CN108563773B - Knowledge graph-based legal provision accurate search ordering method - Google Patents
Knowledge graph-based legal provision accurate search ordering method Download PDFInfo
- Publication number
- CN108563773B CN108563773B CN201810361909.8A CN201810361909A CN108563773B CN 108563773 B CN108563773 B CN 108563773B CN 201810361909 A CN201810361909 A CN 201810361909A CN 108563773 B CN108563773 B CN 108563773B
- Authority
- CN
- China
- Prior art keywords
- subject
- word
- legal
- weight
- combinations
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The invention discloses a knowledge graph-based legal provision accurate searching and sorting method, which comprises the following steps of: s101, inputting keywords to be searched; s102, obtaining a subject word matched with the keyword to be searched and a related word of the subject word from a legal knowledge graph model to form a subject word bank; s103, dynamically generating subject words and weight values of relevant words in a subject word bank of the same case according to the incidence relation with the core subject in the legal knowledge graph model; s104, obtaining subset combinations of the subject word stock, and sequentially forming the subset combinations with m, m-1 and m-2 … 2 elements; s105, respectively taking out a subject word from each subset combination to form a subject word combination finally used for searching; s106, sequencing all subset combinations by taking the number of the topic word combinations finally used for searching and the total weight of the keywords as the basis; and S107, searching in the legal provision database by using the selected subject word combinations after sequencing, and sequencing and displaying the search results.
Description
Technical Field
The invention relates to a legal provision searching and sequencing method, in particular to a method for accurately retrieving legal provisions and preferentially sequencing search results based on a knowledge graph.
Background
Strengthening the construction of China with law enforcement, it is particularly important to advance the basic strategy of controlling the state with law, and the administration with law is an important link of controlling the state with law. In each case, all cases must be judged according to law, and legal provisions related to the cases to be handled often need to be retrieved by judicial personnel except depending on business knowledge and experience of the judicial personnel, so that the legal provisions are accurately and efficiently retrieved, and the urgent needs of other legal provision searching personnel of judicial staff are met.
With the popularization and efficient operation of artificial intelligence, knowledge maps are introduced into the field of legal provision retrieval as a basic knowledge representation method of artificial intelligence. The Knowledge Graph (Knowledge Graph) describes concepts, entities, events and relations among the concepts in the objective world, and the Knowledge Graph is used for describing legal provisions, so that the legal provisions of all categories are connected in series, the concept relations are clearer, and the retrieval of the legal provisions on the basis is more accurate and efficient.
Because the existing legal provision retrieval ordering method is realized on the basis of keyword matching degree, if the legal concepts which are not legal provisions are used as keywords for retrieval, judicial personnel can not obtain ideal legal provision search ordering results in work, or can search available legal provision results only by repeatedly changing the keywords and repeatedly adjusting the combination mode of the keywords.
Disclosure of Invention
The invention aims to solve the technical problem that in the prior art, an ideal legal provision retrieval result cannot be obtained by retrieving through a concept other than a legal concept in the legal provision retrieval, and provides a method for accurately searching and sequencing the legal provision based on a knowledge graph.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the method for accurately searching and sequencing legal provision based on the knowledge graph comprises the following steps:
s101, inputting keywords to be searched;
s102, obtaining a subject word matched with the keyword to be searched and a related word of the subject word from a legal knowledge graph model to form a subject word bank; the associated words include: upper themes, lower themes and label synonyms; the upper theme refers to a theme with a large representative range, the lower theme refers to a theme with a small representative range, and the label synonym is a theme with a similar representative range defined in the label; the subject term attributes include: the name and weight of the subject term are (0, 1);
s103, dynamically generating subject words and weight values of relevant words in a subject word bank of the same case according to the incidence relation with the core subject in the legal knowledge graph model, wherein the higher the weight value is, the closer the scope represented by the core subject is to the weight value;
s104, obtaining subset combinations of the subject word stock, and sequentially forming the subset combinations with the number of m, m-1 and m-2 … 2;
s105, respectively taking out a subject word from each subset combination to form a subject word combination finally used for searching;
s106, sequencing all subset combinations by taking the number of the topic word combinations finally used for searching and the total weight of the keywords as the basis;
and S107, searching in the legal provision database by using the selected subject word combinations after sequencing, and sequencing and displaying the search results.
