG. 13, an example of the distributed representation space to which the words "MAN", "WOMAN", "UNCLE", "AUNT", "KING", and "QUEEN" are mapped is described. Here, as shown in FIG. 13A, a vector from the word “MAN” to the word “WOMAN”, a vector from the word “UNCLE” to the word “AUNT”, and from the word “KING” to the word “QUEEN” These vectors are considered to indicate the same concept, such as “add the concept of a woman excluding the concept of a man from the word”. For this reason, as shown in (B) in FIG. 13, the direction in which the arrows of such vectors point is the same.

  On the other hand, when the distance between the word “KING” and the word “UNCLE” is long, the word “UNCLE” cannot always be easily recalled from the word “KING”. The co-occurrence between the concept indicated by the word "KING" and the concept indicated by the word "UNCLE" is considered to be small. However, even in such a concept with low co-occurrence, a similar concept can be shown when the vector direction is facing, so that serendipity can occur as shown in FIG. It is considered to be highly sexual.

  Therefore, the information providing apparatus 10 is a word having a concept similar to a concept possessed by an idea having high co-occurrence with a word included in the user's utterance as a word having a high possibility of causing serendipity to the user. The words contained in the user's remarks should be provided with words having low co-occurrence. Here, a word having low co-occurrence is a word in which the distance between vectors is large, and it is predicted that, for example, a word or the like belonging to a different field corresponds.

  Therefore, the proposed word extraction unit 23 calculates the distance between the vector of the word set and the vector indicating another word, and identifies the vector whose calculated distance is larger than a predetermined threshold. The word identified in this way is a word including a concept similar to the concept of an idea having high co-occurrence with the user's utterance, and is predicted to be a word having low co-occurrence with the user's utterance. Be done.

  For example, FIG. 14 is a figure for demonstrating an example of the concept which the word which an information provision apparatus proposes has. As shown in FIG. 14, words # 1 to # 3 constituting word pair # 1 belonging to field # 1 have concepts # 1 to # 3, respectively. Here, the concept # 1 is formed by an element group # 1 which is a plurality of elements forming a concept, the concept # 2 is formed by an element group # 2 which is a plurality of elements forming a concept, and the concept # 3 Is formed by an element group # 3 which is a plurality of elements forming a concept. Therefore, the concept of word pair # 1 is considered to be the sum of concepts # 1 to # 3, that is, the sum of element groups # 1 to # 3.

  Here, as shown in FIG. 14A, the possibility of causing serendipity increases as the vector distance increases. However, if the distance is simply separated, the concept is too far from the word set # 1, so serendipity is less likely to occur. Therefore, as shown in (B) in FIG. 14, the suggested word extraction unit 23 is a word having a portion common to the word pair # 1 to such an extent that it can be guaranteed that there is a meaning to the extent that serendipity occurs. Extract More specifically, the proposal word extraction unit 23 is a word having a concept # 4 formed by an element group # 4 similar to the sum of the element groups # 1 to # 3 that the word group # 1 has. It extracts word # 4 belonging to field # 2 which is sufficiently separated from field # 1.

  For example, the suggested word extraction unit 23 extracts a vector whose cosine distance with the vector of the word set is larger than a predetermined threshold (or falls within a predetermined range), and the extracted vector and each of the words included in the word set Calculate the cosine distance of the word vector respectively. Then, the suggested word extraction unit 23 specifies a vector in which the sum of the calculated cosine distances is smaller than a predetermined threshold, and sets the word indicated by the specified vector as a suggested word. That is, the proposed word extraction unit 23 is a distributed expression in which the cosine distance with the sum of the distributed expressions of the word set is larger than a predetermined threshold and the sum of the cosine distances with the distributed expressions of each word is smaller than the predetermined threshold. Extract the word corresponding to.

  As described above, the suggested word extraction unit 23 calculates the sum of the distributed expressions of each word included in the word set, and the word corresponding to the distributed expression whose similarity with the calculated distributed expression sum satisfies a predetermined condition Extract. For example, the proposed word extraction unit 23 is a distributed expression having the same direction as the sum of the distributed expressions of the words contained in the word set (ie, a vector of the word set), and is a distributed expression of the words contained in the word set. A word corresponding to a distributed expression whose distance to the sum is equal to or greater than a predetermined threshold is extracted.

  For example, when the proposed word extracting unit 23 receives a word set such as “graph collective addition”, it calculates the sum of the word “graph” vector, the word “collective” vector, and the word “add” vector. . Next, the proposal word extraction unit 23 specifies the vector of the word “rotation” as a vector whose cosine distance with the calculated sum of vectors is larger than a predetermined threshold. In this case, the suggested word extraction unit 23 calculates the cosine distance between the vector of the word "rotation" and the vector of the word "graph", the cosine distance between the vector of the word "rotation" and the vector of the word "collective", the word "rotation" The cosine distance between each vector and the vector “add” is calculated, and it is determined whether or not the sum of the calculated cosine distances is smaller than a predetermined threshold.

  That is, the proposal word extraction unit 23 determines whether the distance to each word vector constituting the word set vector is smaller than a predetermined threshold. Then, when it is determined that the distance between the word “rotation” vector and each word vector forming the word group vector is smaller than a predetermined threshold, the proposal word extraction unit 23 proposes the word “rotation”. It is a word.

  Returning to FIG. 7, the description will be continued. The output unit 24 outputs the word extracted by the proposal word extraction unit 23 as a proposal. For example, when the suggested word extraction unit 23 extracts the word "rotation", the output unit 24 generates a sentence to make a proposal such as "how to rotate" and transmits the generated sentence to the output device 31 Do. As a result, the output device 31 can output a word extracted by the information providing device 10, that is, a word that causes a serendipity to the user.

  The learning unit 25 performs learning of the distributed expression space stored in the distributed expression space database 14. For example, the learning unit 25 performs morphological analysis of each document included in the document database 13, and uses W2V technology to learn a distributed expression based on the relationship between words included in each document. Then, the learning unit 25 registers the learning result in the distributed expression space database 14.

[4. Flow of processing executed by information providing apparatus 10]
Next, the flow of the process performed by the information providing apparatus 10 will be described with reference to FIGS. First, the flow of the extraction process performed by the information providing device 10 will be described with reference to FIG. FIG. 15 is a flowchart illustrating the flow of the extraction process performed by the information providing apparatus according to the embodiment.

  As illustrated in FIG. 15, when the information providing apparatus 10 acquires input information (step S101), the information providing apparatus 10 specifies a field to which the acquired input information belongs (step S102). Subsequently, the information providing apparatus 10 extracts a word set satisfying a predetermined relationship from the document data belonging to the identified field (step S103). That is, the information providing device 10 extracts a word set from the specific syntax having the structure of the cε dictionary method. Then, the information providing apparatus 10 selects a word set that can cause serendipity from the extracted word set (step S104), and using the distributed expression of the word set, the concept and the similarity caused by the word set satisfy a predetermined condition A word is extracted (step S105). That is, the information providing device 10 extracts words that belong to a field other than the field to which the word set belongs and that has a concept similar to the concept that the word set has. Then, the information providing device 10 outputs the extracted word as a hint (step S106), and ends the process.

  Next, using FIG. 16, the flow of various processes for extracting words that belong to a field other than the field to which the word set belongs and that has a concept similar to the concept that the word set has is more specifically explain. FIG. 16 is a flowchart illustrating the flow of a specific process of the extraction process performed by the information providing apparatus according to the embodiment. Steps S201 to S206 shown in FIG. 16 are more specific examples of the processes of steps S103 to S105 shown in FIG.

  For example, the information providing apparatus 10 performs morphological analysis of document data belonging to the field to which the input information belongs (step S201), and specifies a character string that matches a predetermined pattern (step S202). Then, the information providing device 10 generates a word set from the words included in the specified character string (step S203).

  Further, the information providing apparatus 10 selects a word group that is likely to cause serendipity from the generated word group (step S204), and generates a variation of the word group from the selected word group using an operational verb (step S205). Then, the information providing apparatus 10 inputs the selected word set into W2V, and the similarity including the direction, that is, the similarity not including the direction but having the cosine distance with the vector of the word set within the predetermined range. That is, a word whose sum of cosine distances with the vector of each word included in the word set is equal to or less than a predetermined threshold is extracted (step S206).

  Next, with reference to FIG. 17, a flow of processing for selecting a word set likely to cause serendipity from among word sets and generating variations of the word set from the selected word set will be more specifically described. FIG. 17 is a flowchart for explaining the flow of a specific process of the process in which the information providing apparatus according to the embodiment selects a word set which is likely to cause serendipity. Steps S301 to S303 shown in FIG. 17 are more specific examples of the processes of steps S204 and S205 shown in FIG.

  For example, the information providing device 10 calculates DF of the generated word set (step S301). Further, the information providing apparatus 10 selects a word pair in which the value of the DF of the generated word pair is less than or equal to a predetermined threshold (step S302). Then, the information providing device 10 generates a variation of the selected word set using the operational verb list (step S303).

[5. (Modification)
In the above, an example of the extraction process performed by the information providing apparatus 10 has been described using the aspect illustrated in FIG. 1. However, the embodiments are not limited to this. Hereinafter, variations of the extraction process performed by the information providing apparatus 10 will be described.

[5-1. (Various parameters)
In order to extract words that can cause serendipity, the information providing apparatus 10 described above, for example, selects, from among the extracted word pairs, word pairs whose DF value is a predetermined threshold (hereinafter, referred to as a first threshold). Extracted. In addition, the information providing device 10 is a vector whose cosine distance with the vector of the word set is within a predetermined range (hereinafter referred to as a first range), and with the vector of each word included in the word set. A vector whose sum of cosine distances is equal to or less than a predetermined threshold (hereinafter referred to as a second threshold) was extracted. Here, as the various threshold values adopted by the information providing apparatus 10, arbitrary threshold values can be adopted.

  For example, when the information providing apparatus 10 outputs a word that is difficult to understand at first glance but is prone to serendipity when considered, the value of the first threshold may be set lower. May be smaller. In addition, the information providing device 10 may set the first range narrower.

[5-2. (About word sets that are prone to serendipity
In the above-described example, the information providing apparatus 10 selects a word that is highly likely to cause serendipity from the extracted word set in order to increase the possibility of causing serendipity. For example, the related word extraction unit 20 selects a word set based on the appearance frequency of each word set, the relationship between words, and the like. However, the embodiments are not limited to this.

  For example, the information providing device 10 may preselect a word set that can cause serendipity for each field. Also, such selection processing may be performed in advance by human hands. For example, the information providing apparatus 10 extracts a word set including a plurality of words having a predetermined relationship from the document data, and presents the extracted word set to the operator. Then, the information providing device 10 may preselect a word set selected by the operator as a word set that can cause serendipity.

[5-3. Processing performed by information providing device]
In the above description, the information providing apparatus 10 identifies a plurality of pieces of information having a predetermined relationship from the information of the field to which the input information belongs, and identifies the information having the same concept as the concept generated by the identified plurality of pieces of information Extracted from information belonging to a different field. However, the embodiments are not limited to this. That is, the information providing apparatus 10 can explicitly output information having a discontinuous relationship while holding the implicit connection with the input information out of the information in the field different from the field to which the input information belongs. If there is, the information to be output may be extracted by different processes.

  For example, FIG. 18 is a diagram illustrating an example of a process for outputting information related to information in an input field and information in a different field. In the example shown in FIG. 18, as a variation of the process executed by the information providing apparatus 10, both the information of the field to which the input information belongs and the information of the field different from the field to which the input information belongs are analyzed. An example of a process for outputting information relevant to information has been described.

  For example, as shown in FIG. 18, when the user thinks of a new idea, the fields to which the information to be helpful belongs include a wide variety of fields such as education, power supply, and banking. For this reason, as shown in FIG. 18A, it is difficult for the user to consider a new idea in consideration of information in all these fields.

  Therefore, the information providing device 10 executes the following process. First, the information providing device 10 receives input information from the user. In such a case, the information providing apparatus 10 identifies the field to which the input information belongs. In addition, the information providing apparatus 10 extracts a field (different field) different from the identified field. Then, the information providing apparatus 10 analyzes both information belonging to the field to which the input information belongs and information belonging to a field different from the field, and specifies information relevant to both information. Then, the information providing device 10 outputs the specified information.

  For example, as shown in FIG. 18B, the user inputs arbitrary information such as a word that has come up to the information providing apparatus 10 as input information. In such a case, the information providing apparatus 10 identifies the field to which the input information belongs, as shown in (C) in FIG. For example, if the input information is information such as “propane gas”, the information providing apparatus 10 identifies “gas supply industry” as the field to which the input information belongs.

  Subsequently, the information providing apparatus 10 extracts a field different from the identified field. For example, the information providing apparatus 10 extracts the “advertising business” as a field different from the identified field “gas supply industry”. The information providing apparatus 10 may select a plurality of fields different from the fields to which the input information belongs. Further, the information providing apparatus 10 may select a field (for example, a field in which the distance on the distributed representation space is far) having a meaning different from the field to which the input information belongs using the various processes described above. .

  Then, the information providing apparatus 10 analyzes information on both the field to which the input information belongs and the extracted field, and specifies information related to both information. For example, the information providing apparatus 10 compares attributes, concepts, meanings, and the like of words belonging to the respective fields using the distance and direction between the words in the distributed expression space, and specifies information relevant to both the fields. Further, for example, the information providing device 10 extracts words in which the directions of the words in the distributed representation space are close and the distance is long. Then, the information providing device 10 outputs the extracted word as shown in (F) in FIG.

  As a result of performing such processing, the information providing apparatus 10 can output a combination of words that does not float by ordinary thinking, and thus can generate serendipity for the user. In addition to the process described above, any information analysis method may be adopted as long as the information providing apparatus 10 outputs information related to both the field to which the input information belongs and the different field.

  Note that the information providing apparatus 10 described above can adopt any information as input information. For example, the information providing apparatus 10 adopts, as input information, text, business features, contents such as sounds and images, biometric information, etc., and relates to information in both the field to which the input information belongs and different fields. As information, text, business characteristics, content, biometric information, etc. may be output. In addition, the information providing apparatus 10 can be used within a range in which arbitrary processes are not contradicted among the writing processes described in the above-described embodiments.

[5-4. Others]
Further, among the processes described in the above embodiment, all or part of the process described as being automatically performed may be manually performed, or the process described as being manually performed. All or part of them can be performed automatically by known methods. In addition, information including processing procedures, specific names, various data and parameters shown in the above-mentioned documents and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure 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. It can be integrated and configured. For example, the word space limiting unit 21 and the word space expanding unit 22 illustrated in FIG. 7 may be integrated.

  Moreover, it is possible to combine suitably each embodiment mentioned above in the range which does not contradict process content.

[5-5. program〕
Further, the information providing apparatus 10 according to the embodiment described above is realized by, for example, a computer 1000 configured as shown in FIG. FIG. 19 is a hardware configuration diagram showing an example of a computer that implements extraction processing. The computer 1000 includes a CPU 1100, a RAM 1200, a ROM 1300, an HDD 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 starts 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, sends the data to the CPU 1100, and transmits the data generated by the CPU 1100 to the other devices.

  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 the 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 such as an information providing program stored in a recording medium 1800, which is an example of a non-transitory computer readable storage medium, and provides the read program or data 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 providing apparatus 10 according to the embodiment, the CPU 1100 of the computer 1000 realizes the function of the control unit 16 by executing a program loaded on the RAM 1200. The HDD 1400 also stores data in the storage unit 12, that is, a document database 13, a distributed expression space database 14, and an expanded word database 15. The CPU 1100 of the computer 1000 reads these programs from the recording medium 1800 and executes them. However, as another example, these programs may be acquired from other devices.

[6. effect〕
As described above, when receiving the input information, the information providing apparatus 10 specifies a specific syntax from the input information, and extracts a word group embedded in the specified specific syntax. Further, the information providing apparatus 10 calculates the degree of similarity with the distributed expressions of other words using the distributed expression of the word group included in the specific syntax. Then, the information providing device 10 outputs the information extracted based on the calculated degree of similarity. Therefore, the information providing apparatus 10 can provide information that the user does not think of, so that information that can cause the user to generate serendipity can be output.

  Further, the information providing device 10 specifies a specific syntax using the cε dictionary method. Therefore, the information providing apparatus 10 can output, from a word group included in a sentence having a pattern having a possibility of causing serendipity in a certain field, a word having a feature similar to a feature of the word group. Therefore, it is possible to provide information that is hard to come up with than the user while ensuring the possibility of causing serendipity for the user.

  Moreover, the information provision apparatus 10 has the analogy similar to a word group based on the equivalent conversion theory, and outputs the word which belongs to the field | area different from this word group. Since the user can provide information that is less likely to come to mind while keeping the concept generated by the word group, it is possible to output information that can cause the user to generate serendipity.

  Further, the information providing apparatus 10 corresponds to a distributed expression in which the degree of similarity with the word corresponding to the distributed expression similar to the sum of the distributed expressions of the word group and the sum of the distributed expression of the single instrument group falls within a predetermined range. Extract information. Therefore, the information providing apparatus 10 can output information that can cause the user to generate serendipity.

  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.

  Also, the "section (module, unit)" described above can be read as "means" or "circuit". For example, the control unit can be read as control means or a control circuit.

DESCRIPTION OF SYMBOLS 10 Information provision apparatus 11 Communication part 12 Storage part 13 Document database 14 Distributed expression database 15 Expansion word database 16 Control part 17 Reception part 18 Field identification part 19 Pattern extraction part 20 Related word extraction part 21 Word space limitation part 22 Word space expansion part 23 Suggested Word Extraction Unit 24 Output Unit 25 Learning Unit 30 Input Device 31 Output Device

Claims (7)

A reception unit for receiving input information;
A pattern identification unit that extracts, from input information, a type of business, a specific syntax, and a word group embedded in the specific syntax;
And distributed representation of a word extracted from the information of different industries and the industries, and the included in the identified said pattern, using a distributed representation of the word groups, the similarity calculating section for calculating a degree of similarity,
Outputting, among the words extracted from the information of the different type of industry, a word whose similarity calculated by the similarity calculation unit satisfies a predetermined condition as information serving as an analogy of the input information; A characteristic information providing apparatus. The information providing apparatus according to claim 1, wherein the pattern identification unit uses a cε dictionary method. The information providing system according to claim 1 or 2, wherein the output unit outputs a word having an analogy similar to the word group based on equivalence conversion theory and belonging to a field different from the word group. apparatus. The information providing apparatus according to any one of claims 1 to 3, wherein the output unit outputs a word corresponding to a distributed expression similar to a sum of the distributed expressions of word groups. The said output part outputs the word corresponding to the distributed expression in which the similarity degree with the sum of the distributed expression of a word group falls within a predetermined range. Information provision device. An information providing method performed by an information providing apparatus, comprising:
A reception process for receiving input information;
A pattern identification process that extracts the industry type and specific syntax from the input information and a group of words embedded in the specific syntax,
And distributed representation of a word extracted from the information of different industries and the industries, the included in the identified said pattern, using a distributed representation of the word group, a similarity calculation step of calculating the similarity,
Outputting, among the words extracted from the information of the different types of industry, a word whose similarity calculated in the similarity calculation step satisfies a predetermined condition as information that is analogous to the input information; A characteristic information providing method. Acceptance procedure to accept input information,
A pattern identification procedure for extracting a business type and a specific syntax and a word group embedded in the specific syntax from input information;
And distributed representation of a word extracted from the information of different industries and the industries, the included in the identified said pattern, using a distributed representation of the word groups, and similarity calculation step of calculating the similarity,
An output procedure for outputting a word whose similarity calculated in the similarity calculation procedure satisfies a predetermined condition among the words extracted from the information of the different industry as an information that becomes an analogy of the input information An information providing program characterized by having

Images