CN111339299B - Construction method and device of domain knowledge base - Google Patents
Construction method and device of domain knowledge base Download PDFInfo
- Publication number
- CN111339299B CN111339299B CN202010125394.9A CN202010125394A CN111339299B CN 111339299 B CN111339299 B CN 111339299B CN 202010125394 A CN202010125394 A CN 202010125394A CN 111339299 B CN111339299 B CN 111339299B
- Authority
- CN
- China
- Prior art keywords
- knowledge
- knowledge base
- data
- domain
- reasoning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
- G06N5/025—Extracting rules from data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the application discloses a method and a device for constructing a domain knowledge base, wherein the method comprises the following steps: obtaining entity data according to a preset expert knowledge base, wherein the entity data comprises: concepts and attributes of entities and relationships between entities; extracting text information from the entity data by using a preset extraction technology, and taking the extracted text information as body data; and carrying out knowledge construction on the ontology data by adopting the established reasoning knowledge base, and forming a new domain knowledge base according to the new knowledge obtained by the knowledge construction. By the embodiment, a low-cost knowledge base based on knowledge reasoning is established, an automatic decision making process is realized, the data quality and flexibility of the atlas are improved, and the difficulty and cost of data management are reduced.
Description
Technical Field
The embodiment of the invention relates to a data processing technology, in particular to a method and a device for constructing a domain knowledge base.
Background
Ontology technique
The most common classification method for ontologies is to divide these numerous ontologies into five types, namely a domain ontology, a general or common sense ontology, a knowledge ontology, a linguistic ontology and a task ontology, according to the subject matter of the application of the ontology. And according to the hierarchy and domain dependence of the ontology, guarino et al divide it into four classes: top layer body, field body, task body and application body.
1. Top layer body: the general concepts and relationships between the concepts, such as space, time, events, behavior, etc., are studied, independent of the specific application, completely independent of the defined domain, and thus can be shared over a wide range.
2. Domain body: the relationship between concepts in a specific domain is studied.
3. Task body: some general tasks or related reasoning activities are defined to express concepts and relationships between concepts within a particular task.
4. Application body: to describe some specific applications, reference may be made to specific concepts within domain ontologies, as well as to concepts that appear within task ontologies.
(II) Domain knowledge base
In the research of performing topic analysis of text and content analysis of text, "domain knowledge" is an indispensable basic knowledge, and "domain knowledge base" is an effective way of managing "domain knowledge" of a system, so that the research work of constructing the domain knowledge base has extremely profound significance.
(III) knowledge reasoning
The knowledge graph-oriented knowledge reasoning is related to complete the deep analysis and reasoning of the data. In essence, knowledge graph-oriented knowledge reasoning refers to reasoning out new knowledge or identifying incorrect knowledge in a knowledge graph by adopting certain methods according to the existing knowledge in the knowledge graph. Accordingly, it includes two aspects: knowledge graph completion (knowledge graph completion, knowledge base completion) and knowledge graph denoising (knowledge graph refinement, knowledge graph cleaning). Knowledge graph completion also comprises tasks such as link prediction (link prediction), entity prediction (entity prediction), relation prediction (relation prediction), attribute prediction (attribute prediction) and the like.
At present, the industry still lacks the work of systematically and deeply combing and summarizing knowledge-based reasoning research on knowledge maps. The main problems are:
1. the data management cost of the atlas is high;
2. expert system reasoning capability is not universal;
3. the fact map is not in the place of the industrial field;
4. domain-specific knowledge base construction relies entirely on expert experience.
Disclosure of Invention
The embodiment of the invention provides a method and a device for constructing a domain knowledge base, which can establish a knowledge base with low cost and based on knowledge reasoning, realize an automatic decision process, improve the data quality and flexibility of a map and reduce the difficulty and cost of data management.
In order to achieve the purpose of the embodiment of the present invention, the embodiment of the present invention provides a method for constructing a domain knowledge base, where the method may include:
obtaining entity data according to a preset expert knowledge base, wherein the entity data can comprise: concepts and attributes of entities and relationships between entities;
extracting text information from the entity data by using a preset extraction technology, and taking the extracted text information as body data;
and carrying out knowledge construction on the ontology data by adopting the established reasoning knowledge base, and forming a new domain knowledge base according to the new knowledge obtained by the knowledge construction.