According to the technical scheme, the weight of the associated word is adjusted within the weight range according to the search result in the legal provision database.
In step S107, according to the above technical solution, during the search, the corresponding position of each subject word in the subject word combination in the legal provision document and the number of times of occurrence of the subject word are recorded, the legal provisions are sorted according to the number of times of occurrence of the subject word in the legal provision document, and the legal provision with the largest number of occurrences is arranged at the top of the search result.
According to the technical scheme, the legal knowledge graph model is generated by analyzing and converting files of the constructed legal knowledge graph, the legal knowledge graph model organizes the same case by subject words of related cases according to three main aspects, each aspect is classified, the related subjects are analyzed in the representative range in each classification, and the upper and lower positions or the same-position relation among the subjects is defined; three main aspects include subject, rights obligation and object, legal facts.
According to the technical scheme, the weight of the matched subject word is 1.0, and the weight of the label synonym of the upper subject, the lower subject and the upper and lower subjects is 0.5.
According to the technical scheme, the weight of the matched subject word is 1.0, and the weight of the label synonym of the upper subject of the nth layer, the lower subject of the nth layer and the upper subject of the nth layer is 0.5nWherein n is a natural number.
The invention also provides a system for accurately searching and sequencing legal provisions based on the knowledge graph, which comprises the following steps:
the input module is used for inputting keywords to be searched;
the system comprises a topic word bank generating module, a search module and a search module, wherein the topic word bank generating module is used for acquiring a topic word matched with a keyword to be searched and a relevant word of the topic word from a legal knowledge graph model to form a topic word bank; the associated words include: upper themes, lower themes and label synonyms; the upper theme refers to a theme with a large representative range, the lower theme refers to a theme with a small representative range, and the label synonym is a theme with a similar representative range defined in the label; the subject term attributes include: the name and weight of the subject term are (0, 1);
the weight generation module is used for dynamically generating the weight of the subject words and the associated words in the subject word bank of the same case according to the association relation with the core subject in the legal knowledge graph model, wherein the higher the weight is, the closer the scope represented by the core subject is to the weight;
the subset combination generating module is used for acquiring subset combinations of the subject thesaurus and sequentially forming the subset combinations with the number of m, m-1 and m-2 … 2; respectively taking out a subject word from each subset combination to form a subject word combination finally used for searching; sorting all subset combinations according to the number of the topic word combinations finally used for searching and the total weight of the keywords;
and the search result display module is used for searching in the legal provision database by using the selected subject word combinations after the ordering, and ordering and displaying the search results.
The invention also provides a computer readable storage medium, which comprises a computer program executable by a processor, wherein the computer program specifically executes the method for accurately searching and sorting legal provisions based on the knowledge graph.
The invention has the following beneficial effects: the method adopts the legal knowledge map to describe the logical relationship among legal subject words and the logical relationship among the subject words and other expression words, obtains a subject word bank of keywords to be searched through the analysis and the analysis of the legal knowledge map, sets the weight of the subject words and associated words in the subject word bank, and orders the search results by taking the height of the combined weight as the basis of priority search.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flowchart of a method for accurately searching and ranking legal provisions based on a knowledge graph according to a first embodiment of the present invention;
FIG. 2 is a flowchart of a method for accurately searching and ranking legal provisions based on a knowledge-graph according to a second embodiment of the present invention;
fig. 3 is a legal intellectual map relating to property rights.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1, the method for accurately searching and sorting legal provision based on knowledge graph in the first embodiment of the present invention mainly includes the following steps:
s101, inputting keywords to be searched;
s102, obtaining a subject word matched with the keyword to be searched and a related word of the subject word from a legal knowledge graph model to form a subject word bank; the associated words include: upper themes, lower themes and label synonyms; the upper theme refers to a theme with a large representative range, the lower theme refers to a theme with a small representative range, and the label synonym is a theme with a similar representative range defined in the label; the subject term attributes include: the name and weight of the subject term are (0, 1);
s103, dynamically generating subject words and weight values of relevant words in a subject word bank of the same case according to the incidence relation with the core subject in the legal knowledge graph model, wherein the higher the weight value is, the closer the scope represented by the core subject is to the weight value;
s104, obtaining subset combinations of the subject word stock, and sequentially forming the subset combinations with the number of m, m-1 and m-2 … 2;
s105, respectively taking out a subject word from each subset combination to form a subject word combination finally used for searching;
s106, sequencing all subset combinations by taking the number of the topic word combinations finally used for searching and the total weight of the keywords as the basis;
and S107, searching in the legal provision database by using the selected subject word combinations after sequencing, and sequencing and displaying the search results.
