KR20150030540A - Information fusion method and system - Google Patents

Abstract

The present invention relates to a method and system for fusing information, which fuse information using objects identified from atypical military information and the relationship therebetween. The system for fusing information includes: an object fusion unit configured to extract objects from input information, to identify the relationship between the objects to generate relationship information having a predetermined format, to group the generated relationship information based on a predetermined standard, and to fuse relationship information belonging to a group into a piece of fact information; a knowledge fusion unit configured to infer the fact information using knowledge information stored on ontology, and to generate additional information depending on the inference; and a context fusion unit configured to generate context information by applying at least one of the fact information and the additional information to inference rules defined to observe a specific situation; and an output unit configured to output the generated context information.

Description

INFORMATION FUSION METHOD AND SYSTEM [

The present invention relates to an information fusion method and system, and more particularly, to an information fusion method and system for fusing information by using a relationship between objects and objects identified from atypical military information.

A large amount of data generated in the military field is continuously collected and analyzed for military situation determination and identification of anomalies. It is important to integrate and analyze information in a short period of time in order to determine the situation quickly. However, there is a limit to the amount of information and processing time to be processed by the information analyst, And research on automation through the system is under way.

Conventional information fusion technology is mainly focused on data fusion technology for fusion of regular data collected from sensors. However, in order to judge the actual electric field situation, it is necessary to analyze such sensor data, Information fusion technology is needed.

In addition, there is a need for a technique for collectively managing collected data and for identifying and integrating semantically the same or related data at a level that can retrieve similar documents through text mining.

It is an object of the present invention to provide an information fusion method and system for analyzing collected data and fusing primarily processed information.

It is also an object of the present invention to provide an information fusion method and system for identifying semantic identical or related data to fuse information and infer context information through semantic association.

One embodiment of the present invention relates to an information fusion system, wherein the information fusion system extracts objects from input information, identifies a relationship between the objects, generates relationship information having a predetermined format, An object fusion unit for grouping relationship information on a predetermined basis and fusing relationship information belonging to one group into one fact information, inferring the fact information using knowledge information stored on an ontology, A situation convergence unit for applying the at least one of the fact information and the additional information to a reasoning rule defined for observing a specific situation to generate situation information, And an output unit for outputting the output signal.

In one embodiment, the predetermined format is a triple format including a relative object, a relation, and a target object.

In one embodiment, the object fusion unit aggregates the generated relationship information on the basis of the same reference object.

In one embodiment, in fusing relationship information belonging to one group to one fact information, the object fusion unit may calculate the facts based on the degree of overlapping of the same information among the relation information and the presence or absence of contradictory information And the reliability of the information is calculated.

In one embodiment, the context fusion unit calculates an accuracy index for the context information using the reliability of the fact information and a weight set in the inference rule.

In one embodiment, the output unit outputs the generated context information when the calculated accuracy index satisfies the criterion.

In one embodiment, the knowledge fusion unit may match the object information and the relationship information included in the fact information with a class property and a relation property on the ontology, respectively, And the inference is performed using the knowledge information stored on the ontology.

In one embodiment, the knowledge fusion unit generates the additional information by combining the matched fact information with knowledge information corresponding to an upper or lower concept with respect to the object information.

In one embodiment, the knowledge fusion unit generates the additional information on the basis of the axiom defined in the relation property on the ontology of the matched fact information.

In one embodiment, the reasoning rule is divided into a precondition and a reasoning result, and when the information corresponding to the precondition is input, information corresponding to the reasoning result is defined as the context information .

In one embodiment, the reasoning rule is stored in a semantic web rule language (SWRL) format, which is an inference rule language on the ontology.

In one embodiment of the present invention, the information fusion system further includes a database that is generated by at least one of the object fusion unit, the knowledge fusion unit, and the context fusion unit.

According to another embodiment of the present invention, there is provided an information fusion method, comprising: extracting objects from input information; identifying a relationship between the objects; Creating relationship information having a predetermined format and storing the same in a database, grouping the relation information stored in the database on the basis of the same object, and fusing relationship information belonging to one group into one fact information, Inferring information using intellectual information stored on an ontology and generating additional information according to inference, adding at least one of the fact information and the additional information to inference rules defined for observing a specific situation And generating the status information and outputting the generated status information. It shall be.