In an exemplary embodiment of the present application, the knowledge construction may include: and performing mutual verification and mutual check between the ontology data, and indicating errors in the process of abstracting the ontology data to knowledge.
In an exemplary embodiment of the present application, the method may further include: and in the knowledge construction process, the linkage of the new knowledge and the current knowledge base is realized through a knowledge interrupter.
In an exemplary embodiment of the present application, the implementing, by the knowledge interrupter, the linkage of the new knowledge and the current knowledge base may include:
detecting whether the knowledge which is the same as the new knowledge to be formed exists in the current knowledge base and whether the knowledge contradicted with the new knowledge to be formed exists or not when the new knowledge is formed according to the ontology data;
if the current knowledge base has the same knowledge as the new knowledge to be formed, interrupting the new knowledge abstract process through the knowledge interrupter;
if the knowledge contradicted with the new knowledge to be formed exists in the current knowledge base, the contradictory knowledge is recorded and sent to the research staff for research.
In an exemplary embodiment of the present application, the knowledge construction of the ontology data using the created inference knowledge base may include:
acquiring an evolution time point from a time line in an evolution process of the domain ontology;
importing the ontology data of the evolution time point into the reasoning knowledge base as real-time data;
the reasoning knowledge base identifies the problem characteristics according to the real-time data, and matches and classifies the identified problem characteristics with the fact experience knowledge base and/or the common sense knowledge base to obtain a decision scheme;
evolving the ontology data according to one or more real-time data of the evolution time points and corresponding decision schemes acquired by the reasoning knowledge base, and acquiring a decision target of the domain ontology; and constructing new knowledge according to the decision target.
In an exemplary embodiment of the present application, the acquiring the evolution time point from the timeline in the domain ontology evolution process may include: selecting a time point at which at least one of the preset latitude attributes of the body data changes compared with the previous time point;
wherein, the preset latitude attribute comprises any one or more of the following: classification, grading and staging.
In an exemplary embodiment of the present application, the inference knowledge base may include an inference module; the reasoning knowledge base identifies the problem characteristics according to the real-time data, and the matching of the identified problem characteristics with the fact experience knowledge base and/or the common sense knowledge base in the reasoning knowledge base to obtain a decision scheme comprises the following steps:
the reasoning module performs context awareness, context analysis and context deduction according to the real-time data, identifies the problem features, and matches the identified problem features with a plurality of case features in the fact experience knowledge base and/or common sense knowledge base to obtain a solution; carrying out scheme classification according to the scene deduction result and the solution, and obtaining a matching scheme according to the scene perception result and the scheme classification; and acquiring an optimal scheme according to the matching scheme and a preset treatment scheme, and taking the optimal scheme as a final decision scheme.
In an exemplary embodiment of the present application, the inference module is further configured to complete labeling of the problem feature through a feature dictionary;
the completing the labeling of the problem features by the reasoning module through the feature dictionary may include: isolating the front and rear problem features through the marking points, and carrying out knowledge representation on each marking point; the knowledge representation of the annotation point may be described in one or more of the following: an input element, a state element, and an output element;
the input element includes: an environmental input and a control input;
the state element includes: the nature and characteristics of the scene at a specific moment;
the output element includes: the impact of the state on the external environment and the loss of events due to the state change.
In an exemplary embodiment of the present application, the feature dictionary classifies each feature problem unit generated after labeling the problem feature into a unique code in the feature dictionary, assigns a unique ID number to the fact experience knowledge base and/or the common sense knowledge base for storage, and generalizes the event topic to generate a knowledge representation of the problem feature, and/or,
the reasoning module is also used for determining the relation among the problem features according to the time sequence of the plurality of problem feature units generated after marking in the associated event and acquiring the whole evolution course of the ontology data; the relationship may include a concurrent relationship and a tandem relationship.
The application also provides a construction device of the domain knowledge base, which can comprise a processor and a computer readable storage medium, wherein the computer readable storage medium stores instructions, and when the instructions are executed by the processor, the construction method of the domain knowledge base is realized.
