JP6882372B2 - Decision device, decision method, and decision program - Google Patents

Description

The present invention relates to a determination device, a determination method, and a determination program.

Conventionally, a technique called a knowledge base has been known in which the concept of a person or an event is used as an entity and the relationship between the entities is structured. Further, in order to efficiently create such a knowledge database, there is known a technique of clustering entities and updating relationships between entities based on the result of clustering.

Japanese Patent No. 6088091

“On Emerging Entity Detection” Michael Farber, Achim Rettinger, Boulos El Asmar

However, the above-mentioned techniques may not always provide a useful knowledge database.

For example, as a method of registering entities in the knowledge database, a method of extracting new entities and relationships between entities from contents posted on networks such as the Internet and registering the extracted entities and relationships is considered. Be done. However, when content on the network is used, there is a risk of registering incorrect information as a relationship between entities.

The present application has been made in view of the above, and an object of the present application is to provide a useful knowledge database.

The determination device according to the present application is an acquisition unit that acquires triple candidates including relationship information indicating a relationship between a first object, a second object, and an object from contents on a network, and a candidate acquired by the acquisition unit. On the other hand, a linking unit that associates the content that is the acquisition source of the candidate with the corresponding acquisition source information, and a determination unit that determines the triple to be registered based on the acquisition source information that is associated with the candidate. It is characterized by having.

According to one aspect of the embodiment, a useful knowledge database can be generated.

FIG. 1 is a diagram showing an example of a determination process executed by the information providing device according to the embodiment. FIG. 2 is a diagram showing an example of a search process executed by the information providing device according to the embodiment. FIG. 3 is a diagram illustrating an example of a flow of processing executed by the information providing device according to the embodiment. FIG. 4 is a diagram showing a configuration example of the information providing device according to the embodiment. FIG. 5 is a diagram showing an example of information registered in the entity database according to the embodiment. FIG. 6 is a diagram showing an example of information registered in the triple database according to the embodiment. FIG. 7 is a diagram showing an example of information registered in the reliability table according to the embodiment. FIG. 8 is a flowchart showing an example of the flow of the determination process executed by the information providing device according to the embodiment. FIG. 9 is a flowchart showing an example of the flow of the search process executed by the information providing device according to the embodiment. FIG. 10 is a diagram showing an example of a hardware configuration.

Hereinafter, a determination device according to the present application, a determination method, and a mode for carrying out the determination program (hereinafter, referred to as “embodiment”) will be described in detail with reference to the drawings. It should be noted that this embodiment does not limit the determination device, determination method, and determination program according to the present application. In addition, each embodiment can be appropriately combined as long as the processing contents do not contradict each other. Further, in each of the following embodiments, the same parts are designated by the same reference numerals, and duplicate description is omitted.

[1. Information provider]
First, with reference to FIG. 1, an example of processing executed by the information providing device 10 which is an example of the determination device will be described. FIG. 1 is a diagram showing an example of a determination process executed by the information providing device according to the embodiment. FIG. 1 describes an example of a determination process for determining a new triple to be registered in a knowledge database in which knowledge is systematized and organized information is registered as a process executed by the information providing device 10.

[1-1. Overview of information providing equipment]
The information providing device 10 is an information processing device capable of communicating with the data server 100 and the terminal device 200 via a predetermined network N (see, for example, FIG. 4) such as the Internet, and is, for example, a server device or a cloud system. It is realized by such as. The information providing device 10 may be able to communicate with an arbitrary number of data servers 100 and terminal devices 200 via the network N.

The data server 100 is an information processing device that manages various types of information, and is realized by, for example, a server device or a cloud system. For example, in the data server 100, the relationship between information is defined in advance as a structure, and Wikidata in which data having such a structure is registered and structured data such as an already constructed knowledge database are registered. Further, in the data server 100, semi-structured data to which information indicating attributes is added to various data is registered by using a tag language such as HTML (Hyper Text Markup Language) or XML (Extensible Markup Language). Has been done.

The data server 100 may manage structured data or semi-structured data including information of arbitrary contents. For example, the data server 100 may manage data including a log of a search query in an arbitrary search such as a web search, a route search, a search for a transaction target in an electronic shopping street, a map search, and a content search. Further, for example, the data server 100 may manage news, blogs, SNS (Social Networking Service), etc. posted on the website. Further, the data server 100 may manage explanatory contents for explaining various objects such as a person, an object, a building, contents, and an event. Further, for example, the data server 100 may manage explanatory contents that are contents that explain various objects that can be objects of an entity in a knowledge database, such as Wikipedia (registered trademark) and a dictionary service via the Internet. ..

The terminal device 200 is a terminal device used by a user, and is realized by, for example, a PC (Personal Computer), a server device, various smart devices, and the like. For example, the terminal device 200 has a function of communicating with the information providing device 10 and outputting information as a communication result when receiving input of various information from the user.

Under such a configuration, for example, the information providing device 10 provides various information registered in the knowledge database based on a request from the terminal device 200. For example, the terminal device 200 transmits a search query for searching the knowledge database to the information providing device 10. In such a case, the information providing device 10 searches the knowledge database for information corresponding to the search query, and provides the search result to the user terminal.

[1-2. Knowledge Database Overview]
Here, the knowledge database searched by the information providing device 10 will be described. Various kinds of knowledge are registered in the knowledge database in a systematic and organized state. For example, in the knowledge database, a triple, which is a set of two registered entities and information indicating a relationship between objects indicated by the entities (hereinafter, referred to as "relationship information"), is registered.

Here, an entity is information corresponding to various things in the world such as various things that can be the subject of a person, an object, a building, etc. in the real world, attributes such as occupation and nationality, various states and events, and so on. In the following description, various things corresponding to an entity may be described as "target".

Further, the relationship information is information indicating the relationship between the two entities. More specifically, the relationship information is information indicating the relationship between the first object indicated by the first entity and the second object indicated by the second entity. For example, in the information providing device 10, when the first target is a person, the second target is a book, and the first target is the author of the second target, information such as "author" or "auser" is used as related information. Registration will be accepted.

The entity in the knowledge database of the information providing device 10 may correspond to any thing or event. For example, the information providing device 10 may manage a triple in which the first target and the second target are people, and the first target is the parent or companion of the second target. Further, the information providing device 10 may manage a triple in which the first target and the second target are foods and the first target is the material of the second target. That is, the information providing device 10 manages information indicating relationships and events between objects, that is, a knowledge base, by managing triples indicating arbitrary relationships between arbitrary objects.

[1-3. About decision processing]
Now consider the process of registering a new triple in the knowledge database. In order to register a new triple in such a knowledge database, not only a character string indicating each object but also relationship information indicating how the object has a relationship with other objects is required. In order to efficiently add such triples to the knowledge database, a method of extracting triples from contents posted on the Internet such as news articles can be considered. However, since the content posted on the Internet contains incorrect information, it may not always be possible to extract the correct triple.