In one embodiment, the predetermined format is a triple format including a relative object, a relation, and a target object.

In one embodiment, the method further includes calculating reliability of the fact information based on the degree of overlapping of the same information among the relation information and the presence or absence of inconsistent information.

In one embodiment, the method further comprises calculating an accuracy index for the context information using the reliability of the fact information and the weights set in the inference rule.

In one embodiment, the generated status information is output when the calculated accuracy index satisfies the criterion.

In one embodiment, the step of generating the additional information may include the steps of matching the object information and the relationship information included in the fact information with a class property and a relation property on an ontology, Performing inference using the matched fact information and the knowledge information stored on the ontology, and generating the additional information.

In one embodiment, the reasoning rule is divided into a precondition and a reasoning result, and when the information corresponding to the precondition is input, information corresponding to the reasoning result is defined as the context information .

In one embodiment, the reasoning rule is stored in a semantic web rule language (SWRL) format, which is an inference rule language on the ontology.

 According to the present invention, a large amount of military information that occurs every moment can be integrated and fused semantically by the system. In addition, the analytical judgment logic can be automated through the system, so that the process from information collection to analysis can be processed more quickly and more quickly, which enables analysts to quickly make decisions and make decisions. . In addition, the situation information stored in the database can serve as a knowledge base integrated with information in the military field, and can be utilized variously in the analysis task related to military information.

1 is a block diagram showing a configuration of an information convergence system according to an embodiment of the present invention;
2 is an exemplary diagram for explaining the object fusion unit 120 shown in FIG.
3 is an exemplary diagram for explaining the knowledge fusion unit 140 shown in FIG.
4 is an exemplary diagram for explaining the context fusion unit 160 shown in FIG.
FIG. 5 is a flowchart showing an object fusion method in the information fusion system according to the present invention
6 is a flowchart showing a knowledge fusion method in the information fusion system according to the present invention
FIG. 7 is a flowchart showing a state fusion method in the information fusion system according to the present invention

Hereinafter, the information fusion method and system according to the present invention will be described in detail with reference to the drawings. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In order to facilitate description of the present invention, embodiments of the present invention will be described with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly explain the present invention, parts not related to the description are omitted, and like parts are denoted by similar reference numerals throughout the specification.

1 is a block diagram showing a configuration of an information fusion system according to an embodiment of the present invention.

1, an information fusion system 100 according to the present invention includes an object fusion unit 120, a knowledge fusion unit 140, and a context fusion unit 160. The information fused or generated through the above configuration is stored and managed in the ontology-based context knowledge base 180.

 The present invention fuses information based on the objects (e.g., units, agencies, bases, training, etc.) identified from the military information data of various sources and the attributes of the objects and their relationships among the objects, Information fusion system and method thereof.

 The information fusion system 100 of the present invention includes an object fusion unit 120 for identifying collected information in relation to an object to filter out unnecessary information and fusing the same object on the basis of the same object to generate fact information, A knowledge fusion unit 140 for fusing inferred information according to a semantically connected relation between concepts of information through the combination of the fact information and generating additional information, a decision logic for a situation or an indication to be observed And a situation convergence unit 160 for generating situation information by inferring it.

 In the information fusion system 100 according to the present invention, the object fusion unit 120 identifies object information indicating a specific fact and related information between objects from the collected regular or unstructured military information, And the information is classified according to the same object, time, region, etc., and the duplicated information or inconsistent information is merged into one fact information. In accordance with the degree of redundancy or inconsistency of information in the convergence process, And reflects the reliability of the fused information.

 In the information fusion system 100 according to the present invention, the knowledge fusion unit 140 may associate the abstract concept of each object and the relationship between objects from the fact information fused through the object fusion unit 120, Class information on the ontology model and the class attribute on the ontology model constructed on the ontology model and fused with the knowledge information based on the ontology based on the attribute information, To generate additional information, or to generate additional information through relationship-based reasoning according to the axiom defined in the relationship property.

 In addition, in the information fusion system 100 according to the present invention, the context fusion unit 160 defines a contextual reasoning logic for determining a situation and a symptom to be a target of information fusion, and generates an ontology-based inference rule language SWRL (semantic based rule inference rule for the additional information stored in the knowledge fusion unit 140. The reasoning rule is based on the information reliability and reasoning rules, The status information including the accuracy index of the result is generated and the abnormal situation or the abnormal symptom is determined according to the criterion that affects the determination of the overall field situation and the corresponding situation information and the related information are provided to the user .