Further, according to the search result in the legal provision database, the weight of the associated word can be adjusted within the weight range according to the actual situation.
In step S107, during searching, the corresponding position of each subject term in the subject term combination in the legal provision document and the number of times that the subject term appears are recorded, and the legal provisions are ranked according to the number of times that the subject term appears in the legal provision document, and the legal provision with the largest number of times appears is ranked in the top of the search result.
The legal knowledge map model is generated by analyzing and converting files of the constructed legal knowledge map, the legal knowledge map model organizes the same case by subject words of related cases according to three main aspects, classifies each aspect, analyzes the representing range of the related subject in each classification, and defines the upper and lower positions or the same-position relationship among the subjects; the three main aspects include the subject, the rights obligation object, and the legal facts.
The weight of the matched subject word is 1.0, the weight of the label synonyms of the upper subject, the lower subject and the upper and lower subjects is 0.5-1, the number of layers is sequentially increased, and the weight is changed.
In order to realize the method, the invention also provides a system for accurately searching and sequencing legal provisions based on the knowledge graph, which comprises the following steps:
the input module is used for inputting keywords to be searched;
the system comprises a topic word bank generating module, a search module and a search module, wherein the topic word bank generating module is used for acquiring a topic word matched with a keyword to be searched and a relevant word of the topic word from a legal knowledge graph model to form a topic word bank; the associated words include: upper themes, lower themes and label synonyms; the upper theme refers to a theme with a large representative range, the lower theme refers to a theme with a small representative range, and the label synonym is a theme with a similar representative range defined in the label; the subject term attributes include: the name and weight of the subject term are (0, 1);
the weight generation module is used for dynamically generating the weight of the subject words and the associated words in the subject word bank of the same case according to the association relation with the core subject in the legal knowledge graph model, wherein the higher the weight is, the closer the scope represented by the core subject is to the weight;
the subset combination generating module is used for acquiring subset combinations of the subject thesaurus and sequentially forming the subset combinations with the number of m, m-1 and m-2 … 2; respectively taking out a subject word from each subset combination to form a subject word combination finally used for searching; sorting all subset combinations according to the number of the topic word combinations finally used for searching and the total weight of the keywords;
and the search result display module is used for searching in the legal provision database by using the selected subject word combinations after the ordering, and ordering and displaying the search results.
The method of the present invention may also be stored in a computer readable storage medium comprising a computer program executable by a processor, the computer program specifically performing the method for accurate search ranking of knowledge-graph based legal provision of the above embodiments.