Compared with the prior art, the embodiment of the application can comprise the following steps: obtaining entity data according to a preset expert knowledge base, wherein the entity data can comprise: concepts and attributes of entities and relationships between entities; extracting text information from the entity data by using a preset extraction technology, and taking the extracted text information as body data; and carrying out knowledge construction on the ontology data by adopting the established reasoning knowledge base, and forming a new domain knowledge base according to the new knowledge obtained by the knowledge construction. By the embodiment, a low-cost knowledge base based on knowledge reasoning is established, an automatic decision making process is realized, the data quality and flexibility of the atlas are improved, and the difficulty and cost of data management are reduced.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate and do not limit the invention.
FIG. 1 is a flowchart of a method for constructing a domain knowledge base in accordance with an embodiment of the present invention;
FIG. 2 is a schematic diagram of a method for knowledge construction of ontology data using a created inference knowledge base according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of the overall architecture of an inference knowledge base in accordance with an embodiment of the present invention;
FIG. 4 is a schematic diagram of the evolution process according to an embodiment of the present invention;
fig. 5 is a block diagram of a construction device of a domain knowledge base according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail hereinafter with reference to the accompanying drawings. It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be arbitrarily combined with each other.
The steps illustrated in the flowchart of the figures may be performed in a computer system, such as a set of computer-executable instructions. Also, while a logical order is depicted in the flowchart, in some cases, the steps depicted or described may be performed in a different order than presented herein.
The prior art comprises a structured data management technology, an unstructured data management technology, a knowledge graph construction technology, an ontology construction technology, a knowledge base construction method based on expert experience, an expert system technology and the like. The prior art has a number of applications in industry, but each has shortcomings including:
1. the construction and the use of a common sense knowledge base, an experience knowledge base and an reasoning knowledge base are separated; cannot cooperate to form a unified knowledge base;
2. the prior knowledge base takes retrieval as a main application direction, and the knowledge base is limited to play a role;
3. the knowledge base constructed at present is static, and once constructed, the structure and the knowledge range are basically determined, the updating efficiency is low, and the dynamic adjustment capability is not provided;
4. knowledge quality in a knowledge base cannot be effectively monitored and dynamically optimized, and knowledge quality problems seriously affect the effectiveness of knowledge application;
5. the knowledge graph is used as a knowledge representation form, has a single expression form and lacks dynamic capability.
The core value of the application is that the innovative knowledge base construction method is applied, and the defect of the current knowledge base construction method is overcome. The innovative concept can include: a new method for hierarchical classification of a knowledge base, a new method for constructing a knowledge base based on knowledge reasoning, a new method for constructing a knowledge base based on domain ontology, a new method for constructing a dynamic knowledge base and a knowledge base innovation application based on knowledge reasoning.
The embodiment of the invention provides a construction method of a domain knowledge base, as shown in fig. 1, the method can comprise S101-S103:
s101, acquiring entity data according to a preset expert knowledge base, wherein the entity data can comprise: concepts and attributes of entities and relationships between entities.
S102, extracting text information from the entity data by using a preset extraction technology, and taking the extracted text information as body data.
In the exemplary embodiment of the present application, there are two starting points formed by domain ontology (i.e., ontology data of a knowledge domain), namely, expert experience is one, and the expert summarizes the structure of domain knowledge, abstracts the concepts and attributes of the entities, and abstracts the relationships between the entities to form an expert knowledge base; and secondly, automatically extracting the text information in the expert knowledge base by using a text information extraction technology to automatically form body data. The extraction technique of the text information may be NLP (natural language processing) technique.
In the exemplary embodiment of the application, a rational map can be used as a supplement to the knowledge map, and the rational map adopts 6 elements to represent knowledge, so that the knowledge map has better expressive ability and flexibility than the knowledge map.
S103, knowledge construction is carried out on the ontology data by adopting the established reasoning knowledge base, and a new domain knowledge base is formed according to the new knowledge obtained by the knowledge construction.
In an exemplary embodiment of the present application, the knowledge construction may include: and performing mutual verification and mutual check between the ontology data, and indicating errors in the process of abstracting the ontology data to knowledge.