The database 180 stores knowledge information corresponding to an ontology-based knowledge database, and stores fact information generated by the object fusion unit 120, additional information generated by the knowledge fusion unit 140, May store status information generated by the user.

Meanwhile, the database 180 may include, for example, a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (for example, SD (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read- memory, a magnetic memory, a magnetic disk, or an optical disk.

Although not shown in the figure, the information fusion system 100 includes an output unit, and status information generated by the context fusion unit 140 may be output by the output unit.

The output unit is for generating an output related to visual, auditory or tactile sense, and may include a display unit, an audio output module, an alarm unit, and a haptic module.

As described above, according to the information convergence system 100 of the present invention, it is possible to provide faster and more automated situation convergence information in analyzing a large amount of military information that is frequently generated and determine the situation, .

Hereinafter, the ontology will be described in detail, and the concept of information fusion processed in each configuration included in the information fusion system 100 will be described with reference to FIG. 2 to FIG. 5. FIG.

First, ontology is a conceptual and computer-mediated model of what people see, hear, feel and think about the world through discussions with each other. It is a technology defined by the enemy. Since the ontology represents the agreed knowledge, the ontology is not limited to an individual but is a concept that group members all agree. And there are many stereotypes because the program must be understandable.

Ontology is defined as a set of formal vocabularies that describe the relationship between concepts and concepts belonging to a specific domain as a data model that represents a specific domain in computer science and information science. For example, if a formal vocabulary is used to describe such things as the relationship of a species between a creature and a creature classified as "species-genus-and-species-genus-river-gate-system", and the relationship between English words, it can be called an ontology. An ontology, which is a collection of vocabularies described in a formal language, is used for reasoning and inference.

Due to the development of the information age, the web has appeared, which has led to the transformation of all areas of society including information search, knowledge management and general commerce. Due to the rapid development of the web, the expansion of the scope of the search has required more sophisticated search, and it has facilitated the development of an intelligent information retrieval system. The need for a new tool for information retrieval that can effectively manage web resources has emerged. The ontology is a tool that can implement the semantic web and semantically connect the knowledge concept.

The definition of an ontology that is mainly cited is' an explicitly formal specification (c.f. Tom Gruber, 1993) to conceptualize a field of interest. In other words, it is a clear and detailed explanation of meaning, the use of knowledge, etc. in order to find a common point between information and to express it in one set or category.

Semantic Web technology represents the understanding of the language in the head of a person in a computer language and makes it available to the computer. It is implemented in a web with a special distributed environment. This is to allow the machine to better understand and respond to people's needs such as information retrieval. To be able to truly communicate between people and machines, the machine must be able to understand the language to the level understood by the person. However, the current web, made up of HTML-formatted documents, serves to inform people, but a computer program can not accurately grasp the contents of each document. The ontology starts from this awareness.

Ontology is a technology that makes the same thing as a human being's concept in a computer as a kind of database-like form, while the way to grasp and understand the features contained in them while experiencing each object or events in the world is called conceptualization . The purpose of the ontology to help the program and human beings to share knowledge is to clearly define and describe the concept of the resource that is the subject of the information system so that more accurate information can be found. The ontology-based system has advantages such as the role of information content structure, knowledge sharing and reuse in the field, specification of constraints and assumptions of the domain, and separation of knowledge and processes.

The constituent elements of an ontology can be classified into classes, instances, relations, and properties.

Classes can explain that a class generally refers to a name that we attach to things or concepts. For example, "keyboard," "monitor," and "love" are all classes.

An instance refers to a substance or an object, or a substance or event. In other words, "LG ST-500 Ultra Slim Keyboard", "Samsung Sync Master Wide LCD Monitor", "Love of Romeo and Juliet" is generally an instance. The distinction between classes and instances can vary greatly depending on the application and purpose of use. That is, an object of the same expression can be a class in some cases, or an instance in other cases.

A property is a connection of a class or instance to a specific value to indicate a specific nature or orientation of the class or instance. For example, you can define attributes such as "hasSize" to indicate that "Samsung Sync Master Wide LCD monitor is XX inches."

