CN117332045B - Legal searching method and legal searching system - Google Patents
Legal searching method and legal searching system Download PDFInfo
- Publication number
- CN117332045B CN117332045B CN202311344406.7A CN202311344406A CN117332045B CN 117332045 B CN117332045 B CN 117332045B CN 202311344406 A CN202311344406 A CN 202311344406A CN 117332045 B CN117332045 B CN 117332045B
- Authority
- CN
- China
- Prior art keywords
- event
- analysis
- legal
- laws
- similarity
- 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
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000004458 analytical method Methods 0.000 claims abstract description 134
- 230000033228 biological regulation Effects 0.000 claims description 34
- 238000000605 extraction Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 230000001939 inductive effect Effects 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 3
- 230000006698 induction Effects 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001550 time effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a legal searching method and a legal searching system, which comprise an analysis event pool, a legal resource pool, an event pool classification unit, an event input unit, an event retrieval unit, a depth analysis unit and a legal display unit. According to the method, the required legal provision is retrieved and extracted according to the event process related to laws, so that a retrieval result is more accurate, a target user unfamiliar with the legal provision can be helped to use the legal provision retrieval system better, weight analysis is carried out on all event laws extracted correspondingly by each analysis statement through the arrangement of the depth analysis unit, legal provision and legal provision corresponding to the event laws with higher weight value are extracted according to the weight value, the user does not need to carry out weight comparison on different obtained legal provision one by one, and the user can conveniently and rapidly and effectively read legal provision with higher correlation, and the functionality and convenience of the retrieval system are further improved.
Description
Technical Field
The invention relates to the technical field of legal search, in particular to a legal search method and a legal search system.
Background
Legal provision is a main basis for processing legal problems, along with the rapid development of legal construction and the rapid improvement of legal consciousness of people, the required legal provision can be efficiently and accurately found to bring convenience to people in using law, so that more and more people can go to online searching legal solving ways when encountering problems, and more people generate demands for searching legal provision.
In the related art method for searching legal provision through keywords, the searched result does not consider the applicability of the legal provision, so that the searched result is not required by a user;
meanwhile, there are two main existing legal searching methods, one is to select a search title and/or content and/or attribute according to a search condition, for example: issuing date, issuing institution, efficacy grade, time effect and the like, and showing legal title hit by law; another legal searching method is to display the hit legal provision by taking the legal provision as a basic unit;
the two methods do not consider the multi-level characteristics of law, either only display legal titles, need to read and search needed contents for the second time manually, or adopt a linear law displaying mode to directly display laws, so that precious space of the first few pages is occupied.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a legal searching method and a legal searching system, which solve the problems in the background art.
In order to achieve the above purpose, the invention is realized by the following technical scheme: a legal search method and legal search system includes:
the analysis event pool is used for storing the analyzed historical event package; the historical event package comprises a historical event, a plurality of event section sentences in the historical event and event laws related to the event section sentences;
the historical event is shown as an event related to legal rules, which is reserved after analysis and used as a comparison event, and the event rule is shown as a legal rule of a corresponding event;
the legal provision resource pool is used for storing legal provision corresponding to the event legal provision and legal provision corresponding to the event legal provision;
the event pool classification unit is used for extracting keywords from each event section sentence of the historical event according to a text keyword extraction algorithm, taking the keywords as event labels of the corresponding event section sentences, calculating similarity values of each event label according to a semantic text similarity algorithm, marking the similarity values as text similarity values, carrying out inductive classification on the event section sentences according to the text similarity values, generating a plurality of label sub-pools according to inductive classification results, and locating the generated label sub-pools in the analytic event pool;
the event input unit is used for inputting event details by the target user, and then transmitting the event details to the event retrieval unit, wherein the event details are expressed as detailed processes of corresponding cases when the target user searches legal rules;
the event retrieval unit is used for decomposing event details into a plurality of analysis sentences, then extracting event keywords in each analysis sentence, acquiring corresponding tag sub-pools according to the event keywords, then carrying out similar matching on the analysis sentences and event section sentences in the corresponding acquired tag sub-pools, acquiring event laws related to the corresponding event section sentences according to matching results, and then sending the acquired event laws to the depth analysis unit;
the depth analysis unit is used for carrying out weight analysis on all event laws extracted correspondingly from each analysis statement, obtaining a laws weight table according to the weight analysis result, and then sending the laws weight table to the laws display unit;
the legal item display unit is used for acquiring a legal item weight table generated by a plurality of analysis sentences in the event details, acquiring legal items and legal regulations corresponding to the first event legal items from the legal item resource pool and automatically displaying the legal items and the legal regulations to the target user.
Preferably, the generalized classification mode of the event pool classification unit is as follows:
firstly, setting a first similarity threshold, wherein the first similarity threshold is a preset value;
then, randomly selecting an event label as a first reference label, simultaneously establishing a label sub-pool according to the first reference label, then respectively calculating corresponding text similarity values of other event labels and the first reference label through a semantic text similarity algorithm, and then comparing each text similarity value with a first similarity threshold value;
if the text similarity value is greater than or equal to the first similarity threshold, other corresponding event labels are matched with the first reference label, and event section sentences and event laws corresponding to the corresponding event labels are brought into a label sub-pool established by the first reference label;
if the text similarity value is smaller than the first similarity threshold, other corresponding event labels are not matched with the first reference label, the corresponding event labels are used as second reference labels, a label sub-pool is built according to the second reference labels, then the corresponding text similarity values are calculated by the other event labels and the second reference labels through a semantic text similarity algorithm, and then the text similarity values are compared with the first similarity threshold in the same comparison mode as the text similarity values calculated according to the first reference labels;
and so on, obtaining a plurality of label sub-pools established according to different reference labels;
preferably, the specific manner of the event retrieving unit obtaining the event laws is as follows:
SA1, acquiring the whole content of event details, and importing the whole content into a pre-trained event analysis model;
SA2, then the event analysis model disassembles the event details into a plurality of analysis sentences according to punctuations and sentence semantics;
SA3, then extracting event keywords from each analysis statement by the event analysis model through a text keyword extraction algorithm, and then calculating the similarity value of the event keywords and the event labels through a semantic text similarity algorithm, and recording the similarity value as an event similarity value;
then comparing the event similarity value with a preset second similarity threshold value, and extracting a corresponding label sub-pool according to a comparison result;
S4A, acquiring each analysis statement by the event analysis model, and acquiring a label sub-pool extracted by the analysis statement correspondingly to perform semantic analysis processing;
the semantic analysis processing mode is as follows:
SA41, taking an analysis statement as an example, extracting the analysis statement, and acquiring all event section statements from a label sub-pool which is correspondingly extracted;
SA42, then the event analysis model calculates the similarity value of the analysis sentences and the event section sentences one by one through a semantic text similarity algorithm, and marks the similarity value as a semantic similarity value;
and SA43, comparing the semantic similarity value with a preset third similarity threshold value, and extracting event laws corresponding to the corresponding event section sentences from the corresponding label sub-pools according to the comparison result.
Preferably, in step SA3, the event similarity value is compared with a preset second similarity threshold in the following manner:
if the event similarity value is greater than or equal to a preset second similarity threshold value, the analysis statement is matched with the corresponding event label, and then a label sub-pool corresponding to the event label is extracted;
if the event similarity value is smaller than a preset second similarity threshold value, the analysis statement is not matched with the corresponding event label, and then a label sub-pool corresponding to the event label is not extracted.
Preferably, in step SA43, the semantic similarity value is compared with a preset third similarity threshold in the following manner:
if the semantic similarity value is greater than or equal to a preset third similarity threshold value, the analysis statement is matched with the corresponding event section statement, and then an event law corresponding to the event section statement is extracted;
if the semantic similarity value is smaller than a preset third similarity threshold value, the analysis statement is not matched with the corresponding event section statement, and then the event law corresponding to the event section statement is not extracted.
Preferably, the weight analysis method is as follows:
SB1, selecting an analysis statement, and obtaining the total number ZS of each event law corresponding to the extraction;
SB2, then, in each event law which is extracted correspondingly, repeatedly screening the same event laws, and passing the repeated number FS in each same event law j ；
Meanwhile, in each identical event law, acquiring semantic similarity values of all event laws, and marking the semantic similarity values of each event law as YX i，j ；
SB3, then according to the formula:
calculating weight values Qj of the corresponding extracted different event laws;
wherein, alpha is a preset fixed proportion coefficient set, alpha 1, alpha 2, alpha 3 and alpha 4 are all fixed proportion coefficients in the fixed proportion coefficient set, and lambda 1 and lambda 2 are all preset values;
SB4, then the weight values of the event laws corresponding to the analysis statement are ordered in the order from big to small, and a laws weight table is generated.
Preferably, in the legal regulation weight table correspondingly generated by the analysis statement, legal regulations and legal regulations corresponding to other event legal regulations are actively acquired by the target user except for the first event.
A legal searching method is realized by the legal searching system.
Advantageous effects
The invention provides a legal searching method and a legal searching system. Compared with the prior art, the method has the following beneficial effects:
according to the method, the event section sentences corresponding to the historical event packages and the related event laws are classified by setting the event pool classification unit, so that the event laws related to the event section sentences in the analysis event pools are classified and stored according to the event labels, the retrieval analysis of a subsequent event retrieval unit is facilitated, the fact that the retrieval time is long due to the comparison analysis of irrelevant event section sentences and analysis sentences is avoided, and the retrieval analysis efficiency is further reduced;
according to the method, the event details are input, the event retrieval unit is utilized to acquire corresponding matched event section sentences in the analysis event pool, and corresponding event laws are acquired, so that a target user can retrieve and extract legal provision required by the target user according to the event process related to laws, the retrieval result is more accurate, and further the target user unfamiliar with the legal provision can better use the legal provision retrieval system;
according to the invention, weight analysis is carried out on all event laws and regulations extracted correspondingly by each analysis statement by arranging the depth analysis unit, legal laws and regulations corresponding to the event laws and regulations with higher weight values are extracted according to the weight values and are automatically displayed to a target user, the legal regulations are actively screened for the user, and the user does not need to carry out weight comparison again on different obtained legal regulations one by one, so that the user can conveniently, quickly and effectively read legal regulations with higher relevance, and the functionality and convenience of a retrieval system are further improved.
Drawings
FIG. 1 is a system block diagram of the present invention;
fig. 2 is a flow chart of the method of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the present invention provides the following technical solutions:
as an embodiment of the invention
A legal search system, comprising:
the analysis event pool is used for storing the analyzed historical event package; the historical event package comprises a historical event, a plurality of event section sentences in the historical event and event laws related to the event section sentences;
the historical event is shown as an event related to legal rules, which is reserved after analysis and used as a comparison event, and the event rule is shown as a legal rule of a corresponding event;
the legal provision resource pool is used for storing legal provision corresponding to the event legal provision and legal provision corresponding to the event legal provision;
the event pool classifying unit is used for extracting keywords from each event section sentence of the historical event according to a text keyword extracting algorithm, taking the keywords as event labels of the corresponding event section sentences, calculating similarity values of each event label according to a semantic text similarity algorithm, marking the similarity values as text similarity values, and then carrying out inductive classification on the event section sentences according to the text similarity values to generate a plurality of label sub-pools;
wherein, the semantic text similarity algorithm and the text keyword extraction algorithm are known to the person skilled in the art;
the method comprises the following steps:
firstly, setting a first similarity threshold, wherein the first similarity threshold is a preset value and is used for comparing with a text similarity value;
then, randomly selecting an event label as a first reference label, simultaneously establishing a label sub-pool according to the first reference label, then respectively calculating corresponding text similarity values of other event labels and the first reference label through a semantic text similarity algorithm, and then comparing each text similarity value with a first similarity threshold value;
if the text similarity value is greater than or equal to the first similarity threshold, other corresponding event labels are matched with the first reference label, and event section sentences and event laws corresponding to the corresponding event labels are brought into a label sub-pool established by the first reference label;
if the text similarity value is smaller than the first similarity threshold, other corresponding event labels are not matched with the first reference label, the corresponding event labels are used as second reference labels, a label sub-pool is built according to the second reference labels, then the corresponding text similarity values are calculated by the other event labels and the second reference labels through a semantic text similarity algorithm, and then the text similarity values are compared with the first similarity threshold in the same comparison mode as the text similarity values calculated according to the first reference labels;
and so on, obtaining a plurality of label sub-pools established according to different reference labels;
according to the method, the event section sentences corresponding to the historical event packages and the related event laws are classified by setting the event pool classification unit, so that the event laws related to the event section sentences in the analysis event pools are classified and stored according to the event labels, the retrieval analysis of the subsequent event retrieval unit is facilitated, the fact that the retrieval time is long due to the fact that the unrelated event section sentences are compared and analyzed with the analysis sentences is avoided, and the retrieval analysis efficiency is further reduced.
As embodiment II of the present invention
The first embodiment further includes, on the basis of the first embodiment:
the event input unit is used for inputting event details by the target user, wherein the event details are expressed as detailed processes of corresponding cases when the target user searches legal rules;
the event retrieval unit is used for decomposing event details into a plurality of analysis sentences, then extracting event keywords in each analysis sentence, acquiring corresponding tag sub-pools according to the event keywords, then carrying out similar matching on the analysis sentences and event section sentences in the corresponding acquired tag sub-pools, acquiring event laws related to the corresponding event section sentences according to matching results, and then sending the acquired event laws to the depth analysis unit;
the concrete mode is as follows:
SA1, acquiring the whole content of event details, and importing the whole content into a pre-trained event analysis model;
SA2, then the event analysis model disassembles the event details into a plurality of analysis sentences according to punctuations and sentence semantics;
SA3, then extracting event keywords from each analysis statement by the event analysis model through a text keyword extraction algorithm, and then calculating the similarity value of the event keywords and the event labels through a semantic text similarity algorithm, and recording the similarity value as an event similarity value;
then comparing the event similarity value with a preset second similarity threshold value, and extracting a corresponding label sub-pool according to a comparison result;
the comparison mode is as follows:
if the event similarity value is greater than or equal to a preset second similarity threshold value, the analysis statement is matched with the corresponding event label, and then a label sub-pool corresponding to the event label is extracted;
if the event similarity value is smaller than a preset second similarity threshold value, the analysis statement is not matched with the corresponding event label, and then a label sub-pool corresponding to the event label is not extracted;
S4A, acquiring each analysis statement by the event analysis model, and acquiring a label sub-pool extracted by the analysis statement correspondingly to perform semantic analysis processing;
the semantic analysis processing mode is as follows:
SA41, taking an analysis statement as an example, extracting the analysis statement, and acquiring all event section statements from a label sub-pool which is correspondingly extracted;
SA42, then the event analysis model calculates the similarity value of the analysis sentences and the event section sentences one by one through a semantic text similarity algorithm, and marks the similarity value as a semantic similarity value;
SA43, comparing the semantic similarity value with a preset third similarity threshold value, and extracting event laws corresponding to the corresponding event section sentences from the corresponding label sub-pools according to the comparison result;
the comparison mode is as follows:
if the semantic similarity value is greater than or equal to a preset third similarity threshold value, the analysis statement is matched with the corresponding event section statement, and then an event law corresponding to the event section statement is extracted;
if the semantic similarity value is smaller than a preset third similarity threshold value, the analysis statement is not matched with the corresponding event section statement, and then the event law corresponding to the event section statement is not extracted;
SA5, then sending the acquired event laws to a depth analysis unit;
according to the method and the device, through inputting event details, the event retrieval unit is utilized to acquire corresponding matched event pitch sentences in the analysis event pool and corresponding event laws, so that a target user can retrieve and extract legal regulations required by the target user according to an event process related to laws, the retrieval result is more accurate, and further the target user unfamiliar with the legal regulations can better use the legal regulations retrieval system.
Embodiment III as the present invention
The present embodiment further includes, on the basis of the first embodiment and the second embodiment:
the depth analysis unit is used for carrying out weight analysis on all event laws extracted correspondingly from each analysis statement, obtaining a laws weight table according to the weight analysis result, and then sending the laws weight table to the laws display unit;
the weight analysis mode is as follows:
SB1, taking an analysis statement as an example, obtaining the total number ZS of each event law which is correspondingly extracted;
SB2, then, in each event law which is extracted correspondingly, repeatedly screening the same event laws, and passing the repeated number FS in each same event law j ；
Meanwhile, in each identical event law, acquiring semantic similarity values of all event laws, and marking the semantic similarity values of each event law as YX i，j ；
SB3, then according to the formula:
calculating weight values Qj of the corresponding extracted different event laws;
wherein, alpha is a preset fixed proportion coefficient set, alpha 1, alpha 2, alpha 3 and alpha 4 are all fixed proportion coefficients in the fixed proportion coefficient set, and lambda 1 and lambda 2 are all preset values;
SB4, then sorting the weight values of the event laws corresponding to the analysis statement according to the sequence from big to small, and generating a law weight table;
the legal item display unit is used for acquiring a legal item weight table generated by a plurality of analysis sentences in the event details correspondingly, acquiring legal items and legal regulations corresponding to the first event legal items from a legal item resource pool and automatically displaying the legal items and the legal regulations to a target user;
in a legal regulation weight table correspondingly generated by analysis sentences, legal regulations and legal regulations corresponding to other event legal regulations are actively acquired by a target user except for the first event;
according to the invention, weight analysis is carried out on all event laws and regulations extracted correspondingly by each analysis statement by arranging the depth analysis unit, legal laws and regulations corresponding to the event laws and regulations with higher weight values are extracted according to the weight values and are automatically displayed to a target user, the legal regulations are actively screened for the user, and the user does not need to carry out weight comparison again on different obtained legal regulations one by one, so that the user can conveniently, quickly and effectively read legal regulations with higher relevance, and the functionality and convenience of a retrieval system are further improved.
Referring to fig. 2, the present invention further provides a technical solution: a legal search method, the method implemented by a legal search system, comprising the steps of:
step one, inputting event details in a search system by a target user;
step two, decomposing event details into a plurality of analysis sentences, then extracting event keywords in each analysis sentence, and acquiring corresponding tag sub-pools according to the event keywords;
step three, performing similar matching on the analysis sentences and the event section sentences in the corresponding acquisition label sub-pools, and acquiring event laws related to the corresponding event section sentences according to matching results;
and fourthly, carrying out weight analysis on all event laws corresponding to and extracted from each analysis statement, obtaining a law weight table according to the weight analysis result, then extracting the event laws with the highest weight value in the law weight table, and acquiring law laws and legal regulations corresponding to the event laws from a law resource pool and automatically displaying the law laws and legal regulations to a target user.
And all that is not described in detail in this specification is well known to those skilled in the art.
The foregoing describes one embodiment of the present invention in detail, but the disclosure is only a preferred embodiment of the present invention and should not be construed as limiting the scope of the invention. All equivalent changes and modifications within the scope of the present invention are intended to be covered by the present invention.
Claims (8)
1. A legal search system, comprising:
the analysis event pool is used for storing the analyzed historical event package; the historical event package comprises a historical event, a plurality of event section sentences in the historical event and event laws related to the event section sentences;
the historical event is shown as an event related to legal rules, which is reserved after analysis and used as a comparison event, and the event rule is shown as a legal rule of a corresponding event;
the legal provision resource pool is used for storing legal provision corresponding to the event legal provision and legal provision corresponding to the event legal provision;
the event pool classification unit is used for extracting keywords from each event section sentence of the historical event according to a text keyword extraction algorithm, taking the keywords as event labels of the corresponding event section sentences, calculating similarity values of each event label according to a semantic text similarity algorithm, marking the similarity values as text similarity values, carrying out inductive classification on the event section sentences according to the text similarity values, generating a plurality of label sub-pools according to inductive classification results, and locating the generated label sub-pools in the analytic event pool;
the event input unit is used for inputting event details by the target user, and then transmitting the event details to the event retrieval unit, wherein the event details are expressed as detailed processes of corresponding cases when the target user searches legal rules;
the event retrieval unit is used for decomposing event details into a plurality of analysis sentences, then extracting event keywords in each analysis sentence, acquiring corresponding tag sub-pools according to the event keywords, then carrying out similar matching on the analysis sentences and event section sentences in the corresponding acquired tag sub-pools, acquiring event laws related to the corresponding event section sentences according to matching results, and then sending the acquired event laws to the depth analysis unit;
the depth analysis unit is used for carrying out weight analysis on all event laws extracted correspondingly from each analysis statement, obtaining a laws weight table according to the weight analysis result, and then sending the laws weight table to the laws display unit;
the legal item display unit is used for acquiring a legal item weight table generated by a plurality of analysis sentences in the event details, acquiring legal items and legal regulations corresponding to the first event legal items from the legal item resource pool and automatically displaying the legal items and the legal regulations to the target user.
2. The legal search system of claim 1, where: the induction classification mode of the event pool classification unit is as follows:
firstly, setting a first similarity threshold, wherein the first similarity threshold is a preset value;
then, randomly selecting an event label as a first reference label, simultaneously establishing a label sub-pool according to the first reference label, then respectively calculating corresponding text similarity values of other event labels and the first reference label through a semantic text similarity algorithm, and then comparing each text similarity value with a first similarity threshold value;
if the text similarity value is greater than or equal to the first similarity threshold, other corresponding event labels are matched with the first reference label, and event section sentences and event laws corresponding to the corresponding event labels are brought into a label sub-pool established by the first reference label;
if the text similarity value is smaller than the first similarity threshold, other corresponding event labels are not matched with the first reference label, the corresponding event labels are used as second reference labels, a label sub-pool is built according to the second reference labels, then the corresponding text similarity values are calculated by the other event labels and the second reference labels through a semantic text similarity algorithm, and then the text similarity values are compared with the first similarity threshold in the same comparison mode as the text similarity values calculated according to the first reference labels;
and so on, obtaining a plurality of label sub-pools established according to different reference labels.
3. The legal search system of claim 1, where: the specific mode of the event retrieval unit for acquiring the event laws is as follows:
SA1, acquiring the whole content of event details, and importing the whole content into a pre-trained event analysis model;
SA2, then the event analysis model disassembles the event details into a plurality of analysis sentences according to punctuations and sentence semantics;
SA3, then extracting event keywords from each analysis statement by the event analysis model through a text keyword extraction algorithm, and then calculating the similarity value of the event keywords and the event labels through a semantic text similarity algorithm, and recording the similarity value as an event similarity value;
then comparing the event similarity value with a preset second similarity threshold value, and extracting a corresponding label sub-pool according to a comparison result;
S4A, acquiring each analysis statement by the event analysis model, and acquiring a label sub-pool extracted by the analysis statement correspondingly to perform semantic analysis processing;
the semantic analysis processing mode is as follows:
SA41, selecting an analysis statement, extracting the analysis statement, and acquiring all event section statements from a label sub-pool which is correspondingly extracted;
SA42, then the event analysis model calculates the similarity value of the analysis sentences and the event section sentences one by one through a semantic text similarity algorithm, and marks the similarity value as a semantic similarity value;
and SA43, comparing the semantic similarity value with a preset third similarity threshold value, and extracting event laws corresponding to the corresponding event section sentences from the corresponding label sub-pools according to the comparison result.
4. A legal search system according to claim 3, characterized in that: in step SA3, the event similarity value is compared with a preset second similarity threshold in the following manner:
if the event similarity value is greater than or equal to a preset second similarity threshold value, the analysis statement is matched with the corresponding event label, and then a label sub-pool corresponding to the event label is extracted;
if the event similarity value is smaller than a preset second similarity threshold value, the analysis statement is not matched with the corresponding event label, and then a label sub-pool corresponding to the event label is not extracted.
5. A legal search system according to claim 3, characterized in that: in step SA43, the semantic similarity value is compared with a preset third similarity threshold in the following manner:
if the semantic similarity value is greater than or equal to a preset third similarity threshold value, the analysis statement is matched with the corresponding event section statement, and then an event law corresponding to the event section statement is extracted;
if the semantic similarity value is smaller than a preset third similarity threshold value, the analysis statement is not matched with the corresponding event section statement, and then the event law corresponding to the event section statement is not extracted.
6. The legal search system of claim 1, where: the weight analysis mode is as follows:
SB1, selecting an analysis statement, and obtaining the total number ZS of each event law corresponding to the extraction;
SB2, then, in each event law which is extracted correspondingly, repeatedly screening the same event laws, and passing the repeated number FS in each same event law j ；
Meanwhile, in each identical event law, acquiring semantic similarity values of all event laws, and marking the semantic similarity values of each event law as YX i，j ；
SB3, then according to the formula:
calculating the weight value of each corresponding extracted different event lawQj；
Wherein, alpha is a preset fixed proportion coefficient set, alpha 1, alpha 2, alpha 3 and alpha 4 are all fixed proportion coefficients in the fixed proportion coefficient set, and lambda 1 and lambda 2 are all preset values;
SB4, then the weight values of the event laws corresponding to the analysis statement are ordered in the order from big to small, and a laws weight table is generated.
7. The legal search system of claim 6, where: in a legal rule weight table correspondingly generated by the analysis statement, legal rules and legal regulations corresponding to other event legal rules are actively acquired by a target user except for the first event.
8. A legal searching method, characterized in that it is implemented by a legal searching system as defined in any one of claims 1-7, comprising the steps of:
step one, inputting event details in a search system by a target user;
step two, decomposing event details into a plurality of analysis sentences, then extracting event keywords in each analysis sentence, and acquiring corresponding tag sub-pools according to the event keywords;
step three, performing similar matching on the analysis sentences and the event section sentences in the corresponding acquisition label sub-pools, and acquiring event laws related to the corresponding event section sentences according to matching results;
and fourthly, carrying out weight analysis on all event laws corresponding to and extracted from each analysis statement, obtaining a law weight table according to the weight analysis result, then extracting the event laws with the highest weight value in the law weight table, and acquiring law laws and legal regulations corresponding to the event laws from a law resource pool and automatically displaying the law laws and legal regulations to a target user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311344406.7A CN117332045B (en) | 2023-10-17 | 2023-10-17 | Legal searching method and legal searching system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311344406.7A CN117332045B (en) | 2023-10-17 | 2023-10-17 | Legal searching method and legal searching system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117332045A CN117332045A (en) | 2024-01-02 |
| CN117332045B true CN117332045B (en) | 2024-03-12 |
Family
ID=89278993
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311344406.7A Active CN117332045B (en) | 2023-10-17 | 2023-10-17 | Legal searching method and legal searching system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117332045B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112632225A (en) * | 2020-12-29 | 2021-04-09 | 天津汇智星源信息技术有限公司 | Semantic searching method and device based on case and event knowledge graph and electronic equipment |
| CN116662643A (en) * | 2023-04-26 | 2023-08-29 | 华院计算技术(上海)股份有限公司 | Legal recommendation method, legal recommendation system, electronic device and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014190092A1 (en) * | 2013-05-22 | 2014-11-27 | Quantros, Inc. | Probabilistic event classification systems and methods |
-
2023
- 2023-10-17 CN CN202311344406.7A patent/CN117332045B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112632225A (en) * | 2020-12-29 | 2021-04-09 | 天津汇智星源信息技术有限公司 | Semantic searching method and device based on case and event knowledge graph and electronic equipment |
| CN116662643A (en) * | 2023-04-26 | 2023-08-29 | 华院计算技术(上海)股份有限公司 | Legal recommendation method, legal recommendation system, electronic device and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Dynamic Event Type Recognition and Tagging for Data-driven Insights in Law-Enforcement;Shayan Zamanirad 等;SPRINGER LINK;20200131;1-22 * |
| 基于事件抽取的学科建设知识图谱构建与应用;李家瑞 等;计算机系统应用;20220715;第31卷(第11期);100-110 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117332045A (en) | 2024-01-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103699625B (en) | Method and device for retrieving based on keyword | |
| US9058394B2 (en) | Matching and recommending relevant videos and media to individual search engine results | |
| WO2018120899A1 (en) | Trademark inquiry result proximity evaluating and sorting method and device | |
| US9256649B2 (en) | Method and system of filtering and recommending documents | |
| US20080162514A1 (en) | System and method for generating a relationship network | |
| CN102789464B (en) | Natural language processing methods, devices and systems based on semantics identity | |
| WO2008055120A2 (en) | System and method for summarizing search results | |
| US7162413B1 (en) | Rule induction for summarizing documents in a classified document collection | |
| WO2011127655A1 (en) | Method for keyword extraction | |
| CN102789452A (en) | Similar content extraction method | |
| CN109446313B (en) | Sequencing system and method based on natural language analysis | |
| CN111625621A (en) | Document retrieval method and device, electronic equipment and storage medium | |
| CN106570196B (en) | Video program searching method and device | |
| CN108345694B (en) | Document retrieval method and system based on theme database | |
| Fodil et al. | Theme classification of Arabic text: A statistical approach | |
| CN109948154A (en) | A kind of personage's acquisition and relationship recommender system and method based on name | |
| CN105574004B (en) | A kind of removing duplicate webpages method and apparatus | |
| JP2003150624A (en) | Information extraction device and information extraction method | |
| CN117332045B (en) | Legal searching method and legal searching system | |
| JP4428703B2 (en) | Information retrieval method and system, and computer program | |
| CN111914154B (en) | Intelligent search guiding system and method | |
| JP2000148770A (en) | Device and method for classifying question documents and record medium where program wherein same method is described is recorded | |
| CN117909491B (en) | Document metadata analysis method and system based on Bayesian network | |
| JP2008234557A (en) | Reputation information retrieval device, its method and program | |
| SW | Implementation of Information Retrieval System for Document Search Based on User Keywords |
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 |