In an exemplary embodiment of the present application, after the ontology information is obtained, knowledge construction is performed on the data in the database using an automatic construction technique (e.g., the previously constructed inference knowledge base described above). The method is different from the original knowledge engineering in that the ontology information can be mutually verified and checked, so that errors in the process of data to knowledge abstraction are pointed out, and the knowledge quality is improved.
In an exemplary embodiment of the present application, the method may further include: and in the knowledge construction process, the linkage of the new knowledge and the current knowledge base is realized through a knowledge interrupter.
In an exemplary embodiment of the present application, the implementing, by the knowledge interrupter, the linkage of the new knowledge and the current knowledge base may include:
detecting whether the knowledge which is the same as the new knowledge to be formed exists in the current knowledge base and whether the knowledge contradicted with the new knowledge to be formed exists or not when the new knowledge is formed according to the ontology data;
if the current knowledge base has the same knowledge as the new knowledge to be formed, interrupting the new knowledge abstract process through the knowledge interrupter;
if the knowledge contradicted with the new knowledge to be formed exists in the current knowledge base, the contradictory knowledge is recorded and sent to the research staff for research.
In the exemplary embodiment of the present application, during knowledge extraction, there is a linkage relationship between the database of data and the current knowledge base (i.e., the existing knowledge base), that is, the existing knowledge does not repeatedly perform data processing, and indicates incorrect knowledge, and indicates a contradictory relationship between the data in the database and the existing knowledge.
In the exemplary embodiment of the present application, the linkage of the database and the knowledge base is implemented mainly by means of a knowledge interrupter. That is, when partial knowledge is formed in the data in the database, firstly, inquiring whether the same knowledge exists in the existing knowledge base, and if so, interrupting the abstract process of the knowledge; if a conflict is found, the conflict is recorded and sent to a researcher for manual research.
In the exemplary embodiment of the application, the new knowledge can be saved in a graph database after being formed, and is displayed in the form of a knowledge graph and a rational graph.
In an exemplary embodiment of the present application, how knowledge building of ontology data is implemented using an inference knowledge base will be described in detail below.
In an exemplary embodiment of the present application, as shown in fig. 2, the knowledge construction of the ontology data using the created inference knowledge base may include S201-S204:
s201, acquiring an evolution time point from a time line in the domain ontology evolution process.
In an exemplary embodiment of the present application, a time point in the evolution process may be first determined, and an important time point, i.e., an evolution time point, may be selected therefrom.
In an exemplary embodiment of the present application, the acquiring the evolution time point from the timeline in the domain ontology evolution process may include: selecting a time point at which at least one of the preset latitude attributes of the body data changes compared with the previous time point;
wherein, the preset latitude attribute comprises any one or more of the following: classification, grading and staging.
In an exemplary embodiment of the present application, the principle of selecting a time point for dividing a problem feature may be that at least one of three dimensional attributes of classification, grading and staging of an incident (such as an domain ontology evolution event) is changed at the time point relative to a previous time point.
S202, importing the ontology data of the evolution time point into the reasoning knowledge base as real-time data.
S203, the reasoning knowledge base identifies the problem features according to the real-time data, and matches and classifies the identified problem features with the fact experience knowledge base and/or the common sense knowledge base to obtain a decision scheme.
In an exemplary embodiment of the present application, the overall architecture of the inference knowledge base may be as shown in fig. 3, where the inference knowledge base is established for decision-makers to cope with decisions, and the construction of the inference knowledge base depends on the common sense knowledge base and the fact experience knowledge base.
In the exemplary embodiment of the present application, the inference knowledge base is a bridge for interactive interaction between decision analysis and the real world, so that decisions are established on the basis of real-time information (real-time ontology data), and uncertainty, abnormal and abrupt behaviors inside the system and outside the environment can be reflected in the decisions in real time. The inference knowledge base not only avoids the inefficiency of continuous optimization decision-making, but also realizes the real-time monitoring and intervention of events.
In the exemplary embodiment of the present application, the reasoning basis of the reasoning knowledge base is to identify the problem features of the decision, so as to match the security features of the fact experience knowledge base, thereby obtaining the final decision scheme.
