A fine-grained information extraction method and system based on part-of-speech tagging

Abstract

The present invention provides a fine-grained information extraction method and system based on part-of-speech tagging, the method comprising: encoding and information extraction strategies of pre-stored typical example sentences; performing phrase-level word segmentation on the sentence to be analyzed and tagging the phrases with parts of speech; merging and hiding adjacent phrases according to the encoding strategy; replacing nouns, verbs, and verb phrases in the sentence with parts of speech based on the parts of speech and sentence structure marked by S to form an encoding of the sentence to be analyzed; matching the encoding of the sentence to be analyzed with the encoding of pre-stored typical example sentences; if there is a match with the pre-stored typical example sentences, extracting information from the sentence to be analyzed according to the extraction strategy of the matched typical example sentences; if there is no match, extracting information from the sentence to be analyzed according to the information extraction strategy; and storing the encoding of the sentence to be analyzed and the information extraction strategy. The information extraction process of the present invention can quickly adapt to new information extraction needs.

Description

Technical Field

[0001] The present invention relates to the fields of computer data processing and data mining, and in particular to a method for extracting fine-grained information from text based on part-of-speech tagging using an NLP platform. Background Art

[0002] The use and operation of equipment and supported objects involved in shipboard support operations are accompanied by corresponding operating instructions, providing guidance on implementation and operational safety. These instructions contain a wealth of support operations knowledge, but this knowledge is often distributed across different equipment and supported objects, covering different chapters and sections. Manual retrieval is difficult and prone to omissions.

[0003] Knowledge graph technology can integrate isolated data and data associations into a network-structured knowledge system. This network-like knowledge structure effectively improves the effectiveness and completeness of information retrieval results. The first step in building a knowledge graph is to extract structured data from unstructured or semi-structured text. The technology used to accomplish this is known as information extraction technology.

[0004] Existing information extraction methods mainly include rule-based methods and deep learning-based methods. Deep learning-based methods have problems such as requiring a large amount of corpus annotation during training and long training time.

[0005] Standardized text uses unstructured natural language but possesses inherent structured characteristics. Rule-based methods can be used to extract information from such text. However, in actual text processing, factors such as differences in writing conventions, the inherent expressive characteristics of Chinese, and the technical limitations of NLP analysis tools can lead to inaccurate and incomplete decomposition and identification of entities and relationships when extracting information based solely on rules. Summary of the Invention

[0006] The technical problem to be solved by the present invention is the defects existing in the extraction method of the prior art. In order to solve the above problem, a fine-grained information extraction method and system based on part-of-speech tagging are provided.

[0007] The object of the present invention is achieved in the following manner:

[0008] A fine-grained information extraction method based on part-of-speech tagging, the method comprising the following steps:

[0009] S1: Obtain training sentences, use NLP tools to segment the sentences at the phrase level, and perform part-of-speech tagging on the decomposed phrases;

[0010] Adjacent phrases are merged and hidden according to the encoding strategy to form new part-of-speech tags;

[0011] Based on the new part-of-speech tagging and sentence structure, extract the specific part-of-speech sequence, keywords, and specific punctuation marks corresponding to the phrases in the sentence as the sentence code; the specific parts of speech include nouns, verbs, and verb phrases; the specific punctuation marks are symbols that indicate the parallel relationship between words;

[0012] Extract information from the current sentence by combining the information extraction strategy of the phrase's corresponding part of speech and its preceding and following related parts of speech, characteristic punctuation marks, or characteristic words, and save the sentence encoding and corresponding information extraction strategy of the current sentence as the encoding and information extraction strategy of the pre-stored typical example sentence;

[0013] S2: Segment the sentences to be analyzed at the phrase level and tag the phrases with parts of speech; merge and hide adjacent phrases according to the encoding strategy;

[0014] S3: Based on the parts of speech and sentence structure marked in S2, replace the nouns, verbs, and verb phrases in the sentence with parts of speech to form the code of the sentence to be analyzed;

[0015] S4: matching the encoding of the sentence to be analyzed with the encoding of pre-stored typical example sentences;

[0016] S5: If it matches the pre-stored typical example sentence, information extraction is performed on the sentence to be analyzed according to the extraction strategy of the matched typical example sentence;

[0017] S6: If there is no match, information is extracted from the sentence to be analyzed according to the information extraction strategy in S1; and the encoding of the sentence to be analyzed and the information extraction strategy are stored.