As shown in fig. 2, the second method for searching and ranking legal provision based on knowledge-graph in the embodiment of the present invention includes the following steps:
s01, the domain expert constructs a legal knowledge map in a MindManager tool;
s02, generating a legal knowledge map model KGModel (hereinafter, the legal knowledge map model is abbreviated as KGModel) by analyzing and converting a legal knowledge map file, wherein the KGModel organizes the same case by subject words of related cases according to three main aspects (including subjects, rights obligations and objects, legal facts), classifies the same case in each aspect (for example, the subjects are further divided into natural persons, individual industrial and commercial persons, legal facts and the like), analyzes the representative range of the related subjects in each classification, and defines the upper-lower position or same-position relation among the subjects;
s03, importing a legal provision database, wherein the database contains legal provision documents;
s04, inputting a keyword combination to be searched, wherein the keyword is a keyword combination which is generally proposed in each of three aspects (main body, right obligation and legal fact);
s05, obtaining a subject word matched with the keyword to be searched and a related word of the subject word from the KGModel (as shown in FIG. 3, the KGModel comprises an upper subject SubTopic, a lower subject SubTopic and a Label synonym Label) (wherein the attribute of the subject word comprises a subject word name and a subject word weight), and forming a subject word library; (weight description: weight range is (0,1)]. The weight of the matched subject word is 1.0, the weight of the label synonyms of the upper subject (lower subject) and the upper subject (lower subject) in the 1 layer is 0.5, and the weight of the upper subject (lower subject) and the corresponding label synonyms of the upper subject (lower subject) in the n layers is 0.5n(ii) a For convenience of explanation, the number of the subject layers in the embodiment is 3, and the number of the subject layers can be set as required)
S06, aiming at the fact that the weight of the subject word and the related word in the subject word library of the same case is based on the actual semantic association condition, the weight can be properly adjusted in the weight range;
s07, obtaining subset combinations of the topic lexicon, for example: if the number of the subject words in the subject word library is m, subset combinations with the number of elements of m, m-1 and m-2 … 2 are formed in sequence;
s08, extracting an element (subject word) from each of the n subset combinations to form a subject word combination finally used for searching;
s09, sorting all the combinations according to the number (priority) of the topic word combination finally used for searching and the total weight (secondary) of the keywords in the topic word combination;
and S10, searching in the legal provision database by using the combination-ordered subject word combination and ordering and displaying the search result. During searching, corresponding positions of all subject words in the subject word combination in the legal provision document (when the legal provision database is imported, all legal subject words in the legal knowledge map can be searched in the legal provision database document once, positions of the subject words appearing in the legal provision document are recorded and stored in an XML document) are intersected, the legal provision is sequenced according to the times of the subject words appearing in the document, and the legal provision with the largest occurrence times is ranked in the front of a search result.
To further illustrate the above approach, the following is described in conjunction with specific legal application scenarios:
first, scene description
Case (one):
the A real estate company and the B construction company sign construction engineering contract to ensure that the B company establishes a commodity house for the A company. After the contract is signed, the company A signs a loan contract with the bank C for the project construction fund collection, the company A loans 3000 ten thousand yuan to the bank C, and meanwhile, the company A uses the use right of the developed residential construction land as a mortgage to guarantee the debt. Then, since company A cannot repay the loan of bank C, bank C wants to exercise the mortgage right on the residential construction land and the developed commercial housing of company A.
(II) the judicial personnel need to inquire the problems:
can bank C exercise mortgage over the product room developed by company a?
(iii) keywords for search (input keywords):
commodity house and mortgage right exercising "
Second, the concrete steps
Step 1: importing legal knowledge graph (built by domain experts in MindManager tool) needed to be used in the embodiment (as shown in FIG. 3);
step 2: automatically constructing a model for the legal knowledge graph by analyzing and converting the legal knowledge graph file to form a subject word bank;
and step 3: importing a legal provision database, wherein the database is provided with legal provision documents;
and 4, step 4: keyword for search in input embodiment: "Commodity housing, exercise the mortgage right";
and 5: automatically calculating the similarity of the subject term in the subject term library formed by the term of the commodity room and the KGModel to obtain the subject term matched with the term of the commodity room and the relevant term of the subject term; similar to "exercise the mortgage right", the matching is automatically performed in turn. The matching result is shown in a table A-a keyword matching result table;
TABLE A-keyword match results Table
Step 6: adjusting the weight in the keyword matching result according to the actual semantic association condition of the case, wherein the weight is not changed; for example: the weight of the property is 0.5, if the searcher thinks that the word is not greatly related to the case situation in the case, the weight can be changed into any value of the interval (0,0.5), and then the query is carried out;
and 7: combining the keyword matching result and the modified weight (when the keyword is not modified, the matched original weight is used); if the keyword does not match the same subject term, such as the word "exercise the mortgage right", the completely matched subject term can not be found during matching, and the weight of the word "exercise the mortgage right" input by the user is given as 0.9 during combination and is inserted into the subject term bank;
and 8: acquiring subset combination of the subject word library: when the number of the subject words is k (k >1), sequentially forming set combinations with elements of k, k-1, …, 2 and (2^ k) -k; when k is 1, 1 set combination is obtained, namely the matched subject word set itself; when k is not 1, an element (subject word) is taken from each of the (2^ k) -k sets to form a subject word set finally used for searching. In the above 2-tuple set combination, assuming that a binary set combination is (a, B), where the elements in the set a are { a1, a2}, and the elements in the set B are { B1, B2}, the topic words that will be formed are [ a1B1], [ a1B2], [ a2B1], [ a2B2 ];
and step 9: the number of the subject word combinations finally used for searching is k, the total weight of the keywords in the subject word combinations changes according to a legal database, and all phrase combinations are sequenced by taking the number (priority) and the total weight (secondary) of the subject word combinations as the basis during sequencing;
step 10: a search is conducted in a legal provision database using the combination-ordered topic word combinations. During searching, the corresponding position of each subject word in a subject word combination formed by the words to be searched (the 'commodity room' and the 'mortgage' is exercised) input by a user in the legal provision document (when the legal provision database is imported, all legal subject words in the legal knowledge map can be searched once in the legal provision database document, the positions of the subject words appearing in the legal provision document are recorded and stored in the XML document) is intersected, the legal provision is sequenced according to the times of the subject words appearing in the document, and the legal provision with the largest occurrence times is sequenced at the top of the search result and presented.