In an exemplary embodiment of the present application, as shown in fig. 3, the inference knowledge base may include an inference module; the reasoning knowledge base identifies the problem characteristics according to the real-time data, and the matching of the identified problem characteristics with the fact experience knowledge base and/or the common sense knowledge base in the reasoning knowledge base to obtain a decision scheme comprises the following steps:
the reasoning module performs context awareness, context analysis and context deduction according to the real-time data, identifies the problem features, and matches the identified problem features with a plurality of case features in the fact experience knowledge base and/or common sense knowledge base to obtain a solution; carrying out scheme classification according to the scene deduction result and the solution, and obtaining a matching scheme according to the scene perception result and the scheme classification; and acquiring an optimal scheme according to the matching scheme and a preset treatment scheme, and taking the optimal scheme as a final decision scheme.
In an exemplary embodiment of the present application, a schematic diagram of a decision-making inference process based on an inference knowledge base may be shown in fig. 3. The time point in the evolution process can be determined first through step S201, the important time point is selected, and then the labeling of the problem features is completed by using the feature dictionary.
In an exemplary embodiment of the present application, the inference module is further configured to complete labeling of the problem feature through a feature dictionary;
the completing the labeling of the problem features by the reasoning module through the feature dictionary may include: isolating the front and rear problem features through the marking points, and carrying out knowledge representation on each marking point; the knowledge representation of the annotation point may be described in one or more of the following: an input element, a state element, and an output element;
the input element includes: an environmental input and a control input;
the state element includes: the nature and characteristics of the scene at a specific moment;
the output element includes: the impact of the state on the external environment and the loss of events due to the state change.
In the exemplary embodiment of the present application, knowledge representation of the inference process may be further refined by combining with a decision target, where the labeling point r of the inference isolates the features of the front and rear problems, is a key of the knowledge entity from the time t knowledge instance to the time t to t+ [ delta ] t, and the knowledge representation may be described from three aspects of input element, state element, and output element, and describes the knowledge representation of the inference process, and connects the segments of the front and rear adjacent time. A input element: the research category of the input elements is two aspects of environment input and control input. B state element: the state elements are repeated to describe the property and the characteristic of the scene at a specific moment, and the two aspects of the state and the scene life cycle of the disaster-bearing moment are inspected. C output element: the output element is embodied in two aspects of influence of the state on the external environment and event loss caused by state change.
In an exemplary embodiment of the present application, the feature dictionary classifies each feature problem unit generated after labeling the problem feature into a unique code in the feature dictionary, assigns a unique ID number to the fact experience knowledge base and/or the common sense knowledge base for storage, and generalizes the event topic to generate a knowledge representation of the problem feature, and/or,
in an exemplary embodiment of the present application, as shown in fig. 4, the inference module is further configured to determine a relationship between problem features according to a time sequence of occurrence of a plurality of problem feature units generated after labeling in an associated event, and obtain an evolution whole course of the ontology data; the relationship may include a concurrent relationship and a tandem relationship.
S204, evolving the ontology data according to one or more real-time data of the evolution time points and corresponding decision schemes acquired by the reasoning knowledge base, and acquiring a decision target of the domain ontology; and constructing new knowledge according to the decision target.
In the exemplary embodiment of the present application, after the important time point (time t 1) is acquired, the ontology data at time t1 may be imported into a created inference knowledge base, and the data state is updated in the inference knowledge base, at this moment, the decision targets (such as targets g1, g2, gn) obtained by the last domain ontology evolution are acquired, and at time t1+ [ delta ] t, the decision scheme acquired by the inference knowledge base according to the identified problem feature is acquired, and the decision targets obtained by the last evolution and the decision scheme of this time are combined to perform target evolution, so as to obtain new decision targets (such as targets g4, g5, gm), that is, the domain ontology after further evolution is acquired.
In an exemplary embodiment of the present application, the role of knowledge reasoning through the reasoning knowledge base in the construction process in the knowledge base includes at least the following two points:
1. automatically forming new knowledge:
the formation of new knowledge relies on a wider range of combinations of all fields (or features) in the database (where the body data is stored), not limited to analysis rules indicated by the expert, but rather uses a way similar to random combinations to compare, correlate, analyze the data in the database to form new knowledge.