[0018] The information extraction strategy includes:

[0019] The "subject + predicate + object" sentence structure of NN1 VV NN2 can extract triples: entity 1 represented by NN1 - action relationship embodied by VV - entity 2 represented by NN2, entities and relationships;

[0020] The attributive phrase of NN1 DEG NN2 can extract triples: entity 2 represented by NN2 - belonging relationship - entity 1 represented by NN1;

[0021] The attributive phrase of AD DEG NN can extract the attribute represented by AD from the entity represented by NN;

[0022] The "subject + predicate + parallel object" sentence structure of NN1 VV NN2...NNn can extract triples: entity 1 represented by NN1 - action relationship embodied by VV - entity 2 represented by NN2, ..., entity 1 represented by NN1 - action relationship embodied by VV - entity n represented by NNn;

[0023] "Subject + Predicate + Object" or "Predicate + Object", Sentence ① + "before" | "after" + "Subject + Predicate + Object" or "Predicate + Object", Sentence ②, the process triple can be extracted: Operation Process ① represented by Sentence ① — Operation Process Sequence Relationship — Operation Process ② represented by Sentence ②.

[0024] The encoding strategy includes:

[0025] A. For continuously appearing nouns, the词性 is marked as noun NN, noun phrase NP, personal name NR, proper noun NT, and they are merged to generate a new noun NN;

[0026] B. Phrases with the词性 marked as adjective JJ + noun NN are merged, and the词性 of the merged phrase is noun NN;

[0027] C. For continuously appearing verbs, the词性 is marked as VV, VP and they are merged to generate a new verb;

[0028] D. Phrases with the词性 marked as adverb AD + verb VV are merged, and the词性 of the merged phrase is verb VV; <​​​​​​​​​​​​​​​​​​​​​​​​​In S4, the encoding of the sentence to be analyzed is matched with the encoding of the pre-stored typical example sentences using a method based on sentence structure similarity. Specifically, the difference between sentences is generally represented by the edit distance. In the actual sentence pattern matching process, it is first checked whether the number of verbs contained in the sentence to be matched is the same as that in the sentence pattern. The sentence to be matched is only compared with the encoding of the sentence with the same number of verbs as that to find the sentence pattern with the smallest edit distance as the information extraction template.

[0038] The information extraction includes extracting physical entities, belonging relationships, action relationships, operation process entities, and operation process sequence relationships.

[0039] A system for implementing the fine-grained information extraction method based on part-of-speech tagging, the system comprising:

[0040] Editing and saving module for text to be analyzed: used for browsing and modifying the text to be analyzed;

[0041] Sentence analysis and part-of-speech tagging module: Use NLP tools to segment sentences and tag the decomposed phrases with parts of speech;

[0042] Relational entity extraction and preservation module: uses the results of sentence analysis and part-of-speech tagging to identify entities in sentences and the relationships between entities;

[0043] Custom phrase editing module: It can identify possible proprietary words based on the results of relational entity extraction, and can edit them through the human-computer interaction interface and save them to a specific file of the NLP tool;

[0044] Entity and relationship editing module: provides an editing interface for editing and correcting the relationships, entities, and related label attributes identified from sentences.

[0045] Beneficial effects of the present invention: Compared with the extraction method using deep learning technology, the present invention does not require tedious data labeling and model tuning; at the same time, compared with the traditional extraction method using syntactic analysis, the implementation of the extraction process does not require special induction of information extraction rules, and has strong scalability, clear operating mechanism, and strong interpretability of the extraction results.

[0046] At the same time, its phrase merging based on part of speech helps to identify new entities in the information extraction process.

[0047] Furthermore, the sentence classification and knowledge extraction model functions involved in this method are all customized by the user in advance. Therefore, this information extraction process can quickly adapt to new information extraction needs and can be easily transferred to other fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1This is a schematic diagram of the principle of the knowledge extraction method based on sentence structure matching.

[0049] Figure 2 It is a sentence encoding diagram based on part-of-speech tagging.

[0050] Figure 3 It is a sentence encoding comparison based on part-of-speech tagging.

[0051] Figure 4 This is a schematic diagram of the functional relationship of the ship surface support operation information automatic extraction tool software.

[0052] Figure 5 It is a graphical display of the results of sentence information extraction containing multiple verbs and parallel objects.

[0053] Figure 6 It is a graphical display of the sentence information extraction results including the process. DETAILED DESCRIPTION

[0054] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0055] It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same technical meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0056] A fine-grained information extraction method based on part-of-speech tagging, the method comprising the following steps:

[0057] S1: Obtain training sentences, use NLP tools to segment the sentences at the phrase level, and perform part-of-speech tagging on the decomposed phrases;

[0058] Adjacent phrases are merged and hidden according to the encoding strategy to form new part-of-speech tags;

[0059] Based on the new part-of-speech tagging and sentence structure, extract the specific part-of-speech sequence, keywords, and specific punctuation marks corresponding to the phrases in the sentence as the sentence code; the specific parts of speech include nouns, verbs, and verb phrases; the specific punctuation marks are symbols such as semicolons that indicate the parallel relationship of words;

[0060] Extract information from the current sentence by combining the information extraction strategy of the phrase's corresponding part of speech and its preceding and following related parts of speech, characteristic punctuation marks, or characteristic words, and save the sentence encoding and corresponding information extraction strategy of the current sentence as the encoding and information extraction strategy of the pre-stored typical example sentence;

[0061] S2: Segment the sentences to be analyzed at the phrase level and tag the phrases with parts of speech; merge and hide adjacent phrases according to the encoding strategy;

[0062] S3: Based on the parts of speech and sentence structure marked in S2, replace the nouns, verbs, and verb phrases in the sentence with parts of speech to form the code of the sentence to be analyzed;

[0063] S4: matching the encoding of the sentence to be analyzed with the encoding of pre-stored typical example sentences;

[0064] S5: If it matches the pre-stored typical example sentence, information extraction is performed on the sentence to be analyzed according to the extraction strategy of the matched typical example sentence;

[0065] S6: If there is no match, information is extracted from the sentence to be analyzed according to the information extraction strategy in S1; and the encoding of the sentence to be analyzed and the information extraction strategy are stored.

[0066] The information extraction strategy includes:

[0067] The "subject + predicate + object" sentence structure of NN1 VV NN2 can extract triples: entity 1 (represented by NN1) - action relationship (represented by VV) - entity 2 (represented by NN2), entities and relationships;

[0068] The attributive phrase of NN1 DEG NN2 can extract triples: entity 2 (represented by NN2) - belonging relationship - entity 1 (represented by NN1);

[0069] The attributive phrase of AD DEG NN can extract the attributes (represented by AD) in the entity (represented by NN);

[0070] The "subject + predicate + parallel object" sentence structure of NN1 VV NN2...NNn can extract triples: entity 1 represented by NN1 - action relationship embodied by VV - entity 2 represented by NN2, ..., entity 1 represented by NN1 - action relationship embodied by VV - entity n represented by NNn;

[0071] "Subject + predicate + object" or "predicate + object", sentence ① + "before" | "after" + "subject + predicate + object" or "predicate + object", sentence ②, can extract the process triple: operation process ① represented by sentence ① - the order of operation processes - operation process ② represented by sentence ②.