Therefore, the information providing device 10 executes the following determination process. First, the information providing device 10 acquires triple candidates including the first target, the second target, and the relationship information indicating the relationship between the targets from the contents on the network. Subsequently, the information providing device 10 associates the acquired candidate with the content that is the acquisition source of the candidate and the corresponding acquisition source information. Then, the information providing device 10 determines the triple to be registered from the candidates based on the acquisition source information associated with each candidate.

For example, triples extracted from domain content whose content is verified by multiple users, such as Wikipedia, are more accurate than triples extracted from content posted to domains such as personal blogs and SNS. Estimated to be high. Therefore, the information providing device 10 associates the acquisition source information, which is the acquisition source of the candidate of each triple, with the candidate of each triple. Then, when the acquisition source information associated with a certain triple candidate indicates a reliable acquisition source, the information providing device 10 targets the triple candidate as a registration target.

Here, even if the triples are extracted from the contents posted on personal blogs, SNS, etc., many contents include triples with the same contents, that is, triples showing the same relationship between the same objects. , Such triples are presumed to indicate correct information. Therefore, if the information providing device 10 is a candidate for triples having the same content and the acquisition source information associated with each triple satisfies a predetermined condition, the information providing device 10 targets the triple with the same content as the candidate for each triple. .. That is, the information providing device 10 determines whether or not the reliability of the content indicated by each triple candidate exceeds a predetermined threshold value based on the acquisition source information associated with each triple candidate, and the reliability is predetermined. If the threshold value of is exceeded, the triples with the contents indicated by the candidates of each triple are registered.

For example, the information providing device 10 associates each triple candidate with a reliability that is a numerical value indicating the reliability of the content that is the acquisition source of each triple candidate. To give a more specific example, the information providing device 10 associates the reliability according to the domain from which the candidate of each triple is acquired. That is, the information providing device 10 associates the candidate of each triple with the reliability corresponding to the domain of the acquisition source as the acquisition source information which is the information corresponding to the acquisition source of each triple.

Further, the information providing device 10 identifies a plurality of triple candidates showing the same relationship between the same objects from the candidates of each triple. For example, the information providing device 10 uses various techniques such as normalization to show a first entity that shows the same first object, a second entity that shows the same second object, and relationship information that shows the same relationship. Identify triple candidates that include.

Then, the information providing device 10 determines whether or not the reliability associated with the specified triple candidate satisfies a predetermined condition, and if the reliability satisfies the predetermined condition, it is the same as the specified triple candidate. Triples showing the same relationship between the subjects of are candidates for registration. For example, when the total reliability associated with the specified triple candidate exceeds a predetermined threshold value, the information providing device 10 sets a triple having the same content as the specified triple candidate as a registration candidate. Even if there is only one candidate for the specified triple, the information providing device 10 may determine such a triple candidate as a registration target when the reliability exceeds a predetermined threshold value.

In this way, the information providing device 10 associates the acquisition source of the candidate of each triple with the corresponding acquisition source information with respect to the candidate of each triple. Then, when the acquisition source information associated with the triple candidate indicating a certain relationship between a certain object satisfies a predetermined condition, the information providing device 10 has the same content as the triple candidate or the triple candidate. The triple of is to be registered in the knowledge database. As a result of such processing, the information providing device 10 can register triples with high reliability of the contents in the knowledge database. Therefore, as a result of improving the accuracy of the knowledge database, useful information is provided to the user. Can be done.

In the above-mentioned example, the information providing device 10 associates the reliability of the triple candidate according to the domain of the acquisition source, but the embodiment is not limited to this. For example, the information providing device 10 may associate the acquisition source domain with the triple candidate as the acquisition source information. In such a case, the information providing device 10 may target the triple candidate as long as the domain associated with the triple candidate is set as a highly reliable domain in advance. Further, the information providing device 10 is a triple candidate showing the same content even if each domain associated with a plurality of triple candidates showing the same content is set as a domain with low reliability in advance. When the number of the above exceeds a predetermined threshold value, the candidate of the triple or the candidate of the triple having the same content may be registered as the reliability is guaranteed. That is, if the information providing device 10 determines the triple to be registered among the triple candidates according to the reliability of the acquisition source of the triple candidate, any information can be used as the acquisition source information as the triple candidate. You may tie it.

[1-4. About decision processing]
Hereinafter, an example of the flow of processing executed by the information providing device 10 will be described with reference to FIG. In the following description, the triple candidate acquired from the content may be simply described as "triple", and the triple to be registered may be described as "registered triple". Further, in the following description, a plurality of triples including a first entity indicating the same first object, a second entity indicating the same second object, and relationship information indicating the same relationship are referred to as "same triple". May be described.

First, the information providing device 10 collects various contents from the data server 100 (step S1). For example, the information providing device 10 collects an already constructed knowledge database and various posted contents.

Subsequently, the information providing device 10 acquires triples from each content (step S2). For example, when the content of the acquisition source is the already constructed knowledge database, the information providing device 10 acquires the triples registered in the knowledge database. Further, when the acquisition source content is text-based content such as content posted on SNS or Wikipedia content, the information providing device 10 extracts triples by using morphological analysis, intention analysis, or the like. For example, the information providing device 10 extracts the subject, the predicate, and the object included in the text, and generates a triple in which the subject is the first entity, the object is the second entity, and the predicate is the relationship information. As for the technique for generating triples from such texts, it is assumed that various known techniques can be adopted, and detailed description thereof will be omitted.

For example, in the example shown in FIG. 1, the information providing device 10 is a triple # including the content # 1-1 belonging to the domain # 1, the entity # 1-1A, the entity # 1-1B, and the relationship information # 1-1. Get 1-1. Further, the information providing device 10 acquires the triple # 2-1 including the entity # 2-1A, the entity # 2-1B, and the relationship information # 2-1 from the content # 2-1 belonging to the domain # 2. The information providing device 10 acquires the triple # 3-1 including the entity # 3-1A, the entity # 3-1B, and the relationship information # 3-1 from the content # 3-1 belonging to the domain # 3. The content # 1-1 included in the domain # 1 is structured data, and the content # 2-1 included in the domain # 2 and the content # 3-1 included in the domain # 3 are unstructured data. Suppose that.

Subsequently, the information providing device 10 defeats each triple and associates the acquisition source information with each triple (step S3). For example, the information providing device 10 associates the domain # 1 with the triple # 1-1 as the acquisition source information, associates the domain # 2 with the triple # 2-1 as the acquisition source information, and associates the triple # 3- Domain # 3 is associated with 1 as acquisition source information. In step S3, the information providing device 10 may associate the reliability corresponding to each domain # 1 to # 3 as the acquisition source information.

Subsequently, the information providing device 10 identifies triples showing the same relationship from the acquired triples, and sets the registration target based on the total reliability of the acquisition sources indicated by the acquisition source information given to the identified triples. The registration triple is determined (step S4). For example, the information providing device 10 uses a character string (for example, a synonym for entity # 1-1A) indicating the same object as entity # 1-1A as the first entity from the acquired triples, and entity # 1-1B. The second entity is the character string indicating the same target as the above, and another triple having the character string indicating the same relationship as the relationship information # 1-1 as the relationship information, that is, the same triple is searched. Similarly, the information providing device 10 searches for the same triple with triple # 2-1 and the same triple with triple # 3-1.

For example, in the information providing device 10, the object indicated by the entity # 2-1A and the object indicated by the entity # 3-1A match, and the object indicated by the entity # 2-1B and the object indicated by the entity # 3-1B are one. However, if the relationship indicated by the relationship information # 2-1 and the relationship indicated by the relationship information # 3-1 match, even if the character strings of each entity and the relationship information are different, the triple # 2-1 and the triple # It is determined that 3-1 is the same triple.

A more specific example will be described. For example, the entity # 2-1A included in the triple # 2-1 is the Japanese notation "Shakespeare", the entity # 2-2 is the Japanese notation "King Lear", and the relationship information # 1-1 is. It shall be the "author" in Japanese notation. In addition, entity # 3-1A included in triple # 3-1 is "William Shakespeare" written in Japanese, entity # 2-2 is "King Lear" written in Japanese, and related information # 1-. It is assumed that 1 is the English notation "auther".

In such a case, the entity "Shakespeare" included in triple # 2-1 and the entity "William Shakespeare" included in triple # 3-1 have different character strings, but are the same person (the so-called United Kingdom in the 16th century). It is recognized that it indicates a person, Shakespeare). Further, the relationship information "author" included in triple # 2-1 and the relationship information "auter" included in triple # 3-1 have different character strings, but have the same relationship (that is, the first entity indicates). It is recognized that the person indicates the relationship (relationship that the person wrote the work indicated by the second entity). Therefore, in the information providing device 10, it is assumed that the triple # 2-1 and the triple # 3-1 are triples showing the same relationship, that is, the same triple.

As described above, in the information providing device 10, the triple candidates obtained from different contents show the same relationship with the first entity showing the same first target and the second entity showing the same second target. It was determined whether or not the relationship information was included, and it was determined that the first entity indicating the same first object, the second entity indicating the same second object, and the relationship information indicating the same relationship were included. It is decided whether or not the candidate is to be registered based on the acquisition source information linked to the candidates of a plurality of triples.

It should be noted that such processing can be realized by referring to, for example, a list created in advance, which is a collection of candidate entities indicating the same target. Further, for example, the information providing device 10 can realize such a search for the same triple by referring to an associative memory, a synonym dictionary, and a knowledge database in which synonyms are associated with each other. In addition to such processing, the information providing device 10 may search for the same triple by using, for example, various known techniques of normalization.

Then, the information providing device 10 determines the registered triple based on the acquisition source information of each same triple. For example, when the same triple does not exist in the triple # 1-1, the information providing device 10 specifies the transaction source information "domain # 1" associated with the triple # 1-1. Here, since the content registered in domain # 1 is structured data, it can be considered that a certain degree of reliability is guaranteed. Therefore, the information providing device 10 imparts a reliability "80" higher than a predetermined threshold value (for example, "50") to the domain # 1.

Further, for example, the information providing device 10 identifies the transaction source information "domain # 2" and "domain # 3" associated with the triple # 2-1 and the triple # 3-1 which are the same triples. Here, since the contents registered in the domain # 2 and the domain # 3 are semi-structured data, it is considered that the reliability of the contents is lower than that of the contents of the domain # 1. Therefore, the information providing device 10 imparts the reliability "50" to the domain # 2, and imparts the reliability "30" to the domain # 3. It should be noted that what kind of reliability is given to the content of which domain can be arbitrarily set in advance. For example, the information providing device 10 is semi-structured for the content of structured data. Higher reliability than data content, and higher reliability than domain content that can be posted by the actions of a single user for domain content whose content is scrutinized by multiple users It may be attached.

Then, the information providing device 10 determines whether or not the total reliability of each same triple exceeds a predetermined value (for example, "80"), and if it exceeds, either one of the same triples or These triples with the same contents as the same triples are designated as registered triples. For example, in the information providing device 10, since the reliability of the triple # 1-1 is "80", the entity # 1-1A is set as the entity # 1A, the entity # 1-1B is set as the entity # 1B, and the relationship information # 1 is set. A triple whose relationship information # 1 is -1 is a registered triple # 1. That is, the information providing device 10 sets triple # 1-1 as registered triple # 1.

On the other hand, in the information providing device 10, the sum of the reliability "50" of the triple # 2-1 and the reliability "30" of the triple # 3-1 is "80", so that the triple # 2-1 and the triple # are A triple having the same content as the content of 3-1 is designated as a registered triple # 2. For example, the information providing device 10 has entity # 2A indicating the same object as entity # 2-1A or entity # 3-1A, entity # 2B indicating the same object as entity # 2-1B or entity # 3-1B, and so on. And the triple including the relationship information # 2 showing the same relationship as the relationship information # 2-1 and the relationship information # 3-1 is designated as the registered triple # 2. That is, the information providing device 10 generates a triple in which the contents of the same triple are normalized according to the rules in the knowledge database to be finally generated, and the generated triple is designated as a registered triple.

For example, when triple # 2-1 or triple # 3-1 contains the above-mentioned character string, the information providing device 10 includes "William Shakespeare" written in Japanese as entity # 2A, and Japan as entity # 2B. A triple including the word notation "King Lear" and including the Japanese notation "author" as related information # 2 is designated as a registered triple # 2.

Then, the information providing device 10 adds the acquisition source information corresponding to the acquisition source to each of the registration triples # 1 and # 2, and registers them in the knowledge database. For example, the information providing device 10 associates the registered triple # 1 with the acquisition source information “domain # 1” of the triple # 1-1 that is the source of the registered triple # 1 and registers it in the knowledge database. Further, the information providing device 10 links the registration triple # 2 with the acquisition source information "domain # 2" of the original triple # 2-1 and the acquisition source information "domain # 3" of the triple # 3-1. Attach and register in the knowledge database.

Further, the information providing device 10 receives a search query from the terminal device 200 (step S5). In such a case, the information providing device 10 searches the knowledge database for information corresponding to the search query, and provides the search result to the terminal device 200 (step S6). At this time, the information providing device 10 narrows down the triples, entities, and related information to be searched by using the acquisition source information associated with each registered triple.

[1-5. About search process]
Subsequently, the search process executed by the information providing device 10 will be described with reference to FIG. FIG. 2 is a diagram showing an example of a search process executed by the information providing device according to the embodiment. The search process corresponds to steps S5 and S6 shown in FIG. Further, in the example shown in FIG. 2, as shown in step S4 of FIG. 1, it is assumed that the acquisition source information indicating the acquisition source is associated with the triple in the knowledge database.

For example, the information providing device 10 receives a search query from the terminal device 201 used by the first user (step S1). More specifically, the information providing device 10 accepts a search query and a designation of a domain to be searched. In such a case, the information providing device 10 searches for information corresponding to the search query by using the triple associated with the designated domain (step S2).

For example, in the example shown in FIG. 2, the domain # 1 is associated with the registered triple # 1 including the entity # 1 and the entity # 2, and the registered triple # 2 including the entity # 1 and the entity # 3 is associated with the domain # 1. Domains # 2 and # 3 are associated with, and domains # 1 and # 2 are associated with registration triple # 3 including entity # 1 and entity # 4. Here, when the information providing device 10 receives the "entity # 1" and the search target "domain # 1" as the search query from the first user, the "domain" of the registered triples # 1 to # 3 is used. From the registered triples # 1 and # 3 associated with "# 1", another entity (for example, entity # 2, entity # 4, etc.) having a predetermined relationship with "entity # 1" is searched. Then, the information providing device 10 transmits the search result to the terminal device 201.

On the other hand, when the information providing device 10 receives the search query "entity # 1" and the search target "domain # 2" from the terminal device 202 used by the second user, for example, the registration triples # 1 to # Of 3, other entities (for example, entity # 3, entity # 4, etc.) having a predetermined relationship with "entity # 1" are selected from the registered triples # 2 and # 3 to which "domain # 2" is associated. Search for. Then, the information providing device 10 transmits the search result to the terminal device 202.

That is, the information providing device 10 multiplexes the knowledge database for each acquisition source by associating the acquisition source information with the triple. More specifically, the information providing device 10 generates a knowledge database multiplexed for each domain, that is, a multi-domain knowledge database, by associating the acquisition source domains with triples. Then, the information providing device 10 searches for the information corresponding to the search query received from the user by using the triple in which the designated acquisition source and the corresponding acquisition source information are associated with each other.

In the example shown in FIG. 2, the information providing device 10 associates the domain that is the acquisition source of the triple as the transaction source information, but the embodiment is not limited to this. For example, the information providing device 10 may associate reliability with each triple. More specifically, the information providing device 10 imparts the reliability "80" corresponding to the domain # 1 to the registered triple # 1, and the trust corresponding to the domain # 2 to the registered triple # 2. The reliability "80", which is the sum of the degree "50" and the reliability "30" corresponding to the domain # 3, is given, and the reliability "80" corresponding to the domain # 1 is given to the registered triple # 3. The reliability "130", which is the sum of the domain # 2 and the corresponding reliability "50", is given.

Then, for example, when the reliability specified by the user is "80", the information providing device 10 searches for the registered triples # 1 and # 3 having the reliability of "80" or more, and the user informs the user. When the specified reliability is "100", the registered triple # 3 having a reliability of "100" or more may be searched. Further, for example, the information providing device 10 associates and registers the domain as the acquisition source for each triple, and when the user accepts the designation of the reliability, the information providing device 10 is associated with each triple. The search target may be triples in which the sum of the reliability of the domain and the corresponding reliability exceeds the specified reliability.

[1-6. About the total reliability method]
In the above example, the information providing device 10 determines the registered triple according to whether or not the sum of the reliabilitys according to the acquisition source of the triple exceeds a predetermined threshold value. Here, the information providing device 10 may search for registered triples depending on whether or not the value of the reliability considering various weights exceeds a predetermined threshold value.

For example, the information providing device 10 may set a triple having the same relationship as the same triple as a registered triple when the load sum of the reliabilitys associated with the same triple exceeds a predetermined threshold value. Note that such weights may be adjustable according to, for example, arbitrary information. For example, the information providing device 10 may set a larger weight as the registration time of the content that is the extraction source of the triple is older, the evaluation from other users is higher, or the number of views is larger. Further, the information providing device 10 may set a smaller weight as the registration time of the content that is the extraction source of the triple is newer, the evaluation from other users is lower, or the number of views is lower. .. Further, the information providing device 10 gives a larger weight as the degree of matching between the character string of each entity included in the triple and the character string of the relational information and the character string registered in the reference dictionary information is higher. It may be set.

Further, for example, the reliability of each domain is considered to be relatively determined in addition to the absolutely determined one. Therefore, the information providing device 10 may calculate the load sum of the reliabilitys associated with the same triple by using the weights according to the relevance of the contents that are the acquisition sources of each triple. For example, the information providing device 10 is more reliable than the triple obtained from the domain # 4 when the domain # 5 is relatively more reliable than the triple obtained from the domain # 4. A high value of reliability or weight may be added to calculate the total reliability. Further, for example, when the reliability of the domain # 6 is relatively higher than that of the domain # 5, the information providing device 10 may refer to the triple acquired from the domain # 6 rather than the triple acquired from the domain # 5. The total reliability may be calculated by giving a higher reliability or weight. That is, the information providing device 10 may change the weight when calculating the total reliability of the same triple according to the relative reliability between the domains from which the same triple is acquired.

Further, the information providing device 10 associates the reliability of a value higher than a predetermined threshold value with the triple acquired from the structured data, and the value lower than the predetermined threshold value with respect to the triple acquired from the semi-structured data. You may associate the reliability of. That is, the information providing device 10 may associate the triple acquired from the structured data with a higher reliability than the triple acquired from the semi-structured data.

Further, the information providing device 10 may associate the acquisition source domain with the reliability corresponding to the relationship indicated by the relationship information included in the acquired triple. For example, the information providing device 10 is relatively high when the relational information included in the triple acquired from the first domain in which the information about the book is highly reliable indicates a highly relevant relation with the book such as "author". If the reliability of the value is linked to the triple and the relationship information contained in the triple obtained from the first domain shows a relationship that is not related to the book such as "location", the reliability of the relatively low value is used. It may be tied to a triple.

[1-7. About an example of specific processing]
In the above example, the information providing device 10 associates the acquisition source information with the triple acquired from the content, and determines the registered triple based on the acquisition source information associated with the triple. Here, the information providing device 10 may improve the accuracy of the knowledge database by executing various processes other than the above-described processes.

For example, as described above, the Web may include content that expresses the same relationship between the same objects in different languages. Therefore, the information providing device 10 may perform mapping between each entity and the object indicated by the entity by performing various normalizations. Further, in order to ensure the diversity of the knowledge database, the information providing device 10 acquires triples not only from structured data but also from semi-structured data, but in order to acquire triples from semi-structured data in this way. It may have a function (for example, from which tagged information the triple is extracted, etc.) necessary for the database. Further, the information providing device 10 may further improve the accuracy of the knowledge database by using the information manually input as a reference.

Hereinafter, a specific example of the process in which the information providing device 10 generates the knowledge database will be described with reference to FIG. FIG. 3 is a diagram illustrating an example of a flow of processing executed by the information providing device according to the embodiment. As shown in FIG. 3, the information providing device 10 generates a knowledge database from structured data and structured data by executing a series of processes shown in the system S.

For example, the system S for performing the information providing apparatus 10, ImporterS1, Information ExtractorS2, Attribute ConverterS3, Entitiy MatcherS4, Entitiy ConnectorS5, Id AssignerS6, Additional Data CombinerS7, Entity MergerS8, Object ConverterS9, Attribute CompleterS10, ValidatorS11, and ExporterS12 (hereinafter , "Each process S1 to S12" may be collectively referred to.). The system S may be hardware including a functional configuration for executing each process S1 to S12, or may be various models such as a neural network.

Hereinafter, an example of the processing in each processing S1 to S12 will be described. InterS1 imports data from various schemas as an entity so as to match the schema of the knowledge database to be finally created. For example, InterS1 extracts triples from structured data in which relationships between entities have already been set, and generates data to which acquisition source information is added to the extracted triples.

For example, InterS1 identifies what kind of relationship the relationship information contained in the triple of structured data shows, and specifies the relationship information that shows the specified relationship in the knowledge database that is finally generated. .. For example, InterS1 identifies a first entity that indicates a subject, a second entity that indicates an object, and a triple that is included as relationship information that indicates a predicate. Note that each entity and relationship information may have different character strings even if they indicate the same object or relationship. As a result, the Importer S1 identifies a plurality of triples having different character strings of entities and relationship information as triples showing the same relationship between the same objects in the real world. Further, the Importer S1 associates the acquired triple with the acquisition source information and assigns a unique temporary ID to each triple.

Information Exector S2 acquires triples from semi-structured data such as web text. For example, Information Exector S2 acquires triples from semi-structured data using techniques such as tag analysis and morphological analysis. In addition, Information Exector S2 compares the entities registered in the knowledge database with the entities included in the acquired triples, and associates the entities indicating the same target with each other.

AttributeConverterS3 converts a class of input data into a class of knowledge database. For example, AttributeConverterS3 associates the class given to each entity in the structured data from which the triple is acquired (that is, the information indicating the target attribute indicated by the entity) with the class of each entity in the knowledge database. Refer to the mapping information and convert the class of each entity included in the acquired triple to the knowledge database class. For example, AttributeConverterS3 normalizes the character strings of entities indicating the same target and normalizes the relationship information indicating the same relationship. For example, AttributeConverterS3 acquires a triple obtained by converting the relation information "outer" in English notation into the relation information "author" in Japanese notation.

Entry MatcherS4 matches entities indicating the same target and creates a group of the same entity. For example, the Entry Matcher S4 compares the entities contained in each triple and identifies the entities that are presumed to show the same relationship of the same object. Then, the Entry Matcher S4 assigns a group ID to the specified entity. Note that the Entry Matcher S4 may specify the same triple so that there is no contradiction in the class (attribute) of each entity. Further, in the Entry MatcherS4, when entities having no contradiction in attributes are linked to each other, the entities satisfying the creek structure in the graph structure connecting the entities may be regarded as the same entity.

EntryConnector S5 associates groups of entities with different classes but related to each other. That is, the Entry Connector S5 associates groups of entities that are presumed to have the same target with each other, although the classes are different.

IdAssignerS6 associates an ID (Identifier) assigned to an entity in the final knowledge database with each entity group. For example, IdAssignerS6 identifies an entity that indicates the same target as the grouped entity among the registered entities, and assigns the ID given to the specified entity to the grouped entity.

The Additional Data Comminer S7 reflects the modification of the ID manually assigned by the Id Assigner S6. Further, the Additional Data Comminer S7 associates the ID assigned to the registered entity indicating the same target with the entity acquired by the Information Exector S2.

The Entry Merger S8 integrates a plurality of entities indicating the same object, that is, the same entity into one entity. At this time, the Entry Merger S8 performs addition of reliability and calculation of load sum according to the acquisition source information associated with the entity.

ObjectConverterS9 associates the ID of the registered entity indicating the same target as the object entity among the entities included in the triple.

AttributeCompleterS10 adds new information to the entity. For example, the Attribute Compressor S10 adds information based on a preset ontology. In addition, the Attribute Compressor S10 may discover or add a missing triple based on the ontology. It can be realized by any method such as discovery and addition of such triples, rule base, machine learning, and the like.

Validator S11 deletes or modifies a triple with low reliability based on the acquisition source information. For example, Validator S11 is a registered triple only when the sum of reliability calculated by Something Merger S8 exceeds a predetermined threshold. Note that the Validator S11 may exclude triples that match the blacklist set manually in advance from the registration target regardless of the reliability.

The Explorer S12 registers triples whose reliability exceeds a predetermined threshold value in a knowledge database in a predetermined format. Note that the Explorer S12 may delete information other than the preset information among the information given to each entity of the triple.

[2. Configuration of information providing device]
Hereinafter, an example of the functional configuration of the information providing device 10 described above will be described. FIG. 4 is a diagram showing a configuration example of the information providing device according to the embodiment. As shown in FIG. 4, the information providing device 10 includes a communication unit 20, a storage unit 30, and a control unit 40.

The communication unit 20 is realized by, for example, a NIC (Network Interface Card) or the like. Then, the communication unit 20 is connected to the network N by wire or wirelessly, and transmits / receives information to / from the data server 100 and the terminal device 200.

The storage unit 30 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory (Flash Memory), or a storage device such as a hard disk or an optical disk. Further, the storage unit 30 stores the entity database 31, the triple database 32, and the reliability table 33 (hereinafter, may be collectively referred to as “each database 31 to 33”).

Hereinafter, an example of information registered in each database 31 to 33 will be described with reference to FIGS. 5 to 7. Information about an entity is registered in the entity database 31. For example, FIG. 5 is a diagram showing an example of information registered in the entity database according to the embodiment. As shown in FIG. 5, information having items such as "entity ID", "entity type", "node ID", "node type", and "data" is registered in the entity database 31.

Here, the "entity ID" is an identifier of an entity. Further, the "entity type" is information indicating the type of the entity indicated by the associated "entity ID". For example, whether the entity is an entity indicating a "person" or an entity indicating an "occupation". Information such as is shown. The "node ID" is an identifier of a node associated with the entity indicated by the associated "entity ID". The "node type" is information indicating the type of the node indicated by the associated "node ID", and is information such as whether the node indicates a name, a photograph, an occupation, or the like. Further, the "data" is the data of the node indicated by the associated "node ID".

For example, in the example shown in FIG. 5, the entity ID "E11", the entity type "person", the node ID "I111", the node type "name", and the data "name # 1" are registered in association with each other. In such information, the entity indicated by the entity ID "E11" (that is, the entity E11) is an entity indicating a "person", and the node ID "I111" is registered as a node indicating the "name" of the person indicated by the entity. Indicates that the name is "Name # 1". In the example shown in FIG. 5, conceptual values such as "name # 1" and "photo # 1" are described, but in reality, the entity database 31 contains a person corresponding to the associated entity. Various information indicating the name, photo, date of birth, etc. will be registered.

Information indicating the triple is registered in the triple database 32. For example, FIG. 6 is a diagram showing an example of information registered in the triple database according to the embodiment. In the example shown in FIG. 6, the triple database 32 has items such as "triple ID", "relationship information ID", "type", "first element", "second element", and "acquisition source information". Information is registered.

Here, the "triple ID" is an identifier that identifies the triple. Further, the "relationship information ID" is an identifier that identifies the relational information included in the triple. Further, the "type" is information indicating the relationship between the entities included in the triple. Further, the "first element" and the "second element" are entity IDs of entities included in the triple indicated by the associated "triple ID". Further, the "acquisition source information" is information corresponding to the acquisition source of the associated triple, for example, information indicating the domain of the acquisition source and information indicating the reliability.

For example, in the example shown in FIG. 6, the triple ID "triple # 1", the relationship information ID "C1", the type "occupation", the first element "E11", the second element "E21", and the "domain # 1" are It is registered in association with each other. Such information is registered as a triple indicated by the triple ID "triple # 1" in which the entity E11, the entity E12, and the relationship information C1 are associated with each other, indicating that the entity E21 is the occupation of the entity E11. In addition, such information indicates that this triple is information obtained from "domain # 1".

In the reliability table 33, reliability information according to the acquisition source of the triple is registered. For example, FIG. 7 is a diagram showing an example of information registered in the reliability table according to the embodiment. As shown in FIG. 7, "domain" and "reliability" are registered in association with each other in the reliability table 33. Here, the "domain" is information indicating the domain from which the triple is acquired. For example, in the example shown in FIG. 7, the domain "domain # 1" and the reliability "80" are registered in association with each other. Such information indicates that the reliability of the triple obtained from the contents of the domain indicated by the domain "domain # 1" is "80".

Returning to FIG. 4, the explanation will be continued. The control unit 40 is a controller, and for example, various programs stored in a storage device inside the information providing device 10 by a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) are stored in a RAM or the like. Is realized by executing as a work area. Further, the control unit 40 is a controller, and may be realized by, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

As shown in FIG. 2, the control unit 40 includes an acquisition unit 41, a linking unit 42, a determination unit 43, a determination unit 44, a registration unit 45, and a search unit 46. The functional configuration shown in FIG. 4 does not necessarily correspond to the system S shown in FIG. 3, but the system shown in FIG. 3 can be executed by executing a part or the whole of the processes S1 to S12 shown in FIG. Similarly, the accuracy of the knowledge database can be improved by registering the registered triples in the knowledge database according to the reliability of the triples. In other words, the functional configuration shown in FIG. 4 shows the main functions exhibited by each of the processes S1 to S12 shown in FIG.

The acquisition unit 41 acquires triple candidates including the first target, the second target, and the relationship information indicating the relationship between the targets from the contents on the network. For example, the acquisition unit 41 acquires structured data or semi-structured data from the data server 100, and from the acquired contents, the relationship showing the relationship between the first entity, the second entity, and the object indicated by each entity. Extract triples that are pairs with information. The acquisition unit 41 may acquire triples as content on the network from structured data in which the relationship between targets is defined in advance or semi-structured data in which data attributes are set. ..

The linking unit 42 associates the content to be the acquisition source with the corresponding acquisition source information with respect to the triple acquired by the acquisition unit. For example, the linking unit 42 links the information indicating the content that is the acquisition source of the triple, the information indicating the domain to which the content belongs, or the reliability according to the content or the domain that is the acquisition source to the triple as the acquisition source information. wear. In addition, for example, the associating unit 42 associates the reliability of a value higher than a predetermined threshold value with respect to the triple acquired from the structured data, and with respect to the candidate of the triple acquired from the semi-structured data, from the predetermined threshold value. May be associated with a low reliability.

The determination unit 43 includes triple candidates obtained from different contents, a first entity indicating the same first target, a second entity indicating the same second target, and relationship information indicating the same relationship. Judge whether or not. For example, the determination unit 43 uses various dictionaries, ontology, and the like to determine whether or not each entity and relationship information included in the triple acquired by the acquisition unit 41 indicates the same object or relationship. That is, the determination unit 43 determines whether or not each triple is the same triple.

The determination unit 44 determines the triple to be registered from each candidate based on the acquisition source information associated with each triple. For example, the determination unit 44 sets the triple candidate as a registration target when the reliability associated with the triple satisfies a predetermined condition. Further, the determination unit 44 may set a triple showing the same relationship as the same triple as the registration target when the sum of the reliabilitys and the sum of the loads associated with the same triple exceeds a predetermined threshold value. That is, the determination unit 44 is a plurality of triples acquired from different contents, and the first object indicated by the first entity, the second object indicated by the second entity, and the relationship indicated by the relationship information match. If the reliability associated with each of the plurality of triples satisfies a predetermined condition, triples having the same contents as each triple may be registered.

In addition, the determination unit 44 may calculate the sum of loads by using the weights according to the relevance between the contents that are the acquisition sources. For example, the determination unit 44 may apply a weight according to the relative reliability between the domains of the content to be acquired to the load sum. Further, for example, when the acquisition source information indicating the acquisition source domain is associated with each triple, the determination unit 44 reads and reads the reliability associated with the acquisition source information from the reliability table 33. It may be determined whether or not a triple is to be registered based on whether or not the reliability determines a predetermined condition. For example, the acquisition source indicated by the acquisition source information and the number of the same triples are determined. It may be decided whether or not the triple is to be registered based on whether or not the predetermined condition is satisfied.

The registration unit 45 registers the triple to be registered and the acquisition source information in the database in association with each other. For example, the registration unit 45 registers the information of each entity included in the triple to be registered in the entity database 31. Further, the registration unit 45 registers the entity ID of each entity included in the triple, the relationship information, and the acquisition source information in the triple database 32 in association with each other.

The search unit 46 searches for information corresponding to the search query according to the acquisition source information associated with each triple. For example, the search unit 46 receives a domain designation from the terminal device 200 together with a search query. In such a case, the search unit 46 identifies the triple to which the domain specified from the terminal device 200 is associated as the acquisition source information from the triple database 32, and sets the data of each entity included in the specified triple as the search target. , Search for information that corresponds to the search query.

When the search unit 46 receives the reliability from the terminal device 200, for example, the search unit 46 refers to the triple database 32 and the reliability table 33, and receives the reliability indicated by the associated acquisition source information from the terminal device 200. Identify triples that are more reliable. Then, the search unit 46 may search for the information corresponding to the search query by using the data of each entity included in the specified triple as the search target. After that, the search unit 46 provides the search result to the terminal device 200.

[3. An example of the flow of processing executed by the information providing device]
Subsequently, the flow of the determination process executed by the information providing device 10 will be described with reference to FIG. FIG. 8 is a flowchart showing an example of the flow of the determination process executed by the information providing device according to the embodiment.

First, the information providing device 10 acquires triples from each content (step S101). Subsequently, the information providing device 10 associates the acquisition source information corresponding to the acquisition source of the triple with the triple (step S102). Further, the information providing device 10 identifies triples showing the same relevance (step S103). Then, the information providing device 10 determines whether or not to register the triples showing the relevance shown by these triples based on the acquisition source information associated with each of the specified triples (step S104). After that, the information providing device 10 registers the triple to be registered in the database in a state of associating the acquisition source information (step S105), and ends the process.

Subsequently, the flow of the search process executed by the information providing device 10 will be described with reference to FIG. FIG. 9 is a flowchart showing an example of the flow of the search process executed by the information providing device according to the embodiment.

First, the information providing device 10 determines whether or not the search query has been accepted from the terminal device 200 (step S201), and if not (step S201: No), waits for processing. On the other hand, when the information providing device 10 receives the search query (step S201: Yes), the information providing device 10 identifies the triple corresponding to the acquisition source indicated by the search query (step S202). Then, the information providing device 10 traces the specified triple and searches for the entity corresponding to the search query (step S203). After that, the information providing device 10 provides the search result (step S204), and ends the process.

[4. Modification example]
In the above, an example of the determination process and the search process by the information providing device 10 has been described. However, the embodiment is not limited to this. Hereinafter, variations of the determination process and the search process executed by the information providing device 10 will be described.

[4-1. About the type of entity]
In the above-mentioned example, as an example of the knowledge database, a triple or the like showing the relationship between a person and a work has been described, but the embodiment is not limited to this. That is, the information providing device 10 may update the knowledge database indicating arbitrary things.

[4-2. Device configuration〕
The information providing device 10 does not have to manage the knowledge database by its own device. For example, each database 31 to 33 registered in the storage unit 30 may be stored in an external storage server. Further, the information providing device 10 may be realized by a front-end server that realizes a search process and a back-end server that realizes a decision process. In such a case, the back-end server is provided with each of the units 41 to 45 shown in FIG. 4, and the front-end server is provided with a search unit 46 for exerting a function of searching the knowledge database.

[4-3. Others]
Further, among the processes described in the above-described embodiment, all or a part of the processes described as being automatically performed can be manually performed, and conversely, the processes described as being manually performed. It is also possible to automatically perform all or part of the above by a known method. In addition, the processing procedure, specific name, and information including various data and parameters shown in the above document 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 of the illustrated devices is a functional concept, and does not necessarily have to be physically configured as shown in the figure. That is, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or part of the device is functionally or physically distributed / physically in arbitrary units according to various loads and usage conditions. Can be integrated and configured.

In addition, the above-described embodiments can be appropriately combined as long as the processing contents do not contradict each other.

[4-4. program〕
Further, the information providing device 10 according to the above-described embodiment is realized by, for example, a computer 1000 having a configuration as shown in FIG. FIG. 10 is a diagram showing an example of a hardware configuration. The computer 1000 is connected to the output device 1010 and the input device 1020, and the arithmetic unit 1030, the primary storage device 1040, the secondary storage device 1050, the output IF (Interface) 1060, the input IF 1070, and the network IF 1080 are connected by the bus 1090. Has.

The arithmetic unit 1030 operates based on a program stored in the primary storage device 1040 or the secondary storage device 1050, a program read from the input device 1020, or the like, and executes various processes. The primary storage device 1040 is a memory device that temporarily stores data used by the arithmetic unit 1030 for various calculations, such as a RAM. Further, the secondary storage device 1050 is a storage device in which data used by the arithmetic unit 1030 for various calculations and various databases are registered, such as a ROM (Read Only Memory), an HDD (Hard Disk Drive), and a flash memory. Is realized by.

The output IF 1060 is an interface for transmitting information to be output to an output device 1010 that outputs various information such as a monitor and a printer. For example, USB (Universal Serial Bus), DVI (Digital Visual Interface), and the like. It is realized by a connector of a standard such as HDMI (registered trademark) (High Definition Multimedia Interface). Further, the input IF 1070 is an interface for receiving information from various input devices 1020 such as a mouse, a keyboard, a scanner, and the like, and is realized by, for example, USB.

The input device 1020 includes, for example, an optical recording medium such as a CD (Compact Disc), a DVD (Digital Versatile Disc), or a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), or a tape. It may be a device that reads information from a medium, a magnetic recording medium, a semiconductor memory, or the like. Further, the input device 1020 may be an external storage medium such as a USB memory.

The network IF1080 receives data from another device via the network N and sends it to the arithmetic unit 1030, and also transmits the data generated by the arithmetic unit 1030 to the other device via the network N.

The arithmetic unit 1030 controls the output device 1010 and the input device 1020 via the output IF 1060 and the input IF 1070. For example, the arithmetic unit 1030 loads a program from the input device 1020 or the secondary storage device 1050 onto the primary storage device 1040, and executes the loaded program.

For example, when the computer 1000 functions as the information providing device 10, the arithmetic unit 1030 of the computer 1000 realizes the function of the control unit 40 by executing the program loaded on the primary storage device 1040.

[5. effect〕
As described above, the information providing device 10 acquires triple candidates including the first target, the second target, and the relationship information indicating the relationship between the targets from the contents on the network. Further, the information providing device 10 associates the acquired candidate with the content that is the acquisition source of the candidate and the corresponding acquisition source information. Then, the information providing device 10 determines the triple to be registered from the candidates based on the acquisition source information associated with each candidate. As a result of such processing, for example, the information providing device 10 targets triples acquired from a highly reliable acquisition source, so that reliability can be ensured while automatically expanding the knowledge database. As a result, a useful knowledge database can be provided.

Here, the information providing device 10 may associate the acquisition source information with a reliability, which is a numerical value indicating the reliability of the content. Further, the information providing device 10 may associate the reliability according to the domain of the content as the acquisition source information. Then, the information providing device 10 sets the triple candidate as a registration target when the reliability associated with the triple candidate satisfies a predetermined condition. For example, the information providing device 10 is a plurality of candidates acquired from different contents, and the reliability associated with each of the first target, the second target, and the plurality of candidates whose related information matches is high. When the predetermined condition is satisfied, the triple including the first target, the second target, and the relational information is the registration target. The information providing device 10 sets the first target, the second target, and the first target when the load sum of the reliabilitys associated with a plurality of candidates whose relational information matches exceeds a predetermined threshold value. A triple including the second target and the relational information may be the registration target. As a result of such processing, the information providing device 10 can guarantee the reliability of the triple to be registered, and thus can provide a useful knowledge database.

Further, the information providing device 10 is associated with a plurality of candidates whose first target, second target, and related information match, using weights according to the relevance of the contents that are the acquisition sources of each candidate. The sum of loads of reliability may be calculated. As a result, the information providing device 10 calculates the reliability of the triple in consideration of the relationship between the contents, for example, so that the reliability of the registered triple can be further improved.

Further, the information providing device 10 acquires triple candidates from structured data in which the relationship between targets is defined in advance or semi-structured data in which data attributes are set as contents on the network. .. Here, the information providing device 10 associates the reliability of a value higher than a predetermined threshold value with the triple candidate acquired from the structured data, and determines the triple candidate acquired from the semi-structured data. The reliability of a value lower than the threshold value may be associated. Therefore, the information providing device 10 can realize the automatic expansion of the knowledge database and can provide a highly reliable knowledge database.

Further, in the information providing device 10, the triple candidates acquired from different contents show the same relationship between the first element indicating the same first object and the second element indicating the same second object. And whether or not to include. Then, the information providing device 10 is a plurality of triples determined to include a first element indicating the same first object, a second element indicating the same second object, and relationship information indicating the same relationship. Based on the acquisition source information associated with the candidate, it is determined whether or not the candidate is to be registered. Therefore, the information providing device 10 can guarantee the reliability of the triple registered in the knowledge database.

Further, when the acquisition source information satisfies a predetermined condition, the information providing device 10 includes a triple including a first element indicating the first target, a second element indicating the second target, and relational information indicating the relationship. To be registered. Therefore, the information providing device 10 can guarantee the reliability of the triple to be registered.

Further, the information providing device 10 registers the triple to be registered and the acquisition source information in the database in association with each other. Therefore, the information providing device 10 can provide, for example, a knowledge database capable of realizing a search in consideration of the acquisition source and the reliability of the acquisition source.

Although some of the embodiments of the present application have been described in detail with reference to the drawings, these are examples, and various modifications are made based on the knowledge of those skilled in the art, including the embodiments described in the disclosure column of the invention. It is possible to practice the present invention in other improved forms.

Further, the above-mentioned "section, module, unit" can be read as "means" or "circuit". For example, the imparting unit can be read as a specific means or a specific circuit.

10 Information provider 20 Communication unit 30 Storage unit 31 Entity database 32 Triple database 33 Reliability table 40 Control unit 41 Acquisition unit 42 Linking unit 43 Judgment unit 44 Decision unit 45 Registration unit 46 Search unit 100 Data server 200, 201, 202 Terminals apparatus

Claims (14)

An acquisition unit that acquires triple candidates including relationship information indicating the relationship between the first target, the second target, and the target from the contents on the network, and the acquisition unit.
With respect to the candidate acquired by the acquisition unit, a linking unit that associates the content that is the acquisition source of the candidate with the corresponding acquisition source information, and
A determination device having a determination unit for determining a triple to be registered based on acquisition source information associated with the candidate. The determination device according to claim 1, wherein the linking unit links the reliability, which is a numerical value indicating the reliability of the content, as the acquisition source information. The determination device according to claim 2, wherein the linking unit links the reliability according to the domain of the content as the acquisition source information. The determination device according to claim 2 or 3, wherein the determination unit targets the triple candidate when the reliability associated with the triple candidate satisfies a predetermined condition. The determination unit is a plurality of candidates obtained from different contents, and the reliability associated with each of the first target, the second target, and the plurality of candidates whose relationship information matches is determined. If the predetermined condition is satisfied, the first object, the second object, and the triple including the related information are to be registered, according to any one of claims 2 to 4. Determining device. When the load sum of the reliability associated with the first target, the second target, and the plurality of candidates whose relationship information matches exceeds a predetermined threshold value, the determination unit sets the first target. The determination device according to claim 5, wherein a triple including the second object and the related information is to be registered. The determination unit is associated with a plurality of candidates whose first target, the second target, and the relationship information match, using weights according to the relevance of the contents that are the acquisition sources of each candidate. The determination device according to claim 6, wherein the load sum of the reliability is calculated. The acquisition unit acquires the triple candidate from the structured data in which the relationship between the objects is defined in advance or the semi-structured data in which the data attribute is set as the content on the network. The determination device according to any one of claims 1 to 7. The linking portion associates the triple candidate acquired from the structured data with a reliability of a value higher than a predetermined threshold value, and associates the triple candidate acquired from the semi-structured data with the predetermined threshold value. The determination device according to claim 8, wherein a reliability of a lower value is associated. Determining whether triple candidates obtained from different contents include a first element indicating the same first object, a second element indicating the same second object, and relationship information indicating the same relationship. Has a judgment unit to
The determination unit is associated with a plurality of triple candidates determined to include a first element indicating the same first object, a second element indicating the same second object, and relationship information indicating the same relationship. The determination device according to any one of claims 1 to 9, wherein the candidate is determined whether or not to be registered based on the acquired acquisition source information. When the acquisition source information satisfies a predetermined condition, the determination unit includes a first element indicating the first target, a second element indicating the second target, and a relational information indicating the relationship. The determination device according to claim 10, wherein the determination device is to be registered. The determination device according to any one of claims 1 to 11, wherein the determination unit has a registration unit that registers the triple to be registered and the acquisition source information in a database. The decision method performed by the decision device
The acquisition process of acquiring triple candidates including the relationship information indicating the relationship between the first target, the second target, and the target from the contents on the network, and the acquisition process.
With respect to the candidate acquired by the acquisition process, a linking process of associating the content that is the acquisition source of the candidate with the corresponding acquisition source information, and
A determination method including a determination step of determining a triple to be registered based on acquisition source information associated with the candidate. An acquisition procedure for acquiring triple candidates including relationship information indicating the relationship between the first target, the second target, and the target from the contents on the network, and the acquisition procedure.
A linking procedure for associating the content that is the acquisition source of the candidate with the corresponding acquisition source information to the candidate acquired by the acquisition procedure, and
A decision program for causing a computer to execute a decision procedure for determining a triple to be registered based on the acquisition source information associated with the candidate.

Images