A relation is a relation existing between a class and an instance. It is generally classified into a taxonomic relation and a non-taxonomic relation.

A taxonomic relation is a hierarchical representation that classifies concepts into more broad concepts and specific concepts to classify classes and instances. For example, the "isA" relationship is used to express the conceptual relationship between concepts such as "man is an animal".

A non-taxonomic relation is a relationship that is not a taxonomic relation. For example, "being healthy by exercise" is expressed using the "cause" relationship (causal relationship).

Languages that are mainly used in ontologies include resource description framework (RDF), ontology web language (OWL), and semantic web rule language (SWRL). RDF is an advanced form of XML, consisting of a subject, an object, and a predicate, which simply represents a concept or relationship between instances. Generally, a resource description framework (RDF) is often used to estimate a generic application that does not require complex constraints. OWL (ontology web language) is a language that includes the hierarchy between relationships, and logical constraints within a relationship instance. It is used when precise and logical reasoning is needed. The semantic web rule language (SWRL) is used to define rules for reasoning.

FIG. 2 is an exemplary diagram for explaining the object fusion unit 120 shown in FIG. 1. Referring to FIG. The object fusion unit 120 identifies the collected information in relation to the object, fuses the same information, and performs object fusion processing to generate the fact information.

2, the object fusion unit 120 identifies objects and relationships from collected information, and acquires the collected information in a triple format such as [MIG-23] - [take off] - [ . The triple form may be composed of a singular object, a relation, and a target object (subject-predicate-object).

The converted information is grouped into [MIG-23], which is the same object, and the overlapping information or contradictory contradiction information is filtered by the same content to converge semantically the same information. The result of the above process is that information on [MIG-23] is fused into fact information from a plurality of individual information.

FIG. 3 is an exemplary diagram for explaining the knowledge fusion unit 140 shown in FIG. 1. FIG. The knowledge fusion unit 140 fuses the fact information generated by the object fusion unit 120 with the knowledge information based on the ontology, and fuses the inferred information according to the semantically connected relation between the concepts of the information, The knowledge fusion process is performed.

At this time, knowledge information means information stored in an ontology-based database.

On the other hand, referring to FIG. 6, for example, there is a concept of an upper and a lower concept defined as a subClassOf relation in a concept on an ontology in which an object [00] is matched in fact information generated by the object fusion unit 120. It is possible to generate new additional information for [0th Army], [000th Battalion], etc., inferred by these upper and lower relationships.

In addition, the relationship attribute on the ontology in which the relation [in the fact information] is matched is defined as an inverseOf relation with the relation attribute [holds]. According to this logic of relationship reasoning, new additional information such as [Hwangju base] - [hold] - [MIG-23] can be generated.

The additional information thus generated is information in which the related knowledge information is fused together with the fact information stored in the database 180 (see FIG. 1).

4 is an exemplary diagram for explaining the context fusion unit 160 shown in FIG. The situation fusion unit 160 defines a reasoning rule for a situation or symptom to be observed as an inference rule, and performs context fusion processing for generating context information by reasoning the reasoning rule.

Referring to FIG. 7, for example, in order to integrate a situation, first, a reasoning rule is defined as a reasoning rule for judging a specific situation such as an example of a reasoning rule [a fighter A at a specific base B]. The reasoning rule defines the part corresponding to the precondition and the part corresponding to the inference result. In the example, the assumption that [the fighter A takes off from a specific base B, and the ~ G region lies to the south than the F region] If the conditional information is present, we have defined the result information to be inferred as ['there is a sign of' fighter forward deployment].

At this time, the reasoning rules on the system are expressed in a rule language such as a semantic web rule language (SWRL), which is a type that can be interpreted and processed by an ontology-based rule inference engine, not a natural language sentence like the example.

At least one of the fact information generated by the preceding object fusion unit 120 and the additional information fused by the knowledge fusion unit 140 is determined based on whether or not the inference rule logic defined in the context fusion unit 160 is satisfied And thus infer the new situation information.

In the example, the factual information [MIG-23-Takeoff-Bukhong Base] is found by relationship inference that [MIG-23] is a subclause of [fighter] Take off] is satisfied. In addition, when all of the logic defined in the inference rules is satisfied, along with various existing knowledge information such as the location of the [base station], the location area, the type of the stationed station, coordinates of the location area, - 'Fighter aircraft front station deployment'], as well as information about the specific situation or symptom defined by the inference result. Such inferred outcome information may include relevant information or accuracy indices that may aid in the assessment of the situation.

Hereinafter, the information fusion processing procedure in the information fusion system 100 will be described in detail.

5 is a flowchart illustrating an object fusion method in an information fusion system according to the present invention.

5, when the newly collected information is input to the object fusion unit 120 (200), the object fusion method identifies (201) object information from the corresponding information and identifies relationship information between the objects (202).

The relationship between objects is expressed by converting the identified object information and relationship information into a triple format (203).

The triple form is a form that consists of three types of fact information: subject-predicate-object (object), which is the basic unit of the knowledge base. These triple data can be mapped to class attributes and relation attributes in the ontology model to express the relationship between knowledge information. Also, the triple representation is a format in which the system can load large capacity information at high speed through the triple storage of the database 180, such as n-triple data.

The ontology instance is generated by mapping the data converted into the triple format to the concept of the ontology model (204).

In order to fuse the same information semantically, the generated instance information is arranged in the system, and the instance information mapped according to the category of the situation is arranged (205).

The arranged information is grouped based on the same object (206), and time information and local information are compared and grouped based on the same time and region (207).

If it is determined that the same information is included in the context information 208, the duplicated information is merged into one information 209, and the accuracy index such as the reliability or importance of the information is increased 210.

Identifies contradictory information by comparing configuration information within the context category. Identification of contradictory information can be handled through comparison of the object constituting the instance with relation information. In one embodiment, when information is concurrently generated (211) with the same objects connected to each other, or with the same predicate attribute that can be associated with an object, (212) when information associated with different objects are generated at the same time.

The information identified through the comparison of contradictions is filtered (213) and the accuracy index, such as the reliability or importance of the information, is reduced (214).

The fused object fusion information is provided to the knowledge fusion unit 120 (215).

6 is a flowchart showing a knowledge fusion method in the information fusion system according to the present invention.

Referring to FIG. 6, the knowledge fusion unit 140 classifies the object and the relation information from the inputted information (301) when the instance information fused according to the object is inputted from the object fusion unit 120 (300).

If there is an object of the same concept defined by the sameAs relation on the ontology model for each object (302), additional information is generated through reasoning about the objects (303).

The type (class) of each object is identified (304). If there is a constraint condition (305) for the corresponding relations in the classes, additional information is generated through inference of the class constraint condition (306).

If an upper / lower class according to the subClassOf relation of the classes exists (307), additional information is generated through inference based on the hierarchical relationship (308).

(309), and if there is an axiom (310) such as subPropertyOf, transitiveOf, inverseOf, symmetric defined in the respective elation properties (310) And generates additional information (311).

The inference result information generated through the above process is stored and managed in the database 180 together with the existing object fusion information, and is provided to the context fusion unit 160 (312).

FIG. 7 is a flowchart showing a state convergence method in an information convergence system according to the present invention.

7, when the knowledge fusion information expanded through the combination of the information fused from the collected information and the knowledge model is input (400) from the knowledge fusion unit 140, the context fusion unit 160 extracts the corresponding information (401) the relevant inference rules for contextual reasoning.

If there is an inference rule associated with the input information (402), additional information is generated through ontology-based rule inference processing for each inference rule (403).

The rule inferencing process is performed on existing fusion information and additional generated inference information, and it is repeated until there is no more information to be added.

If there is a change in the context information due to the inferred result (404), the change of the context is identified through calculation of the weight of the context information (405).

Weights of contextual information are classified according to the number of combined fact information in the context category and the importance weight of the fact information. Situational categories are organized hierarchically, and high-status categories are also calculated by summing scores of sub-category categories to identify various situations.

A determination is made as to whether there is an associated impact depending on the additional identification or change in the situation (406).

The impact assessment is to calculate the impact level associated with one or more of the identified category of events, and the impact level is derived from the assessment of the impact score by combining the score derived and the importance of each category by identifying the context category.

If there is an influence change such as an abnormal symptom identification (407), the corresponding information and related information are provided to the user (408).

As described above, according to the present invention, a large amount of military information generated every moment can be integrated and fused semantically by the system. Furthermore, by automating the judgment logic of the analyst through the system, it is possible to process the process from the information collection to the analysis more quickly and quickly, thereby enabling the analyst to promptly judge the situation and make decisions. In addition, the merged fact information, additional information, and situation information can be utilized variously in the analysis task relating to military information as a knowledge base integrated with information in the military field.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention.

Claims (20)

Extracting objects from the input information, generating relationship information having a predetermined format by identifying the relationship between the objects, grouping the generated relationship information according to a predetermined reference, and associating the relationship information belonging to one group with one An object fusion unit that fuses into fact information;
A knowledge fusion unit for inferring the fact information using knowledge information stored on an ontology and generating additional information according to inference;
A situation fusion unit for applying the at least one of the fact information and the additional information to a reasoning rule defined for observing a specific situation to generate situation information; And
And an output unit outputting the generated status information. The method according to claim 1,
Wherein the predetermined format is a triple format composed of a relative object, a relation, and a target object. 3. The method of claim 2,
Wherein the object fusion unit comprises:
And collecting the generated relationship information based on the same subject object. The method of claim 3,
Wherein the object fusion unit comprises:
Wherein reliability of the fact information is calculated based on the degree of overlapping of identical information among the relation information and the presence or absence of contradictory information in fusing relationship information belonging to one group to one fact information Fusion system. 5. The method of claim 4,
The situation fusion unit includes:
Wherein the accuracy index of the context information is calculated using the reliability of the fact information and the weight set in the inference rule. 6. The method of claim 5,
The output unit includes:
And outputs the generated context information if the calculated accuracy index meets a criterion. The method according to claim 1,
The knowledge fusion unit includes:
The object information and the relationship information included in the fact information are respectively matched with a class property and a relation property on the ontology,
And performs inference using the matched fact information and knowledge information stored on an ontology. 8. The method of claim 7,
The knowledge fusion unit includes:
And combines the matched fact information with knowledge information corresponding to an upper or lower concept with respect to the object information to generate the additional information. 8. The method of claim 7,
The knowledge fusion unit includes:
And generates the additional information based on the axiom defined in the relation property on the ontology with the matching fact information. The method according to claim 1,
Wherein the reasoning rule is divided into a precondition and an inference result, and when information corresponding to the precondition is input, the inference rule is defined to infer the information corresponding to the inference result as the context information. 11. The method of claim 10,
Wherein the inference rule is stored in a semantic web rule language (SWRL) format, which is an inference rule language on the ontology. The method according to claim 1,
Further comprising: a database for storing information generated by at least one of the object fusion unit, the knowledge fusion unit, and the context fusion unit. Extracting objects from the input information, identifying relationships between the objects, generating relationship information having a predetermined format using the objects and the relationship, and storing the information in a database;
Grouping the relational information stored in the database on the basis of the same object and fusing the relational information belonging to one group into one fact information;
Inferring the fact information using knowledge information stored on an ontology and generating additional information according to inference;
Generating status information by applying at least one of the fact information and the additional information to inference rules defined for observing a specific situation; And
And outputting the generated context information. 14. The method of claim 13,
Wherein the predetermined format is a triple format composed of a relative object, a relation, and a target object. 14. The method of claim 13,
Further comprising the step of calculating the reliability of the fact information based on the degree of overlapping of the same information among the relation information and the presence or absence of contradictory information. 16. The method of claim 15,
Further comprising the step of calculating an accuracy index for the context information using the reliability of the fact information and the weight set in the inference rule. 17. The method of claim 16,
Wherein the generated context information is output when the calculated accuracy index satisfies the criterion. 14. The method of claim 13,
Wherein the generating the additional information comprises:
Matching the object information and the relationship information included in the fact information with a class property and a relation property on an ontology; And
Performing inference using the matched fact information and knowledge information stored on an ontology, and generating the additional information. 14. The method of claim 13,
Wherein the reasoning rule is divided into a precondition and an inference result, and when the information corresponding to the precondition is input, the inference rule is defined to infer the information corresponding to the inference result as the context information. 20. The method of claim 19,
Wherein the inference rule is stored in a semantic web rule language (SWRL) format, which is an inference rule language on the ontology.

Images