[0072] The encoding strategy includes:

[0073] A. Merge consecutive nouns (part of speech identifiers are NN (noun), NP (noun phrase), NR (personal name), NT (proper noun, organization name)) to generate new nouns;

[0074] B. Combine the phrases with the词性 annotation of JJ+NN (adjective + noun), and the词性 of the combined phrase is NN (noun);

[0075] C. Combine the continuously occurring verbs (词性标识 as VV, VP) to generate a new verb;

[0076] D. Combine the phrases with the词性 annotation of AD+VV (adverb + verb), and the词性 of the combined phrase is VV (verb);

[0077] E. Combine the phrases with the词性 annotation of VV+P (verb + preposition), and the词性 of the combined phrase is VP (verb phrase);

[0078] F. When the combination of noun NN1 + '的' DEG + noun NN2 appears, hide NN1 and DEG, and only retain the noun NN2;

[0079] G. When the combination of adverb AD + '的' DEG + noun NN appears, hide AD and DEG, and only retain the noun NN;

[0080] H. When the combination of noun NN1 + locative word LC + '的' DEG + noun NN2 appears, hide NN1, LC, and DEG, and only retain the noun NN2;

[0081] I. Hide the determiners in the sentence, and the determiners include '全部', '所有', '其它';

[0082] J. Hide all the content from '等' ETC to the end of the sentence;

[0083] K. Hide the participle symbols such as '《' and '》';

[0084] After combining and hiding the phrases in the sentence according to the above strategy, replace the nouns with the词性 of NN, the verbs with the词性 of VV, and the verb phrases with the词性 of VP, and the generated symbol string sequence is the sentence encoding.

[0085] In step S5, the encoding of the to-be-analyzed statement is pattern-matched with the encoding of the pre-stored typical example sentences by using a method based on the sentence structure similarity. Specifically: the differences between sentences are generally represented by the edit distance. In the actual sentence pattern matching process, first check whether the number of verbs contained in the to-be-matched sentence is the same as that in the sentence pattern. The to-be-matched sentence only compares the edit distance with the sentence encoding of the sentence with the same number of verbs, and then finds the sentence pattern with the smallest edit distance from its sentence encoding as the information extraction template.

[0086] The information extraction includes extracting physical entities, ownership relationships, action relationships, operation process entities, operation process sequence relationships, etc.

[0087] Sentence encoding:

[0088] The following example illustrates the information extraction process using the requirements for mooring cables: "The mooring cable should be connected from the aircraft's specific mooring fittings to the deck mooring fittings (hereinafter referred to as sentence ①) and should not compress other aircraft components such as hydraulic shock absorbers, hydraulic pipes, tires, etc. (hereinafter referred to as sentence ②)"

[0089] After treating "mooring rope" and "mooring accessories" as proper nouns, the sentence is segmented and tagged with parts of speech, and the adjacent phrases are merged and hidden according to the above merging rules: "should (VV)" and "from (P)" are merged into "should from (VV)"; "specific (JJ)" and "mooring accessories (NN)" are merged into "specific mooring accessories (NN)"; "connect (VV)" and "to (VV)" are merged into "connected to (VV)", and "deck (NN)" and "mooring accessories (NN)" are merged into "deck mooring accessories (NN)". Figure 2 shown.

[0090] Sentence ① generates the analysis result of {subject (mooring rope, NN) - predicate (should follow, VV) - {object {subordinate structure (aircraft - mooring accessories, NN)} - predicate (connection, VV) - object (deck mooring accessories, NN)}} based on the marked part of speech and the aforementioned information extraction strategy;

[0091] After completing the subject of sentence ② based on the context, the analysis result of {Subject (mooring rope, NN) - Predicate (must not be oppressed, VV) - Object {Parallel structure {Phrase 1 (hydraulic shock absorber, NN), Phrase 2 (hydraulic pipe, NN), Phrase 2 (tire, NN)}}} is generated according to the definitions of part of speech and sentence structure.

[0092] According to the above sentence encoding strategy, the sentence encoding (template) generated by this sentence is:

[0093] ①:NN VV NN VV NN ;

[0094] ②:VVNN, NN, NN.

[0095] Sentence pattern matching:

[0096] Now, the sentence "Qualified personnel shall perform pre-inspection operations in accordance with the provisions of the Naval Aviation Training and Operations Procedure Standardization Manual" (hereinafter referred to as sentence ③) is encoded and compared with the above-mentioned sentence ① according to the above strategy and the edit distance is compared.

[0097] After pre-processing "Naval Aviation Training and Operational Procedure Standardization Manual", "Pre-inspection Operations", "Qualified" and other words as proprietary words and special phrases, phrases and parts of speech are merged and hidden according to "Rule 1" to "Rule 8". The results are compared with the analysis results of sentence ①. Figure 3 shown.

[0098] The feature vectors of both sentences ① and ③ are: NN VP NN VV NN, and their edit distance is 0. Therefore, the information extraction method for sentence ③ can be completely carried out by the extraction method of sentence ①, without the need for further syntactic analysis.

[0099] The sentence analysis functionality of the ship surface support operations information extraction tool developed using the aforementioned method was implemented using the TexSmart offline SDK provided by Tencent Artificial Intelligence Lab. Due to the high number of specialized terms and specialized vocabulary within this knowledge domain, using the pre-trained language model's own general vocabulary alone for text segmentation and part-of-speech tagging was less than ideal. The TexSmart offline SDK, with its ability to add custom vocabulary and entities, identified domain-specific vocabulary and specialized phrases and appended them to the TexSmart offline SDK's corresponding predefined files in a specified format. In subsequent analysis, the SDK interprets these terms as a whole, bypassing any further analysis, ensuring the validity of the segmentation results within this domain. Furthermore, the information extraction tool also generates custom vocabulary from entity names, allowing for dynamic addition and adjustment of these custom vocabulary based on the information extraction results. The content to be analyzed and the analysis results are stored locally as XML files, facilitating cross-reference and verification of the extraction results.

[0100] The logical relationship between the various functions of the tool software is as follows Figure 4 shown.

[0101] Functions of tool-related modules: Text editing and saving module for analyzing, used for browsing and modifying text to be analyzed; sentence analysis and part-of-speech tagging module, which uses TexSmart offline SDK to segment sentences and tag the decomposed phrases with parts of speech; relational entity extraction and saving module, which uses the results of sentence analysis and part-of-speech tagging to identify entities in sentences and the relationship between entities; custom phrase editing module, which can identify possible proprietary vocabulary based on the results of relational entity extraction, and can edit and save it to a specific file of TexSmart offline SDK through the human-computer interaction interface; entity and relationship editing module, which provides an editing interface for editing and correcting the relationships, entities and related label attributes identified from the sentences; Cypher statement generation module, which generates Cypher statements for the currently extracted entities and relationships; xml file access and parsing, which provides the xml stream data access and parsing functions required by the software.

[0102] Information extraction example description

[0103] For sentences ① and ② above, based on the knowledge model of ship surface support operations and combined with the analysis of the sentence structure, the information extraction results are as follows:

[0104] Physical entities (7): mooring ropes, aircraft, mooring accessories, deck mooring accessories, hydraulic shock absorbers, hydraulic pipes, tires, etc.;

[0105] Affiliation (1): Aircraft - tie-down accessories;

[0106] Action relationships (5): Mooring rope - <should> be connected to the mooring accessories - deck mooring accessories, Mooring rope - <must not> press on the hydraulic shock absorber, Mooring rope - <must not> press on the hydraulic pipe, Mooring rope - <must not> press on the tire.

[0107] Based on the above extraction results, generate Cypher statements and submit them to Neo4j for interpretation and execution. The graphical display results are as follows: Figure 5 shown.

[0108] Clauses involving steps are analyzed similarly to the above analysis process. The difference is that the sentence to be analyzed is first divided into different step clauses using keywords representing the sequence of steps. When creating entities later, entities representing the step activities must be created. For example, "Before the aircraft is released from the tow bar, the parking brakes for the aircraft and the tow vehicle are lowered and the wheel chocks are installed." (hereinafter referred to as sentence ④) Due to the presence of the keyword "before", the relevant sentence is partially decomposed into three steps: "the aircraft is released from the tow bar," "the parking brakes for the aircraft and the tow vehicle are lowered," and "the wheel chocks are installed."

[0109] Based on the knowledge model of ship surface support operations and the analysis of the aforementioned sentence structure, the information extraction results are as follows:

[0110] Physical entities (5): parking brake, tractor, aircraft, drawbar, wheel chock;

[0111] Operation steps (3): disengage the drawbar, apply the parking brake, and install the wheel chocks;

[0112] Affiliation (2): tractor-parking brake, aircraft-parking brake;

[0113] Action relationship (1): aircraft-installation-wheel chock;

[0114] The process involves three entity relationships: drawbar - disengage - disengage the drawbar, parking brake - lower - lower the parking brake, wheel chock - install - install wheel chock;

[0115] The sequence of processes (2): Before disengaging the drawbar, lower the parking brake; Before disengaging the drawbar, install the wheel chocks.

[0116] Based on the above extraction results, generate Cypher statements and submit them to Neo4j for interpretation and execution. The graphical display results are as follows: Figure 6 shown.

[0117] The above is only a preferred embodiment of the present invention. It should be pointed out that for those skilled in the art, several changes and improvements can be made without departing from the overall concept of the present invention, and these should also be regarded as the scope of protection of the present invention.

Claims

1. A fine-grained information extraction method based on part-of-speech tagging, characterized by: The method comprises the following steps: S1: Obtain training sentences, use NLP tools to segment the sentences at the phrase level, and perform part-of-speech tagging on the decomposed phrases; Adjacent phrases are merged and hidden according to the encoding strategy to form new part-of-speech tags; Based on the new part-of-speech tagging and sentence structure, extract the specific part-of-speech sequence, keywords, and specific punctuation marks corresponding to the phrases in the sentence as the sentence code; the specific parts of speech include nouns, verbs, and verb phrases; the specific punctuation marks are symbols that indicate the parallel relationship between words; Extract information from the current sentence by combining the information extraction strategy of the phrase's corresponding part of speech and its preceding and following related parts of speech, characteristic punctuation marks, or characteristic words, and save the sentence encoding and corresponding information extraction strategy of the current sentence as the encoding and information extraction strategy of the pre-stored typical example sentence; S2: Segment the sentences to be analyzed at the phrase level and tag the phrases with parts of speech; merge and hide adjacent phrases according to the encoding strategy; S3: Based on the parts of speech and sentence structure marked in S2, replace the nouns, verbs, and verb phrases in the sentence with parts of speech to form the code of the sentence to be analyzed; S4: matching the encoding of the sentence to be analyzed with the encoding of pre-stored typical example sentences; S5: If it matches the pre-stored typical example sentence, information extraction is performed on the sentence to be analyzed according to the extraction strategy of the matched typical example sentence; S6: If there is no match, extract information from the sentence to be analyzed according to the information extraction strategy in S1; and store the encoding of the sentence to be analyzed and the information extraction strategy.

2. The fine-grained information extraction method based on part-of-speech tagging according to claim 1 is characterized in that: The information extraction strategy includes: The "subject + predicate + object" sentence structure of NN1 VV NN2 can extract triples: entity 1 represented by NN1 - action relationship embodied by VV - entity 2 represented by NN2, entities and relationships; The attributive phrase of NN1 DEG NN2 can extract triples: entity 2 represented by NN2 - belonging relationship - entity 1 represented by NN1; The attributive phrase of AD DEG NN can extract the attribute represented by AD from the entity represented by NN; The "subject + predicate + parallel object" sentence structure of NN1 VV NN2…NNn can extract triples: entity 1 represented by NN1 - the action relationship embodied by VV - entity 2 represented by NN2, …, entity 1 represented by NN1 - the action relationship embodied by VV - entity n represented by NNn; "Subject + predicate + object" or "predicate + object", sentence ① + "before" | "after" + "subject + predicate + object" or "predicate + object", sentence ②, can extract the process triple: operation process ① represented by sentence ① - the order of operation processes - operation process ② represented by sentence ②.

3. The fine-grained information extraction method based on part-of-speech tagging according to claim 1 is characterized in that: The encoding strategy includes: A. Mark consecutive nouns as noun NN, noun phrase NP, personal name NR, proper noun NT, and merge them to generate a new noun NN. B. Combine the phrases with the part-of-speech tags of adjective JJ + noun NN, and the part-of-speech of the combined phrase is noun NN; C. Combine the continuously appearing verbs with the part-of-speech tags of VV and VP to generate a new verb VV; D. Combine the phrases with the part-of-speech tags of adverb AD + verb VV, and the part-of-speech of the combined phrase is verb VV; E. Combine the phrases with the part-of-speech tags of verb VV + preposition P, and the part-of-speech of the combined phrase is verb phrase VP; F. When the combination of noun NN1 + '的' DEG + noun NN2 appears, hide NN1 and DEG, and only retain noun NN2; G. When the combination of adverb AD + '的' DEG + noun NN appears, hide AD and DEG, and only retain noun NN; H. When the combination of noun NN1 + locative word LC + '的' DEG + noun NN2 appears, hide NN1, LC, and DEG, and only retain noun NN2; I. Hide the determiners in the sentence, and the determiners include '全部' (all), '所有' (all), '其它' (other); J. Hide all the content from '等' (etc.) to the end of the sentence; K. Hide the segmentation symbols such as '《' and '》'; After combining and hiding the phrases in the sentence according to the above strategy, replace the nouns with the part-of-speech of NN, the verbs with the part-of-speech of VV, and the verb phrases with the part-of-speech of VP with their part-of-speech, and the generated symbol string sequence is the sentence encoding.

4. The fine-grained information extraction method based on part-of-speech tagging according to claim 1 is characterized in that: In step S4, the encoding of the to-be-analyzed statement is pattern-matched with the encoding of the pre-stored typical example sentences by using a method based on the sentence structure similarity. Specifically: the differences between sentences are generally represented by the edit distance. In the actual sentence pattern matching process, first check whether the number of verbs contained in the to-be-matched sentence is the same as that in the sentence pattern. The to-be-matched sentence only performs an edit distance comparison with the sentence encoding with the same number of verbs as it, and then finds the sentence pattern with the smallest edit distance from its sentence encoding as the information extraction template.

5. The fine-grained information extraction method based on part-of-speech tagging according to claim 1 is characterized in that: The information extraction includes extracting physical entities, ownership relationships, action relationships, operation process entities, and the sequence of operation processes.

6. A system for implementing the fine-grained information extraction method based on part-of-speech tagging according to any one of claims 1 to 5, characterized in that: The system includes: To-be-analyzed text editing and saving module: used for browsing and modifying the to-be-analyzed text; Statement analysis and part-of-speech tagging module: uses NPL tools to segment the sentence and perform part-of-speech tagging on the segmented phrases; Relationship entity extraction and saving module: uses the results of statement analysis and part-of-speech tagging to identify the entities in the statement and the association relationships between entities; Custom phrase editing module: can identify possible proper nouns according to the results of relationship entity extraction, and can be edited through a man-machine interaction interface and saved to a specific file of the NPL tool; Entity and relationship editing module: provides an editing interface, and can edit and correct the relationships, entities, and related tag attributes identified from the sentence.

Images