The construction of knowledge in the process limits the excessive combination of data and ensures the feasibility of knowledge construction automation. Illustrating: the personnel in the original knowledge base have father-son relationship but have no grandfather-son relationship, and the relationship between the designer and the person can have a combination relationship, such as grandfather-son relationship, when the ontology knowledge is constructed. When the knowledge base is automatically constructed, the system automatically abstracts grandparent relations from the database, and the quantity and depth of knowledge are increased.
2. Correcting contradictions and errors in knowledge:
in the knowledge construction process, because of human input reasons, expert classification reasons, automatic construction error reasons and the like, the knowledge in the knowledge base is contradicted or erroneous, and more commonly, the knowledge is stale. Using knowledge reasoning techniques, contradictory errors in the knowledge base are automatically discovered and troubleshooted. The actual technical implementation is that the similarity comparison is carried out on the new knowledge and the old knowledge periodically, and the comparison analysis is carried out on the knowledge of the same type and the same content. The specific method can comprise the following steps: when the content of the same kind of data is inconsistent, especially the content of partial entities is completely consistent, but the overall knowledge is inconsistent, the contradictory knowledge is considered.
In the exemplary embodiment of the application, a ontology construction method conforming to the domain features is established in an innovative manner based on an inference knowledge base, the domain knowledge base is continuously established and updated according to the domain ontology and the evolved domain ontology, and a domain knowledge base based on knowledge inference is established in a low-cost and high-usability manner, so that a cross-industry knowledge base system with the domain features is formed, and an automatic decision making process with knowledge basis is provided for decision making; the data quality of the knowledge graph is improved, the flexibility of the knowledge graph is improved, the capability of an expert system is fused, the dependence on expert experience is reduced by combining the innovative method of the fact graph, and the difficulty and cost of data management are reduced.
In the exemplary embodiment of the present application, the present application provides a comprehensive construction method conforming to the characteristics of the domain knowledge base by integrating multiple knowledge sources such as experience fact base, common knowledge base, and the like formed by expert experience:
1. knowledge reasoning technology is respectively used in different types of knowledge bases (such as a common knowledge base, an experience base, an reasoning base and the like), and the knowledge is applied by a method which depends on first-order logical reasoning. The general practice may include: performing structural analysis (text structuring based on NLP) on the problem; forming a query condition; inquiring knowledge in a knowledge base; after the related knowledge is obtained, a conclusion is obtained by applying first-order logical reasoning; checking the rationality of the conclusion in the knowledge base; and returning to a conclusion.
2. In addition to the separate reasoning within each knowledge base, there is also cross analysis and verification prior to each knowledge base, enhancing the comprehensiveness and accuracy of the analysis.
3. The conclusion can be checked by each knowledge base, and supplementary knowledge can be provided, so that a unified demonstration process and conclusion are formed.
The embodiment of the present invention further provides a device 11 for constructing a domain knowledge base, as shown in fig. 5, may include a processor 11 and a computer readable storage medium 12, where the computer readable storage medium 12 stores instructions, and when the instructions are executed by the processor 11, the method for constructing a domain knowledge base is implemented.
Those of ordinary skill in the art will appreciate that all or some of the steps, systems, functional modules/units in the apparatus, and methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. In a hardware implementation, the division between the functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be performed cooperatively by several physical components. Some or all of the components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as known to those skilled in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer. Furthermore, as is well known to those of ordinary skill in the art, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.