In conclusion, the legal knowledge graph model generated by analysis and conversion of the legal knowledge graph can be used for acquiring the corresponding subject words of the keywords to be searched in the knowledge graph and acquiring the associated words of the subject words, namely, the upper subject, the lower subject and the label synonym for retrieval, and the legal entry search results can be obtained from existence to existence and from excellence according to the search word combination formed by the subject words and the associated words, namely, the coverage range of the search results is more comprehensive and more accurate; the invention can also adjust the weight of the subject term according to the actual situation of the case in the same case, and sequence and push the search results according to the combination of the weight of the subject term, thereby helping judicial personnel and other legal provision search personnel to quickly and accurately obtain all legal provisions related to the case and enhancing the practicability of the legal provision search results.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.
Claims (8)
1. A method for accurately searching and sequencing legal provision based on a knowledge graph is characterized by comprising the following steps:
s101, inputting keywords to be searched;
s102, obtaining a subject word matched with the keyword to be searched and a related word of the subject word from a legal knowledge graph model to form a subject word bank; the associated words include: upper themes, lower themes and label synonyms; the upper theme refers to a theme with a large representative range, the lower theme refers to a theme with a small representative range, and the label synonym is a theme with a similar representative range defined in the label; the subject term attributes include: the name and weight of the subject term are (0-1);
s103, dynamically generating subject words and weight values of relevant words in a subject word bank of the same case according to the incidence relation with the core subject in the legal knowledge graph model, wherein the higher the weight value is, the closer the scope represented by the core subject is to the weight value;
s104, obtaining subset combinations of the subject word stock, and sequentially forming the subset combinations with the number of m, m-1 and m-2 … 2;
s105, respectively taking out a subject word from each subset combination to form a subject word combination finally used for searching;
s106, sequencing all subset combinations by taking the number of the topic word combinations finally used for searching and the total weight of the keywords as the basis;
and S107, searching in the legal provision database by using the selected subject word combinations after sequencing, and sequencing and displaying the search results.
2. The knowledge-graph-based legal provision precise search ranking method according to claim 1, characterized in that the weight of the associated words is adjusted within the weight range according to the actual situation according to the search results in the legal provision database.
3. The method as claimed in claim 1, wherein in step S107, the corresponding position of each subject word in the subject word group in the legal provision document and the number of times of occurrence of the subject word are recorded, and the legal provisions are ranked according to the number of times of occurrence of the subject words in the legal provision document, and the legal provision with the largest number of occurrences is presented at the top of the search result.
4. The knowledge-graph-based legal provision precise search ranking method according to claim 1, characterized in that the legal knowledge graph model is generated by parsing and converting the files of the built legal knowledge graph, the legal knowledge graph model organizes the same case by subject words of related cases according to three main aspects, classifies each aspect, analyzes the representing range of the related subject in each classification, and defines the upper, lower or same position relation between the subjects; the three main aspects include subject matter, rights obligations, legal facts.
5. The method of claim 1, wherein the weight of the matched subject term is 1.0, and the weight of the label synonyms of the upper subject, the lower subject and the upper and lower subjects is 0.5.
6. The method of claim 1, wherein the weight of the matched topic word is 1.0, and the weight of the synonym of the label of the upper topic of the nth layer, the lower topic of the nth layer and the upper topic of the nth layer is 0.5nWherein n is a natural number.
7. A knowledge-graph-based legal provision accurate search ranking system, comprising:
the input module is used for inputting keywords to be searched;
the system comprises a topic word bank generating module, a search module and a search module, wherein the topic word bank generating module is used for acquiring a topic word matched with a keyword to be searched and a relevant word of the topic word from a legal knowledge graph model to form a topic word bank; the associated words include: upper themes, lower themes and label synonyms; the upper theme refers to a theme with a large representative range, the lower theme refers to a theme with a small representative range, and the label synonym is a theme with a similar representative range defined in the label; the subject term attributes include: the name and weight of the subject term are (0-1);
the weight generation module is used for dynamically generating the weight of the subject words and the associated words in the subject word bank of the same case according to the association relation with the core subject in the legal knowledge graph model, wherein the higher the weight is, the closer the scope represented by the core subject is to the weight;
the subset combination generating module is used for acquiring subset combinations of the subject thesaurus and sequentially forming the subset combinations with the element number of m, m-1 and m-2 … 2; respectively taking out a subject word from each subset combination to form a subject word combination finally used for searching; sorting all subset combinations according to the number of the topic word combinations finally used for searching and the total weight of the keywords;
and the search result display module is used for searching in the legal provision database by using the selected subject word combinations after the ordering, and ordering and displaying the search results.
8. A computer-readable storage medium comprising a computer program executable by a processor, the computer program tangibly embodying the method of accurate knowledge-graph-based search ranking of legal provisions according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810361909.8A CN108563773B (en) | 2018-04-20 | 2018-04-20 | Knowledge graph-based legal provision accurate search ordering method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810361909.8A CN108563773B (en) | 2018-04-20 | 2018-04-20 | Knowledge graph-based legal provision accurate search ordering method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108563773A CN108563773A (en) | 2018-09-21 |
| CN108563773B true CN108563773B (en) | 2021-03-30 |
Family
ID=63536154
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810361909.8A Active CN108563773B (en) | 2018-04-20 | 2018-04-20 | Knowledge graph-based legal provision accurate search ordering method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108563773B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109947952B (en) * | 2019-03-20 | 2021-03-02 | 武汉市软迅科技有限公司 | Retrieval method, device, equipment and storage medium based on English knowledge graph |
| CN110321408B (en) * | 2019-05-30 | 2023-07-14 | 广东省智湾汇科技有限公司 | Searching method and device based on knowledge graph, computer equipment and storage medium |
| CN110674285A (en) * | 2019-09-18 | 2020-01-10 | 国网安徽省电力有限公司芜湖供电公司 | Intelligent retrieval system and method for power dispatching machine accounts |
| CN110674316B (en) * | 2019-09-27 | 2022-05-31 | 腾讯科技（深圳）有限公司 | Data conversion method and related device |
| CN112579786A (en) * | 2019-09-30 | 2021-03-30 | 北京国双科技有限公司 | Construction method and device of atlas based on record, storage medium and equipment |
| CN110928992B (en) * | 2019-11-21 | 2022-06-10 | 邝俊伟 | Text searching method, device, server and storage medium |
| CN111191042A (en) * | 2019-12-10 | 2020-05-22 | 同济大学 | Knowledge graph path semantic relation-based search accuracy evaluation method |
| CN113127761A (en) * | 2019-12-31 | 2021-07-16 | 中国科学技术信息研究所 | Intelligent sorting method for scientific and technological element retrieval, electronic equipment and storage medium |
| CN111563168B (en) * | 2020-03-03 | 2022-12-13 | 天津蒙比利埃创新网络科技有限公司 | Method for intelligently classifying customs commodities based on AI knowledge graph algorithm |
| CN112182184B (en) * | 2020-09-29 | 2023-04-11 | 国网浙江省电力有限公司 | Audit database-based accurate matching search method |
| CN115905577B (en) * | 2023-02-08 | 2023-06-02 | 支付宝(杭州)信息技术有限公司 | Knowledge graph construction method and device and rule retrieval method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104090863A (en) * | 2014-07-24 | 2014-10-08 | 高德良 | Intelligent legal instrument generating method and system |
| CN105069167A (en) * | 2015-08-28 | 2015-11-18 | 成都六四三六五科技有限公司 | Legal information search method and legal information search apparatus |
| CN105893551A (en) * | 2016-03-31 | 2016-08-24 | 上海智臻智能网络科技股份有限公司 | Method and device for processing data and knowledge graph |
| CN106874695A (en) * | 2017-03-22 | 2017-06-20 | 北京大数医达科技有限公司 | The construction method and device of medical knowledge collection of illustrative plates |
| CN107122444A (en) * | 2017-04-24 | 2017-09-01 | 北京科技大学 | A kind of legal knowledge collection of illustrative plates method for auto constructing |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9104754B2 (en) * | 2011-03-15 | 2015-08-11 | International Business Machines Corporation | Object selection based on natural language queries |
| US9613055B2 (en) * | 2014-05-09 | 2017-04-04 | Sap Se | Querying spatial data in column stores using tree-order scans |
-
2018
- 2018-04-20 CN CN201810361909.8A patent/CN108563773B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104090863A (en) * | 2014-07-24 | 2014-10-08 | 高德良 | Intelligent legal instrument generating method and system |
| CN105069167A (en) * | 2015-08-28 | 2015-11-18 | 成都六四三六五科技有限公司 | Legal information search method and legal information search apparatus |
| CN105893551A (en) * | 2016-03-31 | 2016-08-24 | 上海智臻智能网络科技股份有限公司 | Method and device for processing data and knowledge graph |
| CN106874695A (en) * | 2017-03-22 | 2017-06-20 | 北京大数医达科技有限公司 | The construction method and device of medical knowledge collection of illustrative plates |
| CN107122444A (en) * | 2017-04-24 | 2017-09-01 | 北京科技大学 | A kind of legal knowledge collection of illustrative plates method for auto constructing |
Non-Patent Citations (1)
| Title |
|---|
| 基于CSSCI的国内公共安全研究知识图谱分析;汤志伟等;《现代情报》;20170228;第37卷(第2期);第119-125页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108563773A (en) | 2018-09-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108563773B (en) | Knowledge graph-based legal provision accurate search ordering method | |
| CN111353310B (en) | Named entity identification method and device based on artificial intelligence and electronic equipment | |
| CN109829104B (en) | Semantic similarity based pseudo-correlation feedback model information retrieval method and system | |
| US9715493B2 (en) | Method and system for monitoring social media and analyzing text to automate classification of user posts using a facet based relevance assessment model | |
| US9280535B2 (en) | Natural language querying with cascaded conditional random fields | |
| US20110191335A1 (en) | Method and system for conducting legal research using clustering analytics | |
| CN110674252A (en) | High-precision semantic search system for judicial domain | |
| Nguyen et al. | A math-aware search engine for math question answering system | |
| Remi et al. | Domain ontology driven fuzzy semantic information retrieval | |
| Wang et al. | Data-driven approach for bridging the cognitive gap in image retrieval | |
| CN115563313A (en) | Knowledge graph-based document book semantic retrieval system | |
| CN115374781A (en) | Text data information mining method, device and equipment | |
| CN114491079A (en) | Knowledge graph construction and query method, device, equipment and medium | |
| Khin et al. | Query classification based information retrieval system | |
| Lovelyn Rose et al. | Normalized web distance based web query classification | |
| CN116737758A (en) | Database query statement generation method, device, equipment and storage medium | |
| Kim et al. | Coliee-2018: Evaluation of the competition on case law information extraction and entailment | |
| Golub | Automated subject classification of textual documents in the context of web-based hierarchical browsing | |
| CN111241283B (en) | Rapid characterization method for portrait of scientific research student | |
| Varnaseri et al. | The assessment of the effect of query expansion on improving the performance of scientific texts retrieval in Persian | |
| Leung et al. | CYC based query expansion framework for effective image retrieval | |
| Ingle | Processing of unstructured data for information extraction | |
| Gupta et al. | A survey of existing question answering techniques for Indian languages | |
| Kambau et al. | Unified concept-based multimedia information retrieval technique | |
| Gopianand et al. | XML web quality analysis by employing MFCM clustering technique and KNN classification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |