Method and device for extracting unstructured text information of ship equipment maintenance support

Abstract

The invention discloses a method and device for extracting unstructured text information of ship equipment maintenance support, and the method comprises the steps: constructing a ship equipment maintenance support field ontology model, and then carrying out the syntactic analysis of the unstructured text information of the ship equipment maintenance support through a natural language processing tool; extracting concepts (entities) in the equipment maintenance support field through a syntax dependency relationship, identifying a concept relationship, summarizing to form a rule, and storing the rule in a database; and finally, iteratively operating the unstructured text information by the method to form the unstructured text information equipment maintenance support concepts, entities and a relation graph thereof, thereby providing data and knowledge support for intelligent decision-making application.

Description

Technical Field

[0001] The present invention relates to the technical field of natural language processing, and in particular to a method and device for extracting unstructured text information for ship equipment maintenance and support. Background Art

[0002] In addition to structured data, there are also a large number of unstructured data in the form of text descriptions in ship equipment maintenance and support activities, such as maintenance manuals, technical manuals, equipment principles, etc. These unstructured data in text form contain a large amount of equipment maintenance information, as well as maintenance and support experience and knowledge that exist in an invisible way.

[0003] In the existing technology, the extraction of unstructured information mainly involves manual analysis of unstructured text, followed by structured modeling, forming relational tables, and using computer technology to generate databases. The semantic relationships therein are not considered, so it is impossible to effectively obtain useful information in ship equipment maintenance and support activities. Summary of the invention

[0004] The present invention provides a method and device for extracting unstructured information of ship equipment maintenance and support, which are used to solve or at least partially solve the technical problem in the prior art that useful information in ship equipment maintenance and support activities cannot be effectively obtained.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] The first aspect discloses a method for extracting unstructured text information of ship equipment maintenance and support, comprising:

[0007] Establish the ontology of ship equipment maintenance and support, where the ontology includes the concepts of ship equipment maintenance and support and the semantic relations between different concepts;

[0008] Use Chinese natural language processing tools to perform word segmentation and syntactic analysis on unstructured texts in ship equipment maintenance and support activities to obtain syntactic relations in unstructured texts;

[0009] Analyze the syntactic relationship in the obtained unstructured text to obtain the matching rules between the syntactic relationship and the semantic relationship;

[0010] Based on the ontology of ship equipment maintenance and support domain and the matching rules from syntactic relations to semantic relations, the unstructured text information of ship equipment maintenance and support to be processed is extracted.

[0011] In one implementation, a Chinese natural language processing tool is used to perform syntactic analysis on unstructured text in ship equipment maintenance and support activities to obtain syntactic relations in the unstructured text, including:

[0012] Use Chinese natural language processing tools to perform word segmentation and syntactic analysis on unstructured texts in ship equipment maintenance and support activities to obtain the part of speech of each word in the sentence, its components in the sentence, and the dependency relationship between words;

[0013] According to the part of speech of each word in the sentence, its component in the sentence and the dependency relationship between words, syntactic dependency, adjacent dependency and inter-segment dependency are obtained as syntactic relations.

[0014] In one implementation, the syntactic relationship in the obtained unstructured text is analyzed to obtain a matching rule between the syntactic relationship and the semantic relationship, including:

[0015] Analyze the syntactic relationship in the unstructured text to obtain the parts of speech, the ontology classes to which the two concepts in the syntactic dependency relationship belong, and the components in the sentence;

[0016] The ontological relationship between the two concepts is determined based on the parts of speech, the ontological categories they belong to, and the components in the sentence, and the transformation rules from syntactic dependency to semantic relationship are obtained.

[0017] In one embodiment, the matching rule between the syntactic dependency and the semantic relationship includes the dependency direction between the concepts in the ontology and the concepts in the syntactic dependency. If the dependency direction is the same as the actual relationship direction, the dependency direction is forward, otherwise the dependency direction is backward.

[0018] In one embodiment, the method further comprises:

[0019] Determine whether there is a semantic relationship between two adjacent concepts. If so, analyze the adjacent dependency relationship in the selected technical manual text sample and obtain the transformation rules from adjacent dependency to semantic relationship.

[0020] In one embodiment, the transformation rules from adjacent dependencies to semantic relationships include two attributes, namely, whether to cross words and the crossed keywords, wherein the whether to cross words attribute indicates whether there are structural auxiliary words or conjunctions between two adjacent concepts, and the crossed keywords attribute indicates which keyword is crossed when the whether to cross words attribute is true.

[0021] In one implementation, the unstructured text information of ship equipment maintenance and support to be processed is extracted based on the matching rules between the syntactic relationship and the semantic relationship in the ship equipment maintenance and support domain ontology, including:

[0022] Based on the ontology of ship equipment maintenance and support, concepts are extracted from the unstructured text of ship equipment maintenance and support to be processed;

[0023] The semantic relationship between the extracted concepts is obtained based on the matching rules between syntactic relationship and semantic relationship;

[0024] The extracted concepts and semantic relations are displayed in the form of a knowledge graph.

[0025] Based on the same inventive concept, the second aspect of the present invention provides a device for extracting unstructured text information of ship equipment maintenance and support, comprising:

[0026] An ontology building module is used to establish an ontology for the field of ship equipment maintenance and support, wherein the ontology for the field of ship equipment maintenance and support includes concepts in ship equipment maintenance and support and semantic relationships between different concepts;

[0027] The syntax analysis module is used to perform word segmentation and syntax analysis on the unstructured text in the ship equipment maintenance and support activities using Chinese natural language processing tools to obtain the syntactic relations in the unstructured text;

[0028] A matching rule building module is used to analyze the syntactic relationship in the obtained unstructured text and obtain the matching rules between the syntactic relationship and the semantic relationship;

[0029] The information extraction module is used to extract the unstructured text information of ship equipment maintenance and support to be processed based on the ontology of the ship equipment maintenance and support field and the matching rules between syntactic relations and semantic relations.

[0030] Based on the same inventive concept, the third aspect of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method for extracting unstructured information of ship equipment maintenance and support described in the first aspect is implemented.

[0031] Based on the same inventive concept, the fourth aspect of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for extracting unstructured information of ship equipment maintenance and support described in the first aspect is implemented.

[0032] Compared with the prior art, the advantages and beneficial technical effects of the present invention are as follows:

[0033] The extraction method for unstructured information in ship equipment maintenance support disclosed by the present invention first establishes an ontology in the field of ship equipment maintenance support, and uses Chinese natural language processing tools to perform word segmentation and syntactic analysis on the unstructured text in ship equipment maintenance support activities to obtain the syntactic relationships in the unstructured text; then analyzes the syntactic relationships in the obtained unstructured text to obtain the matching rules between syntactic relationships and semantic relationships; finally, extracts the unstructured text information of ship equipment maintenance support to be processed based on the ontology in the field of ship equipment maintenance support and the matching rules from syntactic relationships to semantic relationships. Through the method of the present invention, the unstructured information in ship equipment maintenance support can be extracted from the perspectives of semantics and knowledge system, and the useful information in the unstructured text of ship equipment maintenance support can be effectively extracted, thereby providing data and knowledge support for intelligent decision-making applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0035] Figure 1 It is the basic technical roadmap for unstructured data extraction in the embodiments of the present invention;

[0036] Figure 2 It is a schematic diagram of the ontology in the field of ship equipment maintenance support constructed in the embodiments of the present invention;

[0037] Figure 3 It is a schematic diagram of the result of syntactic analysis using the Chinese natural language processing tool (FNLP) in the embodiments of the present invention;

[0038] Figure 4 For Figure 3 a visualization schematic diagram of the analysis result;

[0039] Figure 5 It is a schematic diagram of the result of syntactic analysis containing the "de-construction" in the embodiments of the present invention;

[0040] Figure 6 It is a process diagram for constructing the dependency rule library in the embodiments of the present invention;

[0041] Figure 7 It is a schematic diagram of the syntactic dependency transformation rule library in the embodiments of the present invention;

[0042] Figure 8 It is a schematic diagram of the wrong relationship direction in the embodiments of the present invention;

[0043] Fig. 9 is a schematic diagram of an adjacent dependency rule base in an embodiment of the present invention;

[0044] Fig.10 Schematic diagram of the diesel engine fuel pump information extraction result in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In addition to structured data, there is a large amount of unstructured data in the form of text descriptions in ship equipment maintenance and support activities, such as maintenance manuals, technical manuals, equipment principles, etc. These unstructured data in text form contain a large amount of equipment maintenance information, as well as maintenance and support experience and knowledge that exist in an invisible way. How to extract this data and systematically form equipment maintenance and support knowledge to facilitate reuse, sharing and integration is an important part of the informatization and intelligentization of ship equipment maintenance and support.

[0046] Based on this, a method for extracting and acquiring knowledge from unstructured information of ship equipment maintenance and support is proposed. This method constructs an ontology model of the ship equipment maintenance and support field, and proposes to use natural language processing tools to perform syntactic analysis on the unstructured text information of ship equipment maintenance and support. The concepts (entities) in the equipment maintenance and support field are extracted through syntactic dependencies, and the concept relationships are identified. The rules are summarized and stored in the database. Finally, the above method is iteratively run on the unstructured text information to form a graph of unstructured text information equipment maintenance and support concepts, entities and their relationships as the information extraction result, thereby providing data and knowledge support for intelligent decision-making applications.

[0047] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] Embodiment 1

[0049] The present invention discloses a method for extracting unstructured text information of ship equipment maintenance and support, comprising:

[0050] Establish the ontology of ship equipment maintenance and support, where the ontology includes the concepts of ship equipment maintenance and support and the semantic relations between different concepts;

[0051] Use Chinese natural language processing tools to perform word segmentation and syntactic analysis on unstructured texts in ship equipment maintenance and support activities to obtain syntactic relations in unstructured texts;

[0052] Analyze the syntactic relationship in the obtained unstructured text to obtain the matching rules between the syntactic relationship and the semantic relationship;

[0053] Based on the ontology of ship equipment maintenance and support domain and the matching rules from syntactic relations to semantic relations, the unstructured text information of ship equipment maintenance and support to be processed is extracted.

[0054] This method uses the following technical routes to achieve the extraction of unstructured data and knowledge acquisition, such as Figure 1 shown.

[0055] The first step is to establish the domain ontology of ship equipment maintenance and support (including concept extraction and relationship classification). The domain ontology is the premise and key to extracting unstructured text information (such as equipment technical manuals) and acquiring knowledge for ship equipment maintenance and support. The domain ontology of ship equipment maintenance and support standardizes and formalizes the classification of core concepts (i.e., ontology) of equipment maintenance and support in unstructured information of equipment maintenance and support, as well as the types of semantic relationships between different concepts (i.e., ontology), to form a domain ontology set; then, computer software is used to construct the core vocabulary library of the domain ontology of ship equipment maintenance and support based on the domain ontology set, to improve the accuracy of text concept extraction of unstructured information of equipment maintenance and support (such as equipment technical manuals), and to provide sample support for the extraction of text information of equipment technical manuals and knowledge acquisition knowledge representation.

[0056] The second step is to use Chinese natural language processing (FNLP) tools to perform syntactic analysis on the unstructured text (such as equipment technical manual) information in equipment maintenance and support activities to obtain the syntactic relations (relation extraction) in the unstructured text (such as equipment technical manual) information text. Combined with the ontology of ship equipment maintenance and support and its semantic relations,

[0057] The third step is to further explore the syntactic relationship of unstructured text, find the transformation rules from syntactic relationship to semantic relationship, summarize and classify the transformation rules from syntactic relationship to semantic relationship, and establish a rule base. In this way, the concept information is brought into the established rule base, and the ontological relationship (semantic relationship) between concepts can be determined through rule matching. As the information text of unstructured text (such as equipment technical manual) is continuously analyzed, new rules are added to the rule base.

[0058] The fourth step is to extract information from the unstructured text of ship equipment maintenance and support based on the ontology of ship equipment maintenance and support, and the matching rules between syntactic relations and semantic relations. The words after word segmentation are matched with the concepts in the ontology concept library to obtain concepts, and the dependency relations after syntactic analysis are matched with the rules in the relationship transformation rule library (including the transformation rules from syntactic dependency relations to semantic relations, and the transformation rules from adjacent dependency relations to semantic relations) to obtain the corresponding semantic relations. The results of information extraction include the ontology of ship equipment maintenance and support (concepts, semantic relations), and transformation rules, which are displayed in the form of concept maps or knowledge maps using drawing tools.

[0059] The specific implementation process of each part is introduced in detail below.

[0060] 1. Establishing an ontology model for ship equipment maintenance and support

[0061] Based on the technical route mentioned above, a large number of unstructured texts of ship equipment maintenance and support are comprehensively analyzed to establish an ontology model for the field of ship equipment maintenance and support, such as Figure 2 The oval box in the figure is the ontology concept, and the rectangular box is the ontology relationship. The ontology meaning is shown in Table 1.

[0062] Table 1 Meaning of the ontology model for unstructured information extraction

[0063]

[0064]

[0065] The ontology relationship in the unstructured information extraction ontology model is as follows:

[0066] (1)hasSubsystem

[0067] Subsystem relationships indicate that equipment such as parts and components have a lower level of equipment class. For example, in the phrase "pressure of fuel in the inner cavity of the oil pump", the subsystem of "oil pump" is "inner cavity", and the subsystem of "inner cavity" is "fuel". This relationship can be used to represent the structural principle knowledge of equipment in the technical manual.

[0068] (2) hasFunction

[0069] Functional relations indicate that the parts or attributes of equipment have certain functions. For example, in the sentence "the oil pump delivers fuel.", "the oil pump" has the function of "delivering". This relationship can be used to represent the functional principle knowledge of equipment in the technical manual.

[0070] (3)Target

[0071] Object relations represent the object of a function. For example, in the sentence "The oil pump delivers fuel.", the object of the function "delivering" is "fuel". This relation often appears together with the "hasFunction" relation. This relation can also be used to represent the functional principle knowledge of the equipment in the technical manual.

[0072] (4)hasPart

[0073] Partial relationship refers to the part of a component. For example, in the sentence "The lower end of the support rod drives the control sleeve to move to the right.", the "lower end" refers to the part of the "support rod". This relationship is a subsidiary relationship and plays a limiting role on certain directional words.

[0074] (5)hasAttribute

[0075] Attribute relations represent the attributes of a component. For example, in the phrase "pressure of the fuel in the oil pump cavity", the attribute of "fuel" is "pressure". This relation can be used to understand the working status or technical requirements of the entire equipment or its components.

[0076] (6) hasSymptom

[0077] Phenomenon relations represent the phenomenon of a certain attribute. For example, in the sentence "The fuel pressure increases.", the phenomenon of the attribute "pressure" is "increase". When equipment fails, the phenomenon of the equipment is abnormal first, and this can be used as a basis to find the problematic parts.

[0078] (7)hasValue

[0079] A value relationship indicates the value of a certain attribute.

[0080] (8)Downstream

[0081] A sequential relationship indicates a transfer from one action or phenomenon to another. This relationship usually occurs between two predicates in a short sentence or between predicates in different short sentences in a long sentence. For example, in the sentence "The lower end of the support rod drives the control sleeve to move to the right.", the relationship between "drive" and "move" is a sequential relationship.

[0082] (9)Condition

[0083] Conditional relationships indicate that a certain action or phenomenon is premised on a certain action or phenomenon. For example, in the sentence "When the fuel pressure acting on the slide valve exceeds the spring force, the spring is compressed and the slide valve moves.", the function of "moving" will only appear when the condition of "exceeding" is met.

[0084] 2. Using Chinese natural language processing tools for word segmentation and syntactic analysis

[0085] According to the technical route, after establishing the ontology in the field of ship equipment maintenance support, in order to enable the computer to automatically extract text information to obtain knowledge, it is essential to use natural language processing tools (FNLP) to perform syntactic analysis on the text information of ship equipment maintenance support.

[0086] Since the semantic relationships between concepts in unstructured text are implicit and cannot be directly obtained, it is necessary to perform syntactic analysis on the unstructured text to obtain syntactic dependency relationships, and then transform the syntactic relationships into semantic relationships.

[0087] This embodiment uses Chinese natural language processing (FNLP) for syntactic analysis. Taking "The fuel pump transports fuel." as an example, the syntactic analysis results are as Figure 3 shown. In this figure, each square bracket represents the information obtained from the analysis of a word. From left to right, they are the word, part of speech, position of the dependent word, and the component of the word in the sentence. Taking the first line "fuel pump" as an example, its part of speech is "noun", the position of the dependent word is 1, that is, the second word "transports", and the last element "noun" indicates that "fuel pump" acts as the subject in the sentence. When performing syntactic analysis, each sentence will determine a central word (Root) after syntactic analysis. This word is the core of the whole sentence, usually a verb or an adjectival verb. At this time, the words in the whole sentence will directly or indirectly depend on this central word (Root), and the central word (Root) does not depend on any word.

[0088] To increase the readability of the syntactic analysis results, the syntactic analysis results are displayed in the form of a graph through a program. Figure 3 The visualization graph of the results in Figure 4 is as shown. Figure 4 It can clearly show the results of syntactic analysis, and the word segmentation results, part-of-speech tagging, components in the sentence, core words, as well as the dependency relationships and directions are all clear at a glance.

[0089] When performing syntactic analysis, if there is a "de" (的) character in the sentence, the situation where the component in the sentence is a "de-structure" will occur, indicating that this word first depends on the "de" character, and then the "de" character depends on the next word. As Figure 5 shown, in "The rotation speed of the oil pump changes.", the component of the concept "oil pump" in the sentence is a "de-structure", and it depends on the concept "rotation speed" through "de". For this problem, in this embodiment, when looking for dependency relationships, the "de" character should be selectively skipped, and a dependency relationship should be established between the word with the component of "de-structure" and the word that the "de" character depends on. Only in this way is the correct dependency relationship, that is, Figure 5 establish the dependency relationship between "oil pump" and "rotation speed" in

[0090] By performing syntactic analysis on the text, we can obtain information such as the part of speech of each word in the sentence, its components in the sentence, and the dependency relationship between words. The ultimate goal of this method is to obtain the semantic relationship between words and reveal the structure, type, and mutual relationship of the ship equipment maintenance and support knowledge described in the text. The following is a comparative analysis of the dependency relationship obtained by syntactic analysis and the semantic relationship implied in the unstructured text, trying to find the transformation rules or rules from syntactic dependency relationship to semantic relationship.

[0091] 3. Construction of matching rules between syntactic and semantic relations

[0092] This embodiment proposes a method for constructing transformation rules from syntactic dependencies to semantic relationships.

[0093] There are no ready-made rules to refer to in the prior art, and it is necessary to extract rules and establish a rule base through artificial learning methods. The process of manually extracting relationship transformation rules is as follows: Figure 6 First, we use natural language processing tools to perform word segmentation and syntactic analysis on the technical manual text to obtain the syntactic dependency relationship in the text, and then find the transformation rules from the syntactic relationship to the semantic relationship specified by the technical manual domain ontology, and finally add the rules to the rule base.

[0094] When performing syntactic analysis, the two concepts that establish syntactic dependency generally have a certain semantic relationship, which is the main research object for extracting semantic relationships in unstructured information texts (such as technical manuals). According to the analysis, it is necessary to use the part of speech, ontology class, and sentence components of the concept as judgment conditions, and then use certain rules to determine the ontology relationship between the two concepts. Here, through the analysis of syntactic dependency in a large number of samples extracted from unstructured information texts for ship equipment maintenance and support, the transformation rules from syntactic dependency to semantic relationship are summarized as shown in Table 2.

[0095] Table 2 Transformation rules from syntactic dependency to semantic relationship

[0096]

[0097] The transformation rules from syntactic dependencies to semantic relationships in Table 2 are established as follows Figure 7 In the XML document shown in the figure, each element is a rule, and the attributes and meanings of each element are shown in Table 3.

[0098] Table 3 Element attributes and meanings of the syntactic dependency transformation rule base

[0099]

[0100] The rule contains the dependency direction (RelyDirection) of two concepts. If the dependency direction is the same as the actual relationship direction, then RelyDirection is forward; otherwise, it is backward. When analyzing using natural language processing (FNLP), both the subject and object in a sentence depend on the Root in the sentence. According to the dependency rules, the relationship between the Root and the object is Target. Without direction judgment, there will be Figure 8 the error situation shown. Therefore, in the dependency relationship rules, it is necessary to judge the dependency direction.

[0101] During syntactic analysis, syntactic dependency relationships may be incorrect, missing, or incorrect syntactic dependency relationships may be established. It is found that there are often syntactic dependency relationships between two adjacent concepts, but syntactic analysis often makes mistakes. To correct the errors in syntactic dependencies, adjacent dependency analysis and rule matching need to be carried out.

[0102] Adjacent dependency analysis is to establish a dependency relationship between two adjacent concepts, and then judge whether there is a semantic relationship between the two adjacent concepts by matching adjacent dependency transformation rules. Similar to syntactic dependency rules, when determining the adjacent relationship, it is necessary to handle the case of the word "的". In addition, there may be other structural particles or conjunctions, such as "就", "和", etc. When establishing adjacent rules, it is necessary to judge whether such words appear. If they appear, the adjacent dependency relationship is directly established across the word. However, the words crossed will affect the semantic relationship between concepts. For example, when establishing an adjacent dependency relationship across the word "和", the semantic relationship between concepts may be a coordinate relationship "IsCoordinateWith". This type of situation needs to be added to the adjacent dependency transformation rule library. Through the analysis of adjacent dependency relationships in the selected technical manual text samples, the transformation rules from adjacent dependencies to semantic relationships obtained are shown in Table 4.

[0103] Table 4 Transformation Rules from Adjacent Dependency Relationships to Semantic Relationships

[0104]

[0105] Establish the transformation rules from adjacent dependency relationships to semantic relationships in Table 4 into an XML document as Fig. 9 shown, where each element is a rule.

[0106] Compared with syntactic dependency relationship rules, each element in the adjacent relationship rules adds the following two attributes:

[0107] (1) SpanOrNot: Whether to cross a character. There may be structural particles or conjunctions such as "的" and "和" between two concepts. Therefore, when determining adjacent rules, it is necessary to judge whether there are structural particles or conjunctions. If they appear, it is true; otherwise, it is false.

[0108] (2) SpanKey: The cross-keyword. This attribute is only given when the above attribute is true, otherwise it is empty. Currently, the main keywords include words such as "的", "和", "至", "于", "就", etc.

[0109] During adjacent dependency analysis, the attributes of adjacent concepts are matched with the rules in the established adjacent dependency rule library. If the attributes of two concepts match a certain rule, then the relationship in that rule is the relationship existing between the two concepts, and the direction therein is the direction of the relationship, that is, the direction of the directed connection line in the concept graph.

[0110] IV. Information Extraction

[0111] The following takes the unstructured text information of the working principle of the internal pressure control valve in the fuel injection pump of a diesel engine as an example to elaborate an example of structured text information extraction and knowledge acquisition for ship equipment maintenance support. The text information is as follows:

[0112] "The spool valve is under the combined action of the fuel pressure and the spring force in the inner cavity of the fuel pump. When the rotational speed of the fuel pump is different, the amount of fuel delivered by the fuel transfer pump is different, and the fuel pressure in the inner cavity of the fuel pump also changes accordingly. When the fuel pressure acting on the spool valve exceeds the spring force, the spring is compressed and the spool valve moves. When the spool valve moves to open the oil return hole on the internal pressure control body, the fuel in the inner cavity of the fuel pump flows back to the suction side of the fuel transfer pump through the oil return hole, and the fuel pressure in the inner cavity of the fuel pump drops to balance with the spring force. Therefore, the magnitude of the spring force determines the magnitude of the fuel pressure in the inner cavity of the fuel pump."

[0113] The extraction result is as Fig.10 shown.

[0114] Through the method of this embodiment, the unstructured text of ship equipment maintenance support is fused into a more intuitive graphical form. After being fused into this intuitive form, the relational database and graph database can be used to mine the above maintenance support data to form their respective main databases, and various algorithms are used to extract and generate various maintenance support decision index data. This method is used for iteration again, so as to provide data and knowledge support for intelligent decision-making applications, such as maintenance operation guidance for various equipment failures, intelligent inspection, and technical status assessment during the equipment management and use process.

[0115] Embodiment 2

[0116] Based on the same inventive concept, this embodiment discloses an extraction device for unstructured text information of ship equipment maintenance support, including:

[0117] An ontology building module is used to establish an ontology for the field of ship equipment maintenance and support, wherein the ontology for the field of ship equipment maintenance and support includes concepts in ship equipment maintenance and support and semantic relationships between different concepts;

[0118] The syntax analysis module is used to perform word segmentation and syntax analysis on the unstructured text in the ship equipment maintenance and support activities using Chinese natural language processing tools to obtain the syntactic relations in the unstructured text;

[0119] A matching rule building module is used to analyze the syntactic relationship in the obtained unstructured text and obtain the matching rules between the syntactic relationship and the semantic relationship;

[0120] The information extraction module is used to extract the unstructured text information of ship equipment maintenance and support to be processed based on the ontology of the ship equipment maintenance and support field and the matching rules between syntactic relations and semantic relations.

[0121] Since the device introduced in the second embodiment of the present invention is a device used to implement the method for extracting unstructured information of ship equipment maintenance and support in the first embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, the person skilled in the art can understand the specific structure and deformation of the device, so it is not repeated here. All devices used in the method in the first embodiment of the present invention belong to the scope of protection of the present invention.

[0122] Embodiment 3

[0123] Based on the same inventive concept, the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed, the method described in the first embodiment is implemented.

[0124] Since the computer-readable storage medium introduced in the third embodiment of the present invention is the computer-readable storage medium used to implement the method for extracting unstructured information of ship equipment maintenance and support in the first embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, those skilled in the art can understand the specific structure and deformation of the computer-readable storage medium, so it is not repeated here. All computer-readable storage media used in the method of the first embodiment of the present invention belong to the scope of protection of the present invention.

[0125] Embodiment 4

[0126] Based on the same inventive concept, the present application also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the method in Embodiment 1 is implemented when the processor executes the above program.

[0127] Since the computer device introduced in the fourth embodiment of the present invention is a computer device used to implement the method for extracting unstructured information of ship equipment maintenance and support in the first embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, the person skilled in the art can understand the specific structure and deformation of the computer device, so it is not repeated here. All computer devices used in the method in the first embodiment of the present invention belong to the scope of protection of the present invention.

[0128] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0129] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0130] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic creative concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the embodiments of the present invention without departing from the spirit and scope of the embodiments of the present invention. Thus, if these modifications and variations of the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for extracting unstructured text information of ship equipment maintenance and support, characterized in that: include: Establish the ontology of ship equipment maintenance and support, where the ontology includes the concepts of ship equipment maintenance and support and the semantic relations between different concepts; Use Chinese natural language processing tools to perform word segmentation and syntactic analysis on unstructured texts in ship equipment maintenance and support activities to obtain syntactic relations in unstructured texts; Analyze the syntactic relationship in the obtained unstructured text to obtain the matching rules between the syntactic relationship and the semantic relationship; Based on the ontology of ship equipment maintenance and support domain and the matching rules from syntactic relations to semantic relations, the unstructured text information of ship equipment maintenance and support to be processed is extracted.

2. The method for extracting unstructured text information of ship equipment maintenance and support according to claim 1, characterized in that: The Chinese natural language processing tools are used to perform syntactic analysis on the unstructured text in the ship equipment maintenance and support activities, and the syntactic relations in the unstructured text are obtained, including: Use Chinese natural language processing tools to perform word segmentation and syntactic analysis on unstructured texts in ship equipment maintenance and support activities to obtain the part of speech of each word in the sentence, its components in the sentence, and the dependency relationship between words; According to the part of speech of each word in the sentence, its component in the sentence and the dependency relationship between words, syntactic dependency, adjacent dependency and inter-segment dependency are obtained as syntactic relations.

3. The method for extracting unstructured text information of ship equipment maintenance and support according to claim 2, characterized in that: The syntactic relations in the obtained unstructured text are analyzed to obtain matching rules between syntactic relations and semantic relations, including: Analyze the syntactic relationship in the unstructured text to obtain the parts of speech, the ontology classes to which the two concepts in the syntactic dependency relationship belong, and the components in the sentence; The ontological relationship between the two concepts is determined based on the parts of speech, the ontological categories they belong to, and the components in the sentence, and the transformation rules from syntactic dependency to semantic relationship are obtained.

4. The method for extracting unstructured text information of ship equipment maintenance and support according to claim 1, characterized in that: The matching rules between syntactic dependencies and semantic relations include the dependency direction between concepts in the ontology and concepts in the syntactic dependencies. If the dependency direction is the same as the actual relationship direction, the dependency direction is forward, otherwise the dependency direction is backward.

5. The method for extracting unstructured text information of ship equipment maintenance and support according to claim 3, characterized in that: The method further comprises: Determine whether there is a semantic relationship between two adjacent concepts. If so, analyze the adjacent dependency relationship in the selected technical manual text sample and obtain the transformation rules from adjacent dependency to semantic relationship.

6. The method for extracting unstructured text information of ship equipment maintenance and support according to claim 5, characterized in that: The transformation rules from adjacent dependencies to semantic relations include two attributes, namely whether to cross words and the crossed keywords. The whether to cross words attribute indicates whether there are structural auxiliary words or conjunctions between two adjacent concepts, and the crossed keywords attribute indicates which keyword is crossed when the whether to cross words attribute is true.

7. The method for extracting unstructured text information of ship equipment maintenance and support according to claim 1, characterized in that: Based on the ontology of ship equipment maintenance and support, the matching rules between syntactic relations and semantic relations are used to extract the unstructured text information of ship equipment maintenance and support to be processed, including: Based on the ontology of ship equipment maintenance and support, concepts are extracted from the unstructured text of ship equipment maintenance and support to be processed; The semantic relationship between the extracted concepts is obtained based on the matching rules between syntactic relationship and semantic relationship; The extracted concepts and semantic relationships are displayed in the form of a knowledge graph.

8. A device for extracting unstructured text information of ship equipment maintenance and support, characterized in that: include: An ontology building module is used to establish an ontology for the field of ship equipment maintenance and support, wherein the ontology for the field of ship equipment maintenance and support includes concepts in ship equipment maintenance and support and semantic relationships between different concepts; The syntax analysis module is used to perform word segmentation and syntax analysis on the unstructured text in the ship equipment maintenance and support activities using Chinese natural language processing tools to obtain the syntactic relations in the unstructured text; A matching rule building module is used to analyze the syntactic relationship in the obtained unstructured text and obtain the matching rules between the syntactic relationship and the semantic relationship; The information extraction module is used to extract the unstructured text information of ship equipment maintenance and support to be processed based on the ontology of the ship equipment maintenance and support field and the matching rules between syntactic relations and semantic relations.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, it implements the method for extracting unstructured text information of ship equipment maintenance and support as described in any one of claims 1 to 7.

10. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method for extracting unstructured text information of ship equipment maintenance and support as described in any one of claims 1 to 7 is implemented.

Images