Online query and reading method and system for Chinese language and literature database

Abstract

The application relates to the technical field of information retrieval, in particular to a Chinese language and literature database online query reading method and system, which comprises the following steps: extracting a word meaning item, a literature reference, a meaning length and a boundary mark, screening a candidate set, analyzing a subject pronoun and a verb collocation, positioning a paragraph and marking a word meaning item position, extracting a phrase structure comparison version offset, marking a keyword meaning number and a position, outputting a meaning, a source and a semantic path, and generating a comparison prompt; in the application, the positioning, display and structure alignment of the literature are optimized through semantic mapping, the accuracy and consistency of the target literature are improved, relevant content is accurately screened in combination with context information and version comparison, the structure offset between versions is automatically adjusted, the dependence on manual keyword setting is reduced, meanwhile, the generated word meaning comparison prompt group is optimized to improve the literature retrieval experience, the user can more quickly and accurately understand the word meaning and context relationship, and the information acquisition precision and efficiency are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of information retrieval, in particular to a Chinese language and literature database online query reading method and system. BACKGROUND

[0002] The technical field of information retrieval involves the organization, storage, indexing, and querying of text data, mainly including natural language-based text analysis, structured semantic processing, keyword matching, and associated information positioning methods, aiming to achieve fast access to specific information targets by constructing retrieval logic and query mechanisms. The methodological features of this field are reflected in the use of steps such as lexical analysis, text comparison, and language model construction to efficiently locate and identify massive information resources. The traditional Chinese language and literature database online query reading method refers to a data management solution for ancient and modern Chinese language and literature text resources. It addresses the issue of how to handle complex language structures and unstructured text data in Chinese language and literature documents during information retrieval. Traditional methods use a term indexing mechanism based on specific document classification numbers such as TP3913 information retrieval systems, and perform text matching and distribution through manually set keywords or general keyword extraction algorithms to achieve online browsing and preliminary filtering of target materials. This method generally uses fixed classification matching, word frequency statistics, or pattern recognition methods to complete information calling and display.

[0003] The main limitations of existing technologies lie in the handling of complex text structures and semantic associations. Traditional information retrieval methods often rely on fixed vocabulary classification and pattern matching, using manually set keywords for text filtering. However, this approach cannot fully address the diverse language structures and complex semantic relationships in documents. Especially when dealing with ancient Chinese language and literature documents with rich context and context dependence, keyword matching often leads to misleading results, resulting in decreased accuracy and relevance of the search results. Existing systems also fail to effectively handle differences between document versions, lacking a dynamic adjustment mechanism, which makes it difficult to automatically handle structural shifts and content inconsistencies when comparing the same information in different versions. This processing method limits the comprehensiveness and accuracy of information, making it difficult for users to quickly and accurately locate information during use, affecting retrieval efficiency and reading experience. SUMMARY

[0004] To solve the technical problems existing in the prior art, the embodiments of the present application provide a Chinese language and literature database online query reading method.

[0005] To achieve the above-mentioned purposes, the present application adopts the following technical solutions:

[0006] The Chinese language and literature database online query reading method comprises the following steps:

[0007] S1: obtain the word sense item of the search term, extract the core ancient literature citation frequency, the length of the explanation text and the number of sense item boundary annotations, integrate in combination with the literature time sequence weight, screen the word sense item meeting the condition, and generate a semantic candidate set;

[0008] S2: based on the word sense item in the semantic candidate set, extract the subject pronoun and verb collocation structure, analyze the associated and sorted path of the semantic subject and action verb in the explanation, and generate a context semantic mapping path sequence;

[0009] S3: according to the word sense item in the priority path in the context semantic mapping path sequence and the chapter information, locate the corresponding paragraph in the version index, mark the target word sense item paragraph number and literature location, construct the retrieval structure, and generate a version paragraph retrieval scheduling table;

[0010] S4: call the content of the version paragraph retrieval scheduling table, extract the target word sense item associated phrase structure and order, compare the structure order offset between versions, adjust the paragraph position exceeding the set standard, and generate a word sense item mapping paragraph alignment result;

[0011] S5: based on the word sense item mapping paragraph alignment result, mark the word sense number and position of the key word in the paragraph, organize the reading area display structure, output the word sense explanation, literature source and semantic path information, and generate a word sense comparison prompt block.

[0012] As a further scheme of the present application, the semantic candidate set includes a candidate word sense item list, a word sense feature parameter table, an entry threshold value identifier, and a score summary. The context semantic mapping path sequence includes a path sequence number set, a relationship strength level set, a subject-verb collocation snapshot set, and an explanation subject mapping label set. The version paragraph retrieval scheduling table includes a version identification code, a paragraph number list, a literature location coordinate, and a retrieval routing item. The word sense item mapping paragraph alignment result includes an associated phrase structure table, a position order sequence, a structure order offset result, and a position adjustment record. The word sense comparison prompt block includes a word sense number annotation area, a key word position index, a literature source item, a path prompt information card, and a reading area display layout.

[0013] As a further scheme of the present application, the specific steps of S1 are:

[0014] S101: obtain the word sense item corresponding to the search term in the Chinese language and literature database, collect the citation frequency, the length of the explanation text and the number of sense item boundary annotations of the word sense item in the core ancient literature, arrange the data into a unified format, and generate a literature word sense parameter;

[0015] S102: Based on the literature word meaning parameter, call the literature time weight difference age ancient book literature word meaning item frequency processing, and compare with the interpretation text length and the number of meaning item boundary mark, generate the word meaning index;

[0016] S103: According to the word meaning index, the word meaning item is screened, the word meaning item set meeting the condition is reserved, and the word meaning item in the set is arranged according to the predetermined sorting standard, and the semantic candidate set is generated.

[0017] As a further scheme of the application, the specific steps of S2 are:

[0018] S201: Obtain the word meaning item in the semantic candidate set, extract the subject pronoun and verb collocation structure of the paragraph where the word meaning item is located, arrange the subject pronoun and verb into a combination record, and generate the paragraph collocation combination quantity.

[0019] S202: Based on the paragraph collocation combination quantity, call the semantic subject and behavior verb in the word meaning interpretation, compare the subject pronoun and verb collocation structure in the paragraph, judge the corresponding relationship and record the strength result, and obtain the semantic correlation strength value.

[0020] S203: According to the semantic correlation strength value, the path corresponding to the word meaning item is sorted, the path result is merged according to the paragraph position, and the context semantic mapping path sequence is generated.

[0021] As a further scheme of the application, the specific steps of S3 are:

[0022] S301: Obtain the priority path in the context semantic mapping path sequence, extract the word meaning item and chapter information under the path, detect the paragraph position corresponding to the word meaning item in the version index, and record the paragraph number and index mark, generate the paragraph index number quantity.

[0023] S302: Based on the paragraph index number quantity, call the paragraph number and literature position containing the target word meaning item, judge the corresponding relationship of the paragraph number and position data, screen the combination structure capable of uniquely identifying the paragraph, and obtain the paragraph positioning matching value.

[0024] S303: According to the paragraph positioning matching value, establish the calling structure of the paragraph number and literature position in the version index, merge all combination relationships into retrieval paths in turn, and generate the version paragraph retrieval scheduling table.

[0025] As a further scheme of the application, the specific steps of S4 are:

[0026] S401: Call the paragraph number and literature position in the version paragraph retrieval scheduling table, extract the associated phrase structure and appearance order of the target word meaning item in the paragraph, compare the number and position relationship between adjacent phrases, and generate the phrase order mark quantity.

[0027] S402: Based on the phrase order marking quantity, the offset degree of the structure order in the difference version is calculated, the number difference value of the paragraph is compared with the set order offset standard value one by one, the paragraph set exceeding the standard is obtained, and the order offset difference value is generated;

[0028] S403: According to the order offset difference value, the position of the paragraph exceeding the standard is adjusted, the numbering and literature position are rearranged, the paragraph order is called, all the adjusted paragraph positions are mapped to the corresponding word meaning item combination, and the word meaning item mapping paragraph alignment result is generated.

[0029] As a further scheme of the application, the specific steps of S5 are:

[0030] S501: Based on the word meaning item mapping paragraph alignment result, the word meaning number corresponding to the keyword and the position index in the paragraph are marked segment by segment, the occurrence frequency and position difference value of the keyword in the paragraph are called, the combination of the number and the position index is operated, and the word meaning position marking quantity is generated;

[0031] S502: According to the word meaning position marking quantity, the word meaning number marked in the paragraph is sorted, the position index of the keyword corresponding to the number in the reading area is matched with the paragraph structure, the sorting difference value is counted and compared with the set display reference value, and the reading area sequence difference value is obtained;

[0032] S503: The reading area sequence difference value is called, the paragraphs corresponding to the difference value exceeding the display reference value are reorganized, the word meaning explanation, literature source and path information are output according to the paragraph order, the content is combined into a prompt block, and the word meaning comparison prompt block is generated.

[0033] As a further scheme of the application, the core ancient literature refers to the representative historical literature in the field of Chinese language and literature, including Shangshu, Zuo Zhuan, and classical literature of Han Dynasty in Shiji;

[0034] The literature time sequence weight refers to the weighted value given to the word meaning item in different literature according to the age, school and influence in language evolution of the literature;

[0035] The semantic candidate set refers to all candidate word meaning item sets filtered by multi-dimensional parameters;

[0036] The subject pronoun refers to the pronoun representing the subject in the sentence;

[0037] The verb collocation structure refers to the grammar structure composed of a verb and associated nouns, adjectives or other modifiers;

[0038] The context semantic mapping path sequence refers to the path set generated according to the semantic relationship sorting in the context.

[0039] As a further scheme of the present application, the version index comprises different document versions, paragraph positions and chapter structures;

[0040] The paragraph number and document position refer to identifiers in the document;

[0041] The version paragraph retrieval schedule refers to a retrieval order and retrieval structure generated based on the different version documents and the target paragraph in the query process;

[0042] The word sense item associated phrase structure refers to a phrase, word group or sentence structure closely associated with the target word sense item in the document paragraph, and the structure is consistent with the semantics of the word sense item;

[0043] The structure order offset refers to the order difference of phrases or word groups in sentences in different version documents. By calculating the structure order difference between different versions, the order and adjustment are performed;

[0044] The word sense item mapping paragraph alignment result refers to the paragraph alignment result generated by adjusting the structure order between versions;

[0045] The word sense comparison prompt block refers to a word sense comparison framework displayed to the user during reading, including word sense interpretation, document source and path information.

[0046] Based on the same invention concept, an online query reading system for Chinese literature database is also proposed, which is used to implement the online query reading method for Chinese literature database as described above, comprising:

[0047] The word sense candidate module obtains the word sense item frequency of citation, the interpretation text length, the number of boundary annotations and the document time sequence weight in the core ancient document, and sets a screening threshold to screen the word sense items, and generates a semantic candidate set;

[0048] The semantic path module extracts the subject pronoun and verb collocation based on the word sense items in the semantic candidate set in the corresponding paragraph, calculates the matching strength of the semantic subject and the action verb and sorts them, and generates a context semantic mapping path sequence;

[0049] The version positioning module locates the paragraph number and document position in the version index and records the coordinates according to the priority path in the context semantic mapping path sequence, combines the word sense item and chapter information, and generates a version paragraph retrieval schedule;

[0050] The paragraph alignment module calls the version paragraph retrieval schedule, extracts the phrase structure and position order of the target word sense item, calculates the order offset between different versions and adjusts the paragraphs exceeding the threshold, and generates a word sense item mapping paragraph alignment result;

[0051] The reading prompt module marks the keyword meaning number and position in the reading area based on the paragraph alignment result of the word meaning item mapping, organizes the display of literature source and path information, and generates a word meaning comparison prompt group.

[0052] Compared with the prior art, the application has the advantages and positive effects that:

[0053] In the application, the positioning, display and structure alignment of the literature are optimized through semantic mapping, the accuracy and consistency of the target literature are improved, the related content is accurately screened in combination with context information and version comparison, the structure deviation between versions is automatically adjusted, the dependence on manual keyword setting is reduced, the generated word meaning comparison prompt group optimizes the literature retrieval experience, the user can more quickly and accurately understand the meaning of the vocabulary and the context relationship, and the accuracy and efficiency of information acquisition are improved. BRIEF DESCRIPTION OF DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0055] Figure 1 The step flowchart of the application is shown in the figure.

[0056] Figure 2 The S1 refinement schematic diagram of the application is shown in the figure.

[0057] Figure 3 The S2 refinement schematic diagram of the application is shown in the figure.

[0058] Figure 4 The S3 refinement schematic diagram of the application is shown in the figure.

[0059] Figure 5 The S4 refinement schematic diagram of the application is shown in the figure.

[0060] Figure 6 The S5 refinement schematic diagram of the application is shown in the figure.

[0061] Figure 7 The system module diagram of the application is shown in the figure. DETAILED DESCRIPTION

[0062] The technical solutions in the application will be described below with reference to the drawings.

[0063] In order to make the technical problems, technical solutions and advantages of the application clearer, the following will be described in detail with reference to the drawings and specific embodiments.

[0064] Please refer to Figure 1The embodiment of the present application provides a Chinese language and literature database online query reading method, which comprises the following steps:

[0065] S1: Obtain the word meaning item corresponding to the search term in the Chinese language and literature database, extract the frequency of citation, interpretation text length and meaning item boundary mark number in the core ancient literature, integrate the parameters in combination with the literature time sequence weight, screen the word meaning items meeting the requirements, and generate a semantic candidate set;

[0066] The Chinese language and literature database is a digital literature database including ancient literature and modern Chinese academic data, and the database includes ancient Chinese vocabulary, grammar and multi-version data of changes;

[0067] The core ancient literature refers to the representative historical literature in the field of Chinese language and literature, including the classic literature of Han Dynasty such as Shangshu, Zuo Zhuan and Shiji;

[0068] The literature time sequence weight refers to the weighted value given to the word meaning item in different literature according to the age, school and influence of the literature in language evolution;

[0069] The semantic candidate set refers to a collection of all candidate word meaning items screened through multi-dimensional parameters (including frequency and interpretation length);

[0070] S2: Based on the word meaning item in the semantic candidate word meaning set, the subject pronoun and verb collocation structure in the corresponding paragraph are extracted, the association with the semantic subject and behavior verb in the word meaning interpretation is judged, the path is sorted according to the association strength, and the context semantic mapping path sequence is generated;

[0071] The subject pronoun refers to the pronoun representing the subject in the sentence, which is used to reflect the theme of the sentence in the context;

[0072] The verb collocation structure refers to the grammatical structure of verb and associated noun, adjective or other modifiers, which is used to capture the use mode of verb in the context;

[0073] The association strength refers to the matching degree between the subject pronoun and the semantic subject in the word meaning, and the behavior verb and the behavior verb in the word meaning, which is used to judge the association closeness of the word meaning item and the context;

[0074] The context semantic mapping path sequence refers to the path set generated according to the semantic relationship sorting in the context, which is used to accurately match the query word meaning item;

[0075] S3: According to the word meaning item of the priority path in the context semantic mapping path sequence and the chapter information, the corresponding paragraph is located in the version index, the paragraph number and literature position containing the target word meaning item are marked, the retrieval structure is constructed, and the version paragraph retrieval scheduling table is generated;

[0076] Version index is an index including different versions of documents, paragraph positions and chapter structures, which is used to locate the content in the documents;

[0077] Paragraph number and document position refer to identifiers in the documents, which are used to mark the target paragraph and the specific position in the document, and help to retrieve the content of the target paragraph;

[0078] Version paragraph retrieval scheduling table refers to the retrieval order and retrieval structure generated based on the different version documents and the target paragraph in the query process, which is used to organize the order and index of the paragraph content in the retrieval system;

[0079] S4: Call the content in the version paragraph retrieval scheduling table, extract the target word sense item associated phrase structure and position order, compare the structure order offset between different versions, adjust the position of the paragraph whose order offset exceeds the set standard, and generate the word sense item mapping paragraph alignment result;

[0080] Word sense item associated phrase structure refers to the phrase, word group or sentence structure closely associated with the target word sense item in the document paragraph, which is consistent with the semantics of the word sense item;

[0081] Structure order offset refers to the order difference of phrases or word groups in sentences in different versions of documents. By calculating the structure order difference between different versions, the order is sorted and adjusted;

[0082] Word sense item mapping paragraph alignment result refers to the paragraph alignment result generated by adjusting the structure order between versions to ensure semantic consistency and structure matching;

[0083] S5: Based on the word sense item mapping paragraph alignment result, mark the word sense number and position corresponding to the key words in the paragraph, organize the reading area display structure, output the word sense interpretation, document source and path information, and generate the word sense comparison prompt block;

[0084] Word sense comparison prompt block refers to the word sense comparison framework displayed in the user reading process, including word sense interpretation, document source and path information, which is used to improve the convenience of document comparison and understanding.

[0085] Semantic candidate set includes candidate word sense item list, word sense feature parameter table, selection threshold identification, score summary, context semantic mapping path sequence includes path sequence number set, relationship strength level set, subject-verb collocation snapshot set, interpretation subject mapping marker set, version paragraph retrieval scheduling table includes version identification code, paragraph number list, document position coordinates, retrieval routing item, word sense item mapping paragraph alignment result includes associated phrase structure table, position order sequence, structure order offset result, position adjustment record, word sense comparison prompt block includes word sense number annotation area, key word position index, document source item, path prompt information card, reading area display layout.

[0086] Please refer to Figure 2 The specific steps of S1 are as follows:

[0087] S101: Obtain the word meaning items corresponding to the search term in the Chinese language and literature database, collect the frequency of occurrence, the length of the explanatory text and the number of boundary annotations of the word meaning items in the core ancient literature, arrange the data into a unified format, and generate the literature word meaning parameters;

[0088] When obtaining the word meaning items corresponding to the search term in the Chinese language and literature database, first map the term to the number in the database one by one, for example, the meaning item number of "Ren" is A1, A2, A3, then collect the frequency of occurrence of these meaning items in the core ancient literature, for example, A1 appears 15 times in "Analects of Confucius" and 8 times in "Mencius", A2 appears 6 times in different literature, then count the length of the explanatory text, the explanatory text of A1 is 120 characters, the explanatory text of A2 is 85 characters, and the explanatory text of A3 is 60 characters, and at the same time, count the number of boundary annotations, A1 has 4 boundaries, A2 has 2 boundaries, and A3 has 3 boundaries, arrange the above collected frequency, length and boundary annotation number into corresponding records, arrange them into a data set in a fixed order, each record consists of three parameters, and obtain the literature word meaning parameters.

[0089] S102: Based on the literature word meaning parameters, call the literature time sequence weight to process the frequency of occurrence of word meaning items in different age ancient literature, and combine the length of the explanatory text and the number of boundary annotations to generate the word meaning index;

[0090] Based on the literature word meaning parameters, the ancient literature of different ages needs to be weighted, and the time sequence weight of different ages is set, for example, the first Qin Dynasty is 1.2, the Han Dynasty is 0.9, and the Tang Dynasty is 0.7, when a meaning item appears 10 times in the literature of the first Qin Dynasty and 5 times in the Han Dynasty, the weighted frequency of occurrence is 16.5, then the length of the explanatory text and the number of boundary annotations are compared, the threshold value of the length of the explanatory text is set to 100 characters, if the number of explanatory characters exceeds 100, it is recorded as 1, otherwise as 0, the reference value of the number of boundary annotations is 3, if it is greater than or equal to 3, it is recorded as 1, otherwise as 0, for example, the length of the explanatory text of A1 is 120 characters, which is greater than the threshold value and is recorded as 1, the number of boundary annotations is 4, which is greater than the reference value and is recorded as 1, the weighted frequency is added to the two recorded values to obtain a comprehensive value, the larger the value, the more prominent the weight in the parameter set, and the same operation is performed on all meaning items to obtain the word meaning index.

[0091] S103: According to the word meaning index, screen the word meaning items, retain the word meaning item set that meets the conditions, and arrange the word meaning items in the set according to the predetermined sorting standard to generate the semantic candidate set;

[0092] According to the word meaning index, all word meaning items are screened, and the screening threshold is set to 10. If the value is greater than or equal to 10, it is retained, otherwise discarded. For example, A1 value is 17.5, which meets the condition and is retained, and A2 value is 7.3, which is discarded. The retained meaning items after screening are combined into a set, for example, C = {A1, A3}. Then arrange them in order from large to small according to the index value. If the values are the same, adjust the order from short to long according to the length of the explanation text. For example, A1 value is 17.5 and A3 value is 11.2. A1 is arranged before A3. If A3 and A4 have the same value, compare the number of words in the explanation text. The one with fewer words is arranged in front. After sorting, a new set sequence is obtained, and a semantic candidate set is generated.

[0093] Please refer to Figure 3 The specific steps of S2 are as follows:

[0094] S201: Obtain the word meaning items in the semantic candidate set, extract the subject pronoun and verb collocation structure in the paragraph where the word meaning item is located, arrange the subject pronoun and verb into a combination record, and generate the paragraph collocation combination amount.

[0095] When obtaining the word meaning items in the semantic candidate set, the sentence structure of the paragraph where the word meaning item is located needs to be checked item by item, and the subject pronoun and verb collocation are extracted. First, identify the pronoun in the sentence through syntactic decomposition method, and then locate the verb action collocated with it. For example, in the sentence "Its speech is good", "its" is the subject pronoun and "speech" is the verb collocation. In the sentence "He attacks the city", "he" and "attack" form a collocation. Record these collocations one by one according to the paragraph number to form a unified combination record, which contains three elements: paragraph number, subject pronoun and verb. To maintain consistency, a fixed text format can be used for recording, such as "D1-its-speech" and "D2-he-attack". If the same collocation structure appears in the same paragraph, the number of times is added, for example, "its-speech" appears twice in D1, which is recorded as "D1-its-speech x 2". As the number of processed paragraphs increases, these collocation combinations are aggregated into a whole data set, which can be regarded as a paragraph collocation combination amount, which is used for subsequent semantic comparison operations.

[0096] S202: Based on the paragraph collocation combination amount, call the semantic subject and behavior verb in the word meaning explanation, compare the subject pronoun and verb collocation structure in the paragraph, judge the corresponding relationship and record the strength result, and obtain the semantic association strength value.

[0097] Based on the paragraph collocation combination quantity, the semantic subject and behavior verb in the word meaning disambiguation need to be called for comparison processing. First, the disambiguation text is sorted, and the semantic subject such as “person”, “official”, “lord” and the like is recognized. Then, the verbs related to the subject are extracted from the disambiguation sentence, for example, “love”, “guard” and “walk”. Next, the semantic combination table is formed, such as “person-love”, “official-guard” and “lord-walk”. Then, the subject pronoun in the paragraph collocation combination is sequentially corresponded to the semantic combination table. If the object pointed by the pronoun is consistent with the semantic subject and the verb meaning is consistent, it is determined as strong correspondence. If only part of it matches, it is medium correspondence. If it is completely inconsistent, it is weak correspondence. In order to ensure quantification, the strong correspondence can be set as 100%, the medium correspondence as 50%, and the weak correspondence as 0%. For example, when “its-walk” in D1 corresponds to the disambiguation “person-walk”, it can be considered as 100%. When “their-attack” in D2 corresponds to the disambiguation “official-guard”, only the verb part is inconsistent, so it is recorded as 0%. For example, when “certain-follow” in D3 corresponds to the disambiguation “official-walk”, the subject is inconsistent and the verb is close, so it can be marked as 50%. All paragraph combinations are compared one by one, the results are recorded and unified to form the semantic correlation strength value, which provides a basis for subsequent sorting.

[0098] S203: According to the semantic correlation strength value, the path corresponding to the word meaning item is sorted, the path results are merged according to the paragraph position, and the context semantic mapping path sequence is generated.

[0099] According to the semantic correlation strength value, the path corresponding to each word meaning item needs to be sorted. Each path is arranged in order according to the strength value. If a certain meaning item appears in multiple paragraphs with different strength results, for example, A1 corresponds to D1 with a strength of 100% and D2 with a strength of 50%, D1 is arranged first and then D2. If A2 corresponds to D3 with a strength of 80% and D4 with a strength of 30%, D3 is arranged first and then D4. When the strength values are the same, the paragraph order is sorted. For example, if the strength of two paragraphs is 70%, the paragraph number with a smaller number is arranged in front. After sorting, the results are merged according to the paragraph position. The paths of different meaning items in the same paragraph are merged. For example, D1 contains A1, D2 contains A1 and A3, and D3 contains A2. After integration, a continuous context semantic mapping path sequence is formed. This sequence can cover all paragraphs and reflect the arrangement results of each meaning item, and a complete context semantic mapping path sequence is obtained.

[0100] Please refer to Figure 4 , the specific steps of S3 are as follows:

[0101] S301: Obtain the priority path in the context semantic mapping path sequence, extract the word meaning item and chapter information under the path, detect the paragraph position corresponding to the word meaning item in the version index, and record the paragraph number and index mark to generate the paragraph index number quantity.

[0102] The priority path in the context semantic mapping path sequence is obtained by retrieving each path in the sequence, comparing the number of occurrences of the path and the semantic correlation strength to determine the priority. For example, when a certain path appears more than 5 times in the sequence and the strength value is greater than 70%, it can be marked as a priority path. Then, relevant word sense items and chapter information are extracted under the priority path. These word sense items are matched with the corresponding paragraphs one by one. The paragraph location where the word sense item is located is detected in the version index. The detection process needs to scan the paragraph number one by one, and the corresponding paragraph entries are recorded. Then, index markers are added to the entries, such as paragraph numbers D1 and D3, corresponding word sense item A can be recorded as D1-A and D3-A. If the same word sense item appears multiple times in a paragraph, the number of occurrences is accumulated and recorded as D1-A x 2. Such records are uniformly included in the paragraph index number amount. After the accumulation of multiple paragraph data, a set containing paragraph numbers and index markers is formed, which is used for subsequent retrieval and call.

[0103] S302: Based on the paragraph index number amount, the paragraph number containing the target word sense item and the literature location are called. The paragraph number and the location data are matched to determine the combination structure that can uniquely identify the paragraph, and the paragraph positioning matching value is obtained.

[0104] Based on the paragraph index number amount, the paragraph number containing the target word sense item and the literature location data are called. The paragraph number and the literature location are matched and judged. First, the index number amount is expanded, such as D1-A, D3-A, and D2-B. Then, the corresponding literature location data is retrieved from the literature index table, such as paragraph D1 corresponding to P12, paragraph D3 corresponding to P27, and paragraph D2 corresponding to P15. The paragraph number and the literature location are matched one by one. If one paragraph number corresponds to multiple locations, the combination structure that can uniquely identify the paragraph needs to be selected. The selection criteria can be based on the uniqueness of the location number and the matching degree of the paragraph number. When there is a offset between the location number and the paragraph number, the unique value can be selected by comparing the size of the offset, such as paragraph D1 corresponding to location P12.1 and P12.3. The smallest offset can be used to determine P12.1 as the unique matching value. Each paragraph number only retains one corresponding literature location, forming the paragraph positioning matching value. After processing, unique combination results such as D1-P12.1, D3-P27.2, and D2-P15.1 are obtained.

[0105] S303: According to the paragraph positioning matching value, the call structure of the paragraph number and the literature location is established in the version index. All combination relationships are merged into retrieval paths one by one to generate a version paragraph retrieval scheduling table.

[0106] According to the paragraph positioning matching value, it is necessary to establish the call structure of the section number and the document position in the version index. First, the section number is combined with the matching value to form a mapping relationship, for example, D1 corresponds to P12.1, D2 corresponds to P15.1, and D3 corresponds to P27.2. Then, all the mapping relationships are merged in sequence to form a retrieval path. In the merging process, the order of the section numbers is kept consistent with the original order of the chapters. If there are the same matching values, they are arranged according to the paragraph numbers. After integration, the complete retrieval path link is obtained. Then, these paths are recorded in the dispatch table one by one. The dispatch table contains three types of information, such as section number, document position, and positioning matching value. For example, it can be recorded as D1-P12.1, D2-P15.1, and D3-P27.2. This tabular text structure can be regarded as a version paragraph retrieval dispatch table, which is used for subsequent quick positioning of each paragraph.

[0107] Please refer to Figure 5 The specific steps of S4 are as follows:

[0108] S401: Call the section number and the document position in the version paragraph retrieval dispatch table, extract the associated phrase structure and the occurrence order of the target word sense item in the paragraph, compare the numbering and position relationship between adjacent phrases, and generate a phrase order annotation quantity;

[0109] Call the section number and the document position in the version paragraph retrieval dispatch table, read the dispatch table entries one by one, take the section number and the corresponding document position as the basic reference items, extract the associated phrase structure of the target word sense item in the paragraph in sequence, and record them according to the actual occurrence order. The phrase structure can be separated sentence by sentence in combination with the natural semantic flow in the paragraph when it is extracted. For example, the word sense item X appears in the document position P12 in the paragraph D1, forming the phrases "operation unit X1" and "operation unit X2", which are annotated as Z1 and Z2 in sequence. When there are multiple phrases in the paragraph, the numbering needs to be continued to form continuous marks such as Z1, Z2, Z3, etc. Then, the numbering and position relationship between adjacent phrases are compared. The comparison method is based on the line spacing difference in the document position. The distance between each pair of phrases is recorded as a numerical range. For example, when the line spacing between adjacent phrases is less than 10 lines, it is annotated as a close relationship, when the line spacing is between 10 and 20 lines, it is annotated as an intermediate relationship, and when the line spacing is greater than 20 lines, it is annotated as a far relationship. After all the phrase relationships are summarized, the phrase order annotation quantity is formed. This annotation quantity not only contains the phrase numbering and the order, but also includes the qualitative description of the position difference, such as close, intermediate, or far. Each record can be regarded as a sequence relationship of a group of adjacent phrases, and a complete phrase order annotation set is obtained, which is used for subsequent structure comparison and order difference judgment.

[0110] S402: Based on the phrase order annotation quantity, the offset degree of the structure order in the difference version is calculated. The numbering difference value of the paragraph is compared with the set order offset standard value one by one, the paragraph set that exceeds the standard is obtained, and the order offset difference value is generated.

[0111] Based on the phrase order annotation quantity, the structural order deviation between different versions needs to be calculated. First, the phrase order annotations of the two versions are expanded piece by piece, and the distance difference between the same phrase pairs is compared one by one. For example, the distance annotation of Z1 and Z2 in version V1 is adjacent relationship, and the distance annotation of Z1 and Z2 in version V2 is intermediate relationship. It is determined that there is an order deviation between the phrase pair. Then the difference is converted into a deviation value. The deviation value can be set according to the difference level. For example, the difference between adjacent and intermediate is 2, the difference between intermediate and far is 3, and the difference between adjacent and far is 5. The obtained value is compared with the pre-set order deviation standard value one by one. The standard value can be set to 5. When the difference value is less than or equal to 5, it is determined to be within the standard range. When the difference value is greater than 5, it is determined to be out of range. The paragraph to which the phrase pair belongs is included in the deviation paragraph set. For example, in paragraph D2, the difference value of Z2 and Z3 is 7, so the paragraph is recorded as a deviation paragraph, and the order deviation difference value 7 is generated and archived. After comparison, all paragraphs form a set, which contains all paragraphs that exceed the standard and the corresponding difference value, providing a basis for subsequent order adjustment.

[0112] S403: According to the order deviation difference value, the paragraphs exceeding the standard are adjusted in position. The paragraph order is rearranged by calling the number and literature position. The positions of all adjusted paragraphs are mapped to the corresponding word sense item combination to generate the word sense item mapping paragraph alignment result.

[0113] According to the order deviation difference value, the paragraphs in the set need to be adjusted in position. First, read the set entries one by one, and analyze the paragraph number and literature position combined difference value. When the difference value is between 1 and 5, the original position can be kept unchanged. When the difference value is between 6 and 10, the paragraph needs to be moved one position forward or backward. When the difference value is between 11 and 20, the paragraph needs to be moved two positions. If the value exceeds 20, the moving range will be further increased. In this way, the order is adjusted paragraph by paragraph. The adjusted paragraphs are rearranged into a new sequence, and the logical connection relationship between the paragraph numbers is not disturbed during the rearrangement. Then, the new paragraph position is mapped to the original word sense item combination. For example, paragraph D3 is originally located at P27, but it is moved to P26 due to a difference value of 8, and it is re-established with the combination of word sense items X and Y. It is recorded as D3-P26-(X, Y). Then, other paragraphs are processed in turn to generate a new word sense item mapping paragraph alignment result. This result contains paragraph numbers, updated position numbers and word sense item combinations, which completely covers all adjusted paragraphs and realizes the unified structure after order alignment.

[0114] Please refer to Figure 6 , the specific steps of S5 are:

[0115] S501: Based on the paragraph alignment result of the word sense item mapping, the word sense number corresponding to the keyword and the position index in the paragraph are annotated segment by segment, the occurrence frequency and position difference value of the keyword in the paragraph are called, the combination of the number and the position index is operated, and the word sense position annotation quantity is generated;

[0116] Based on the paragraph alignment result of the word sense item mapping, the word sense number corresponding to the keyword and the position index in the paragraph need to be annotated segment by segment. First, the segment numbers contained in the alignment result are read one by one, and the keywords under each segment number are assigned numbers in accordance with the order of appearance. For example, the paragraph D1 contains the keywords "search", "compare" and "sort", which are annotated as W1, W2 and W3 respectively. Then, the position index of each keyword in the text is determined according to the order of the words in the paragraph. Assuming that "search" is located at the 8th word position, "compare" is located at the 15th word position, and "sort" is located at the 21st word position, the indices are P8, P15 and P21 respectively. Then, the occurrence frequency of the keyword is counted. When the keyword appears multiple times, the specific position of each occurrence needs to be recorded. For example, "search" appears twice in the paragraph, with indices P8 and P24 respectively. The index record of this number is {P8, P24}. At the same time, the difference value between the indices needs to be calculated. If P24 minus P8 is 16, then the difference value 16 corresponding to W1 is recorded and bound with the number. W2 and W3 appear only once, so the difference value is 0. When the paragraph contains multiple word sense numbers, the collection structure of the number, index and difference value needs to be formed. For example, {W1: P8, P24-difference value 16} {W2: P15-difference value 0} {W3: P21-difference value 0}. This collection is the word sense position annotation quantity, which can record the keyword number, occurrence position and relative difference value completely. The same processing needs to be performed in different paragraphs, so that each paragraph has the corresponding word sense position annotation quantity.

[0117] S502: According to the word sense position annotation quantity, the annotated word sense number in the paragraph is sorted, the position index of the keyword corresponding to the number in the reading area is matched with the paragraph structure, the sorting difference value is counted and compared with the set display reference value, and the reading area sequence difference value is obtained;

[0118] According to the word sense position marking quantity, the word sense numbers marked in the paragraph need to be sorted. First, all numbers are arranged in ascending order according to the position index, for example, in paragraph D2, the position index of W1 is P6, the position index of W2 is P14, and the position index of W3 is P19. The sorting order is W1, W2, and W3 in turn. Then, the number order is compared with the natural language order of the paragraph. If the number order is completely consistent with the actual order, the difference value is 0. If the order is different, the difference value needs to be calculated. The difference value is measured by the number of displacements between the position indexes. For example, W1 is in the first position in the sorting order, but in the second position in the natural language order of the paragraph. The difference value is recorded as 1. After comparing all the numbers, a difference value set in the paragraph is formed, for example, {W1:1, W2:0, W3:0}. Then, the set is compared with the display reference value. The display reference value can be set as 3. When the difference value is less than or equal to 3, it is recorded as within the standard range. When the difference value is greater than 3, it is recorded as out of range. For example, in paragraph D2, the difference value of W1 is 1, the difference value of W2 is 0, and the difference value of W3 is 4. It is determined that W3 is out of range. A reading area sequence difference value set is formed. The set clearly marks the difference value corresponding to each number and compares the reference value, so as to obtain the deviation items existing in each paragraph.

[0119] S503: Call the reading area sequence difference value. The paragraphs with difference values exceeding the display reference value are reorganized. The word sense explanation, literature source, and path information are output in paragraph order. The content is combined into a prompt block to generate a word sense comparison prompt block.

[0120] The difference value of the reading area sequence is called, and the paragraphs corresponding to the difference values exceeding the display reference value need to be reorganized. First, the abnormal number and the difference value are extracted, for example, the difference value set of paragraph D3 is {W1:0, W2:5, W3:2}, and the difference value of W2 is 5, which is greater than the reference value 3, so it is determined that D3 needs to be adjusted. The adjustment method is to rearrange the position of the number in the paragraph according to the difference value. The numbers with a difference value between 1 and 3 remain in their original positions, the numbers with a difference value between 4 and 6 are moved forward or backward by 1 index interval, the numbers with a difference value between 7 and 10 are moved by 2 index intervals, and if the difference value exceeds 10, the moving range is increased according to the difference value. For example, W2 has a difference value of 5, so it needs to be moved by 1 index interval from the original position P20 to P19. After completing the position correction, the number is combined with the word meaning explanation and the literature source, for example, W2 explanation is "search condition", literature source is a certain file page 15, paragraph 2, path information is "path / section / 15 / 2", then record W2-explanation-source-path. Then all the numbers are combined in turn to form a prompt group block. The structure of the prompt group block is the combination of all word meaning numbers corresponding to the paragraph number and their explanation, source, path. For example, the prompt group block of paragraph D3 is {W1-explanation-source-path, W2-explanation-source-path, W3-explanation-source-path}. The prompt group block set of all paragraphs constitutes the online reading word meaning comparison prompt group block, which is arranged in sequence and retains complete explanation information.

[0121] Please refer to Figure 7 Based on the same inventive concept, a Chinese literature database online query reading system is also proposed for implementing the above-mentioned Chinese literature database online query reading method, comprising:

[0122] The word meaning candidate module obtains the frequency of occurrence, the length of the explanation text, the number of boundary annotations, and the literature time sequence weight of the search word item in the core ancient literature, and sets a screening threshold to screen the word meaning items, and generates a semantic candidate set;

[0123] The semantic path module extracts the subject pronoun and verb collocation in the corresponding paragraph based on the word meaning items in the semantic candidate set, calculates the matching strength of the semantic subject and the action verb and sorts them, and generates a context semantic mapping path sequence;

[0124] The version positioning module locates the paragraph number and the literature position in the version index according to the priority path in the context semantic mapping path sequence, combines the word meaning items and the chapter information, records the coordinates, and generates a version paragraph retrieval scheduling table;

[0125] The paragraph alignment module calls the version paragraph retrieval scheduling table, extracts the phrase structure and position sequence of the target word meaning item, calculates the sequence offset between different versions, and adjusts the paragraphs exceeding the threshold, and generates a word meaning item mapping paragraph alignment result;

[0126] The reading prompt module aligns the paragraph based on the word sense item mapping, marks the key word sense number and position in the reading area, organizes the display of the literature source and path information, and generates a word sense comparison prompt group.

[0127] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An online query reading method for Chinese literature database, characterized in that, The method comprises the following steps: S1: obtaining the word meaning item of the search term, extracting the core ancient literature citation frequency, the explanation text length and the number of boundary annotations of the meaning item, integrating the literature time sequence weight, screening the meaning item meeting the conditions, and generating a semantic candidate set; The specific steps of S1 are: S101: obtaining the word meaning item corresponding to the search term in the Chinese language and literature database, collecting the citation frequency, explanation text length and boundary annotation number of the meaning item in the core ancient literature, arranging the data into a unified format, and generating literature word meaning parameters; S102: based on the literature word meaning parameters, calling the literature time sequence weight to process the citation frequency of the meaning item in the ancient literature of different ages, and comparing the explanation text length and the number of boundary annotations to generate a word meaning index; S103: according to the word meaning index, screening the word meaning item, retaining the word meaning item set meeting the conditions, and arranging the word meaning items in the set according to the predetermined sorting standard to generate a semantic candidate set; S2: based on the word meaning item in the semantic candidate set, extracting the subject pronoun and verb collocation structure, analyzing the semantic subject and action verb associated with the explanation, and sorting the path to generate a context semantic mapping path sequence; The specific steps of S2 are: S201: obtaining the word meaning item in the semantic candidate set, extracting the subject pronoun and verb collocation structure in the paragraph where the word meaning item is located, arranging the subject pronoun and verb into a combination record, and generating a paragraph collocation combination amount; S202: based on the paragraph collocation combination amount, calling the semantic subject and action verb in the word meaning explanation, comparing the subject pronoun and verb collocation structure in the paragraph, judging the corresponding relationship and recording the strength result, and obtaining a semantic correlation strength value; S203: according to the semantic correlation strength value, sorting the path corresponding to the word meaning item, merging the path results according to the paragraph position, and generating a context semantic mapping path sequence; S3: according to the word meaning item and the chapter information in the preferred path in the context semantic mapping path sequence, locating the corresponding paragraph in the version index, marking the paragraph number and literature position of the target word meaning item, constructing a retrieval structure, and generating a version paragraph retrieval scheduling table; S4: calling the content of the version paragraph retrieval scheduling table, extracting the target word meaning item associated phrase structure and order, comparing the structure order offset between versions, adjusting the paragraph position exceeding the set standard, and generating a word meaning item mapping paragraph alignment result; S5: based on the word meaning item mapping paragraph alignment result, marking the word meaning number and position of the key word in the paragraph, organizing a reading area display structure, outputting the word meaning explanation, literature source and semantic path information, and generating a word meaning comparison prompt block.

2. The method of claim 1, wherein the Chinese literature database is queried on-line. The semantic candidate set includes a candidate word sense item list, a word sense feature parameter table, an entry threshold value identifier, and a score summary. The context semantic mapping path sequence includes a path sequence number set, a relationship strength level set, a subject-verb collocation snapshot set, and a paraphrase subject mapping label set. The version paragraph retrieval scheduling table includes a version identification code, a section number list, a document location coordinate, and a retrieval routing item. The word sense item mapping paragraph alignment result includes an associated phrase structure table, a position sequence, a structure sequence offset result, and a position adjustment record. The word sense comparison prompt block includes a word sense number annotation area, a keyword position index, a document source item, a path prompt information card, and a reading area display layout.

3. The online query and reading method for the Chinese language and literature database according to claim 1, characterized in that, The specific steps of S3 are: S301: Obtain a priority path in the context semantic mapping path sequence, extract the word sense item and the chapter information under the path, detect the paragraph location corresponding to the word sense item in the version index, record the paragraph number and the index label, and generate a paragraph index number amount; S302: Based on the paragraph index number amount, call the section number and the document location containing the target word sense item, judge the corresponding relationship between the section number and the location data, filter the combination structure that can uniquely identify the paragraph, and obtain a paragraph positioning matching value; S303: According to the paragraph positioning matching value, establish the retrieval structure of the section number and the document location in the version index, merge all combination relationships into a retrieval path in turn, and generate a version paragraph retrieval scheduling table.

4. The online query and reading method for the Chinese language and literature database according to claim 3, characterized in that, The specific steps of S4 are: S401: Call the section number and the document location in the version paragraph retrieval scheduling table, extract the associated phrase structure and the appearance order of the target word sense item in the paragraph, compare the number and location relationship between adjacent phrases, and generate a phrase order annotation amount; S402: Based on the phrase order annotation amount, calculate the offset degree of the structure order in the difference version, compare the number difference value of the paragraph with the set order offset standard value one by one, obtain the paragraph set that exceeds the standard, and generate an order offset difference value; S403: According to the order offset difference value, adjust the position of the paragraph that exceeds the standard, call the number and the document location to rearrange the paragraph order, map all adjusted paragraph positions to the corresponding word sense item combination, and generate a word sense item mapping paragraph alignment result.

5. The method of claim 4, wherein the Chinese literature database is queried on-line. The specific steps of S5 are: S501: Based on the word sense item mapping paragraph alignment result, annotate the word sense number and the position index of the keyword in the paragraph one by one, call the appearance number and the position difference value of the keyword in the paragraph, operate on the combination of the number and the position index, and generate a word sense position annotation amount; S502: According to the word sense position annotation amount, sort the word sense number annotated in the paragraph, match the position index of the keyword in the reading area with the paragraph structure corresponding to the number, count the sorting difference value and compare it with the set display reference value, and obtain a reading area sequence difference value; S503: Call the reading area sequence difference value, reorganize the paragraphs corresponding to the difference value that exceeds the display reference value, output the word sense paraphrase, the document source, and the path information according to the paragraph order, combine the contents into a prompt block, and generate a word sense comparison prompt block.

6. The method for online query and reading of Chinese literature database according to claim 1, characterized in that, The core ancient literature refers to the representative historical literature in the field of Chinese language and literature, including Shangshu, Zuo Zhuan, and the classic literature of the Han Dynasty; The literature time weight refers to the weighted value given to the word meaning item in different literature according to the age, school, and influence on language evolution of the literature; The semantic candidate set refers to the set of all candidate word meaning items filtered by multi-dimensional parameters; The subject pronoun refers to the pronoun representing the subject in the sentence; The verb collocation structure refers to the grammatical structure of a verb and its associated nouns, adjectives, or other modifiers; The context semantic mapping path sequence refers to the path set generated by sorting according to the semantic relationship in the context.

7. The method of claim 1, wherein the Chinese literature database is queried on-line. [0077] The version index includes different literature versions, paragraph positions, and chapter structures; The paragraph number and literature position refer to the identifier in the literature; The version paragraph retrieval scheduling table refers to the retrieval order and retrieval structure generated based on the different version literature and the target paragraph in the query process; The word meaning item associated phrase structure refers to the phrase, word group, or sentence structure closely associated with the target word meaning item in the literature paragraph, which is consistent with the semantics of the word meaning item; The structure order offset refers to the order difference of phrases or word groups in sentences in different versions of literature. By calculating the structure order difference between different versions, the order is sorted and adjusted; The word meaning item mapping paragraph alignment result refers to the paragraph alignment result generated by adjusting the structure order between different versions; The word meaning comparison prompt block refers to the word meaning comparison framework displayed to the user during reading, including word meaning interpretation, literature source, and path information.

8. An online query reading system for Chinese literature database, which is used to realize the online query reading method for Chinese literature database according to any one of claims 1-7, characterized in that, It includes: The word meaning candidate module obtains the word meaning item frequency of the search word in the core ancient literature, the interpretation text length, the number of boundary annotations, and the literature time weight, and sets the screening threshold to screen the word meaning items, generating a semantic candidate set; The semantic path module extracts the subject pronoun and verb collocation based on the word meaning items in the semantic candidate set in the corresponding paragraph, calculates the matching strength of the semantic subject and the action verb and sorts them, and generates a context semantic mapping path sequence; The version positioning module locates the paragraph number and literature position in the version index according to the priority path in the context semantic mapping path sequence, combines the word meaning item and chapter information, records the coordinates, and generates a version paragraph retrieval scheduling table; The paragraph alignment module calls the version paragraph retrieval scheduling table, extracts the phrase structure and position order of the target word meaning item, calculates the order offset between different versions, and adjusts the paragraphs that exceed the threshold, generating a word meaning item mapping paragraph alignment result; The reading prompt module marks the key word meaning number and position based on the word meaning item mapping paragraph alignment result in the reading area, organizes the display of literature source and path information, and generates a word meaning comparison prompt block.

Images