Claims (7)
1. The construction method of the domain knowledge base is characterized by comprising the following steps:
obtaining entity data according to a preset expert knowledge base, wherein the entity data comprises: concepts and attributes of entities and relationships between entities;
extracting text information from the entity data by using a preset extraction technology, and taking the extracted text information as body data;
and carrying out knowledge construction on the ontology data by adopting the established reasoning knowledge base, wherein the knowledge construction comprises the following steps:
acquiring an evolution time point from a time line in an evolution process of the domain ontology, wherein the method comprises the following steps: selecting a time point at which at least one of the preset dimension attributes of the body data changes compared with the previous time point;
wherein the preset dimension attribute comprises any one or more of the following: classification, grading and staging;
importing the ontology data of the evolution time point into the reasoning knowledge base as real-time data;
the reasoning knowledge base comprises a reasoning module;
the reasoning knowledge base identifies the problem characteristics according to the real-time data, and matches and classifies the identified problem characteristics with the fact experience knowledge base and/or the common sense knowledge base to obtain a decision scheme, which comprises the following steps:
the reasoning module performs context awareness, context analysis and context deduction according to the real-time data, identifies the problem features, and matches the identified problem features with a plurality of case features in the fact experience knowledge base and/or common sense knowledge base to obtain a solution; carrying out scheme classification according to the scene deduction result and the solution, and obtaining a matching scheme according to the scene perception result and the solution classification; acquiring an optimal scheme according to the matching scheme and a preset treatment scheme, and taking the optimal scheme as a final decision scheme;
evolving the ontology data according to one or more real-time data of the evolution time points and corresponding decision schemes acquired by the reasoning knowledge base, and acquiring a decision target of the domain ontology; constructing new knowledge according to the decision target;
and forming a new domain knowledge base according to the new knowledge obtained by knowledge construction.
2. The method for constructing a domain knowledge base according to claim 1, wherein the knowledge construction comprises: and performing mutual verification and mutual check between the ontology data, and indicating errors in the process of abstracting the ontology data to knowledge.
3. The method for constructing a domain knowledge base according to claim 1, further comprising: and in the knowledge construction process, the linkage of the new knowledge and the current knowledge base is realized through a knowledge interrupter.
4. The method for building a domain knowledge base according to claim 3, wherein said implementing the linkage of the new knowledge and the current knowledge base by the knowledge interrupter comprises:
detecting whether the knowledge which is the same as the new knowledge to be formed exists in the current knowledge base and whether the knowledge contradicted with the new knowledge to be formed exists or not when the new knowledge is formed according to the ontology data;
if the current knowledge base has the same knowledge as the new knowledge to be formed, interrupting the new knowledge abstract process through the knowledge interrupter;
if the knowledge contradicted with the new knowledge to be formed exists in the current knowledge base, the contradictory knowledge is recorded and sent to the research staff for research.
5. The method for constructing a domain knowledge base according to claim 1, wherein the reasoning module is further configured to complete labeling of the problem features through a feature dictionary;
the reasoning module completes the labeling of the problem features through the feature dictionary and comprises the following steps: isolating the front and rear problem features through the marking points, and carrying out knowledge representation on each marking point; the knowledge representation of the annotation point is described in terms of one or more of the following: an input element, a state element, and an output element;
the input element includes: an environmental input and a control input;
the state element includes: the nature and characteristics of the scene at a specific moment;
the output element includes: the impact of the state on the external environment and the loss of events due to the state change.
6. The method according to claim 5, wherein each feature question unit generated by labeling the question feature is classified by the feature dictionary into a unique code in the feature dictionary, unique identification ID numbers are assigned to the fact experience knowledge base and/or the common sense knowledge base for storage, and knowledge representation of the question feature is generated by summarizing the subject matter of the event,
the reasoning module is also used for determining the relation among the problem features according to the time sequence of the plurality of problem feature units generated after marking in the associated event and acquiring the whole evolution course of the ontology data; the relationship comprises a concurrency relationship and a precedence relationship.
7. A construction apparatus for a domain knowledge base, comprising a processor and a computer readable storage medium having instructions stored therein, wherein the construction method for a domain knowledge base according to any one of claims 1-6 is implemented when the instructions are executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010125394.9A CN111339299B (en) | 2020-02-27 | 2020-02-27 | Construction method and device of domain knowledge base |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010125394.9A CN111339299B (en) | 2020-02-27 | 2020-02-27 | Construction method and device of domain knowledge base |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111339299A CN111339299A (en) | 2020-06-26 |
| CN111339299B true CN111339299B (en) | 2023-06-02 |
Family
ID=71185675
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010125394.9A Active CN111339299B (en) | 2020-02-27 | 2020-02-27 | Construction method and device of domain knowledge base |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111339299B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112100156B (en) * | 2020-09-15 | 2024-02-20 | 北京百度网讯科技有限公司 | Method, device, medium and system for constructing knowledge base based on user behaviors |
| CN112347532A (en) * | 2020-11-06 | 2021-02-09 | 北京机械设备研究所 | Cable laying sequence determination method and device and computer equipment |
| CN112508249A (en) * | 2020-11-27 | 2021-03-16 | 清华大学 | Method and device for constructing emergency deduction graph structure and method and device for deducting emergency |
| CN112579782A (en) * | 2020-12-28 | 2021-03-30 | 上海明略人工智能(集团)有限公司 | Data processing method, knowledge management system, electronic device, and readable storage medium |
| CN112784064B (en) * | 2021-01-28 | 2023-05-12 | 电子科技大学 | Hierarchical super-dimensional knowledge graph construction method for social network |
| CN115269751B (en) * | 2022-05-10 | 2023-05-30 | 泰瑞数创科技(北京)股份有限公司 | Method for constructing geographic entity space-time knowledge graph ontology library |
| CN115033716B (en) * | 2022-08-10 | 2023-01-20 | 深圳市人马互动科技有限公司 | General self-learning system and self-learning method based on same |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103699663A (en) * | 2013-12-27 | 2014-04-02 | 中国科学院自动化研究所 | Hot event mining method based on large-scale knowledge base |
| WO2017040290A1 (en) * | 2015-08-28 | 2017-03-09 | Iqstrategix, Inc. | Knowledge acquisition, visualization and storage system |
| CN108196880A (en) * | 2017-12-11 | 2018-06-22 | 北京大学 | Software project knowledge mapping method for automatically constructing and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8407169B2 (en) * | 2008-06-26 | 2013-03-26 | Glenn Hofford | Expert system and method |
-
2020
- 2020-02-27 CN CN202010125394.9A patent/CN111339299B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103699663A (en) * | 2013-12-27 | 2014-04-02 | 中国科学院自动化研究所 | Hot event mining method based on large-scale knowledge base |
| WO2017040290A1 (en) * | 2015-08-28 | 2017-03-09 | Iqstrategix, Inc. | Knowledge acquisition, visualization and storage system |
| CN108196880A (en) * | 2017-12-11 | 2018-06-22 | 北京大学 | Software project knowledge mapping method for automatically constructing and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111339299A (en) | 2020-06-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111339299B (en) | Construction method and device of domain knowledge base | |
| CN110889556B (en) | Enterprise operation risk characteristic data information extraction method and extraction system | |
| CN109446343B (en) | Public safety knowledge graph construction method | |
| US10484413B2 (en) | System and a method for detecting anomalous activities in a blockchain network | |
| CN102982097B (en) | Domain for Knowledge based engineering data quality solution | |
| CN111159249A (en) | Target identification method, device and system based on knowledge graph and storage medium | |
| CN102930023A (en) | A data quality solution based on knowledge | |
| CN111538794B (en) | Data fusion method, device and equipment | |
| CN114710368B (en) | Security event detection method and device and computer readable storage medium | |
| Park et al. | Multi-entity bayesian networks learning in predictive situation awareness | |
| CN116562370A (en) | Emergency knowledge graph-based sudden public health event scenario deduction method | |
| CN115858513A (en) | Data governance method, data governance device, computer equipment and storage medium | |
| CN115034596A (en) | Risk conduction prediction method, device, equipment and medium | |
| CN114818353A (en) | Train control vehicle-mounted equipment fault prediction method based on fault characteristic relation map | |
| CN111125069A (en) | Data cleaning and fusing system | |
| US11875234B2 (en) | Systems and/or methods for machine-learning based data correction and completion in sparse datasets | |
| CN112818003A (en) | Execution risk estimation method and device for query task | |
| US20220247620A1 (en) | Identification of Clusters of Elements Causing Network Performance Degradation or Outage | |
| CN112559641B (en) | Pull chain table processing method and device, readable storage medium and electronic equipment | |
| CN111737319B (en) | User cluster prediction method, device, computer equipment and storage medium | |
| CN112905808A (en) | Knowledge graph construction method and device and electronic equipment | |
| CN112905612A (en) | Knowledge card construction method and device | |
| CN111967671A (en) | Cross-border active user identification method and device based on support vector data field description | |
| CN114546754A (en) | Automatic intelligent monitoring method and system and map data cloud platform | |
| CN112613072B (en) | Information management method, management system and management cloud platform based on archive big data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |