CN111352988B

CN111352988B - Big data warehouse storage, analysis and extraction system aiming at legal information

Info

Publication number: CN111352988B
Application number: CN202010132115.1A
Authority: CN
Inventors: 吴怡
Original assignee: Chongqing Best Daniel Robot Co ltd
Current assignee: Chongqing Best Daniel Robot Co ltd
Priority date: 2020-02-29
Filing date: 2020-02-29
Publication date: 2023-05-23
Anticipated expiration: 2040-02-29
Also published as: CN111352988A

Abstract

The invention relates to the field of big data processing, in particular to a big data warehouse storage, analysis and extraction system aiming at legal information, which comprises the following components: the input module is used for sending a data request of legal requirements of a user; the operation module is used for receiving the data request of legal requirements of the user and operating the data request; the acquisition module is used for acquiring a data request with condition information and sending the condition information; the extraction module is used for receiving the condition information and extracting an operation result according to the condition information; and the output module is used for outputting the extracted operation result in real time. The invention can reduce the occupancy rate of the CPU of the server and improve the response rate during peak time, and prevent the system from being blocked and crashed, thereby improving the user experience.

Description

Big data warehouse storage, analysis and extraction system aiming at legal information

Technical Field

The invention relates to the field of legal service, in particular to a large data warehouse storage, analysis and extraction system for legal service information.

Background

With the gradual enhancement of legal consciousness of people, public and law institutions have come out of various legal consultation and inquiry systems so that the public can acquire the latest legal dynamics in time. The document CN110059193A discloses a legal consultation system based on statistical analysis of legal semantic parts and document big data, relates to the technical field of judicial, and particularly relates to a system which is oriented to the field of judicial and can perform multi-round natural language interaction with a user, dynamically generate consultation reports and solve legal problems of the user; based on a massive machine pre-labeling referee document data set, an interactive mode capable of describing legal problems encountered by the user through a spoken question-answering mode is provided for the user through introducing a neural network model, and a customized consultation report generated based on continuously added legal texts can be provided for the user through a natural text information extraction technology. The method gets rid of the prior legal consultation based on expert system, which can only provide the fixed questionnaire and answer combination which are input in advance, and is more close to the effect of manual consultation.

Because of the wide and numerous operators in China, the server needs to extract, analyze, backup and other processes on massive data in the peak period of legal service demand of users, which occupies large CPU utilization rate of the server, reduces the response speed of the server and affects the user experience.

Disclosure of Invention

The invention provides a large data warehouse storage, analysis and extraction system for legal information, which solves the technical problems of high utilization rate occupation of a CPU (Central processing Unit) of a server, low response speed and poor user experience of the conventional legal service system in the peak period of legal service demand of users.

The basic scheme provided by the invention is a large data warehouse storage, analysis and extraction system aiming at legal information, comprising the following steps: the input module is used for sending a data request of legal requirements of a user; the operation module is used for receiving the data request of legal requirements of the user and operating the data request; the acquisition module is used for acquiring a data request with condition information and sending the condition information; the extraction module is used for receiving the condition information and extracting an operation result according to the condition information; and the output module is used for outputting the extracted operation result in real time.

The working principle of the invention is as follows: the server firstly judges whether the access amount of the user exceeds a preset threshold value, and if the access amount exceeds the preset threshold value, the access amount is in a peak period. When the user access is in the peak period, the server requests to perform simulation operation according to legal requirement data of the user, and the operation result is not directly extracted after the operation is finished. But firstly extracting the condition information in the data request, and then extracting the operation results in batches and successively according to the condition information, so that the occupancy rate of the CPU of the server is reduced. The invention has the advantages that: the CPU occupancy rate of the server can be reduced during peak hours, and the system is prevented from being blocked and crashed, so that the user experience is improved.

According to the invention, the operation results are extracted in batches and successively according to the condition information of the data request, instead of extracting the operation results simultaneously, so that the operation pressure of the system is relieved, and the operation speed is increased.

Further, the operation module is further configured to obtain an instant usage rate of the server CPU, and if the obtained instant usage rate of the server CPU is greater than a first threshold, operate the data request in batches; and ensuring that the instant utilization rate of the CPU of the server is smaller than or equal to a first threshold value when each batch of data requests run. Thus, the utilization rate of the CPU of the server can be ensured to be always kept below the first threshold value, and the first threshold value can be set manually. For example, the first threshold is set to a low value at ordinary times and set to a high value at peak times, so that the response rate of the system can be improved as much as possible.

Further, the operation module is also used for acquiring the instantaneity data in the data request of the legal requirement of the user and sequencing the data request according to the instantaneity requirement. Thus, the system can process according to the instant requirement in sequence, and the operation pressure is relieved.

Further, the operation module is also used for identifying legal fields and specific categories corresponding to legal requirements of the user. According to the legal system of China, cases can be divided into three legal fields of civil affairs, criminals and administrative affairs, and the procedures and rules for processing different legal fields are greatly different. On the other hand, each legal field contains a plurality of specific categories. For example, the civil law field includes a plurality of specific categories such as contracts, infringements, rights of matter, marital, etc., and each specific category has a unique rule. Therefore, it is necessary to distinguish between systems that operate orderly and efficiently.

Further, the operation module is also used for clustering the data requests of legal requirements of the user. Although the needs of users are diverse and various for each particular category of cases, the needs of some users are essentially common. Users of the same type separated by the clustering algorithm have great similarity, and similar users are all integrated into the same type. Therefore, the types of users in each specific category can be reduced, and legal requirements of the users can be conveniently and intensively processed, so that the efficiency is improved, and the pressure of a server is reduced.

Further, the operation module is also used for detecting the included behavior targets required by the law of the user. Thus, the method is convenient for deeply analyzing the user demands and refining the information really having guiding effect on the user.

Further, the system also comprises a storage module for storing the data request and the operation result data of legal requirements of the user. This allows a large amount of user demand information to be collected, providing a large sample of data for optimization of the system.

Further, a learning module is also included for optimizing the algorithm of the system. Thus, the system is convenient to acquire new knowledge or skills according to the data request and the operation result data of legal requirements of the user, and the existing knowledge structure is reorganized to continuously improve the performance of the system.

Further, the system also comprises a feedback module for inquiring the user and feeding back according to the answer of the user. Thus, various information required for solving the user problem can be acquired as much as possible, and the operation result is corrected.

Further, the system comprises a rating module for collecting the rating, suggestion and opinion of the system. By knowing the user's ratings, suggestions and comments, the system is facilitated to be improved, thereby improving the user's experience.

Drawings

FIG. 1 is a block diagram of a system architecture of a first embodiment of a large data warehouse storage, analysis, extraction system for legal information according to the present invention.

Fig. 2 is a system architecture diagram of a second embodiment of the large data warehouse storage, analysis, and extraction system for legal information according to the present invention.

Detailed Description

The following is a detailed description of the embodiments.

Example 1

The embodiment of the large data warehouse storage, analysis and extraction system aiming at the legal information is basically shown in the figure 1, and comprises an input module, an operation module, an acquisition module, an extraction module and an output module.

At some point, many users (say, 20000 people) are simultaneously inputting respective legal requirements at the input end of the legal intelligent service robot, which may be an input interface. These legal requirements are then sent to the operating module, which is a server.

After receiving the data request of the legal requirement of 20000, the server firstly identifies the legal field and the specific category corresponding to the legal requirement of 20000 users through the keywords in the data request. For example, the server recognizes 15000 of these 20000 users as belonging to civil cases, 4000 as belonging to criminal cases, and 1000 as belonging to administrative cases. Of the 15000 civil cases 7000 belonged to contract disputes, 3000 belonged to infringement disputes, 2000 belonged to marital disputes, and 3000 belonged to property right disputes. Of 4000 criminal cases, 500 are harmful to public safety, 500 are damaging to socially market economy, 1000 are infringing citizen personal rights, democratic rights, 2000 are property crimes.

The data requests for legal requirements of the user in each specific category are then clustered. The present embodiment employs a k-means clustering algorithm (k-means clustering algorithm) for clustering. Before clustering, keywords in the data request of each user need to be extracted, in this embodiment, TF-IDF algorithm is adopted, and specific steps can refer to the prior art. After the keyword extraction is completed, clustering can be performed. Firstly, randomly selecting K keywords as initial clustering centers; second, each object is allocated to the cluster center nearest to it; third, recalculating a clustering center; fourth, if convergence, outputting a clustering result; if not, executing the first step. For a specific details of the clustering algorithm, reference is also made to the prior art. For example, the keywords extracted from 7000 contract disputes include "material", "goods", "delivery", "interest", "loan", "guarantee", "rent", "renew lease" … clustering algorithm, and the contract disputes containing keywords such as "material", "goods", "delivery" are classified into the category of buying and selling contract disputes, and the total number is 4000; contract disputes containing keywords such as interest, loan, guarantee and the like are classified into borrowing contract disputes, and the total number of contract disputes is 1000; contract disputes containing keywords such as "rent", "renew rent", "transfer rent" and the like are classified into the list of lease contract disputes, and a total of 2000 contract disputes.

And then, acquiring the instantaneity data in the data request of the legal requirement of the user, and sequencing the data request according to the instantaneity requirement. In this embodiment, the instantaneity refers to the remaining period during litigation, specifically, the shorter the remaining period, the higher the instantaneity requirement. Since national folk regulations state that citizens have not properly exercised rights in time to cause the aging period of litigation to pass, the debt becomes a natural debt and the execution force of the debt is reduced. Specifically, for 4000 trade contract disputes, the time keywords are acquired, the remaining deadlines during the corresponding litigation are calculated through the time keywords, the remaining deadlines are ranked in front, and the remaining deadlines are ranked in the back. For example, the third party and the fourth party input legal data requests on 1 st 7 th 2018, the third party disputes on 1 st 7 th 2017, and the fourth party disputes on 1 st 2016 7 th. According to the general rule of civil law, which starts to be implemented on 110 2017, the litigation time period of the buying and selling contract is 3 years, the remaining time period of the litigation time period of the Zhang three is 2 years, and the remaining time period of the litigation time period of the Lifour is 1 year. Thus, the situation of the fourth plum is urgent relative to the third plum, and should be preferentially processed, then the data request of the fourth plum should be arranged in front of the third plum. Similarly, 4000 business contract disputes may all be ordered.

And finally, acquiring the instant utilization rate of the CPU of the server. If the data requests of the 4000 persons for buying and selling contract dispute legal requirements are recorded into the server, the CPU utilization rate of the server is displayed to be larger than a first threshold value, and then the data requests are run in batches; and ensuring that the instant utilization rate of the CPU of the server is smaller than or equal to a first threshold value when each batch of data requests run. For example, the first threshold is artificially set to 85%, and if 4000 data requests are simultaneously run, the CPU usage is displayed as 95%. At this time, the batch operation is performed in the order described above, the data request amount of the operation is reduced, and the first 3500 pieces are processed: if the CPU utilization rate is reduced to below 85%, the 3500 person data request can be directly operated; if the CPU usage rate does not drop below 85% (for example, 90%), the data request amount of the current operation is reduced again, and the operation is not started until the CPU usage rate drops below 85%. The operation of each batch of data requests is performed in a similar manner to ensure that the CPU utilization is below 85%. In addition, the first threshold value may be set to a low value, such as 70%, at ordinary times, so that the response rate of the system may be improved as much as possible. At run-time, the included behavioral objectives of the user's legal requirements are also detected. Such as whether the user is simply consulting, is aware of litigation risk, or is truly complaint.

Then, condition information in the data requests is acquired, and then the running result is extracted according to the condition information. Such condition information includes the age, physical condition, cultural degree, etc. of the user, and the operation result of the user who is older and worse is preferentially extracted, so that the waiting time can be reduced. And outputting the operation result in real time while extracting the operation result. The output mode can be paper or electronic; the method can also be in a two-dimension code form, and a user can check the result only by scanning the two-dimension code.

Example two

One difference from the embodiment is only that: as shown in fig. 2, the system further comprises a storage module, a learning module, a feedback module and an evaluation module.

And after the server is operated, storing the data request and operation result data of legal requirements of the user. This provides a large sample of data for optimization of the system. The learning module optimizes the system by adopting a machine learning algorithm according to the stored large sample data, so that the system can conveniently acquire new knowledge or skills according to the data request and operation result data of legal requirements of users, and reorganizes the existing knowledge structure to continuously improve the performance of the system.

Before outputting the operation result, inquiring whether the user is satisfied with the result. And if the data request is not satisfied, prompting the user to modify, add or delete the data request required by law, and then feeding back according to the answer of the user, and putting the data request modified, added or deleted by the user into the next sequence for rerun. In addition, before the operation result is output, the user is prompted to make an evaluation on the work of the system, and the evaluation, suggestion and opinion of the user on the system are collected. By knowing the user's ratings, suggestions and comments, the system is facilitated to be improved, thereby improving the user's experience.

Example III

The only difference from the embodiment is that the run module is also used for recognition of daily language. In this embodiment, there is contract disputes between the users Li IV and Zhang III, and the content is approximately as follows: "5 tons of river sand are purchased from the plum four in 6 months of … Zhang Sanyu and 1 month, a river sand buying and selling contract is signed in the same day, the plum four transports the river sand warranty to a stone dam nearby Zhang Sanjia within 15 days of the contract, and the Zhang Sanjia receives the river sand to pay for once. And when the plum four transports the river sand to a stone dam nearby Zhang Sanjia, the price of the river sand is required to be paid for Zhang Sanjia by 2 ten thousand yuan in 6 th 2010. Furthermore, the plum four refers to "ding gold", and the difference is whether gold is fixed or gold fixed at this time.

First, pre-judgment is performed: outputting a subscription if no more than 20% or one-time payable information is included; if more than 20% or more of paid-out information is included, a subscription is output. For example, zhang three and Liu four have a convention of "ding gold" of 3500 yuan, which is less than 20% of the price (20000×0.2=4000), and "ding gold" should be fixed gold; conversely, if the contract "ding gold" for Zhang three and Li four is 4500 yuan, the "ding gold" is greater than 20% of the price (20000×0.2=4000), and the "ding gold" should be an order. For another example, zhang three and Lifour agree that "ding gold" is one-time clear, and the "ding gold" should be fixed gold; conversely, if the three and four items of land are paid off twice or three times, the land should be a subscription.

In fact, the distinction between subscription and gold is not clear due to the limited legal knowledge of the parties: (1) the contract for delivering the subscription is a slave contract, which is about to deliver the subscription and is unpaid, and does not form violation of the master contract; and the contract for delivering the contract fee is a part of the main contract, and the contract fee is delivered according to the contract and is not delivered, namely, the violation of the main contract is formed. (2) When the party who delivers and receives the subscription does not fulfill the contractual debt, no loss or double return of the pre-payment results occur and the subscription can only be made as a damage reimbursement. (3) The amount of the fund is not more than 20% of the amount of the main contract standard; while the amount of the subscription is in accordance with the free agreement between the parties, the law is generally not limiting. (4) The subscription has a guarantee property, but the subscription is only a unilateral action, and has no obvious guarantee property.

Thus, a principal may develop a contract that "rates 5000 yuan" that exceeds 20% of the contract's amount, with the excess not legally effecting the rate. At this point, the pre-determined decision cannot determine whether the 5000 elements are subscription or subscription. Then, the subsequent judgment is needed to be carried out, three options of 'fixed gold', 'ordered gold' and 'unknown' are output for the user to select, and if the user selects fixed gold or ordered gold, the result is directly output. In case the user chooses not to know, further decisions need to be made based on the information about Zhang three and Li four conventions. For example, if Li IV mentions "if I pull Hesha you don't pay for the tail, 5000I don't go back", it is seen that 5000 blocks have a guaranteed nature. Then the Zhang three and Liu four conventions should be "fixed" while those 1000 blocks that exceed 20% of the price have no fixed effectiveness. For another example, if Zhang San mentions that if you have salad the river, this 5000 blocks offset 5000 blocks, i pay only 15000 yuan, then this 5000 blocks has the effect of prepayment, and Zhang San and Lisi four conventions should be "order".

The foregoing is merely an embodiment of the present invention, and a specific structure and characteristics of common knowledge in the art, which are well known in the scheme, are not described herein, so that a person of ordinary skill in the art knows all the prior art in the application day or before the priority date of the present invention, and can know all the prior art in the field, and have the capability of applying the conventional experimental means before the date, so that a person of ordinary skill in the art can complete and implement the present embodiment in combination with his own capability in the light of the present application, and some typical known structures or known methods should not be an obstacle for a person of ordinary skill in the art to implement the present application. It should be noted that modifications and improvements can be made by those skilled in the art without departing from the structure of the present invention, and these should also be considered as the scope of the present invention, which does not affect the effect of the implementation of the present invention and the utility of the patent. The protection scope of the present application shall be subject to the content of the claims, and the description of the specific embodiments and the like in the specification can be used for explaining the content of the claims.

Claims

1. The large data warehouse storage, analysis and extraction system for the legal information is characterized by comprising the following components:

the input module is used for sending a data request of legal requirements of a user;

the operation module is used for receiving the data request of legal requirements of the user and operating the data request;

the acquisition module is used for acquiring a data request with condition information and sending the condition information;

the extraction module is used for receiving the condition information and extracting an operation result according to the condition information;

the output module is used for outputting the extracted operation result in real time;

the operation module is also used for identifying daily languages, and the operation module is used for distinguishing whether the daily languages are the subscription or the subscription:

first, pre-judgment is performed: outputting a subscription if no more than 20% or one-time payable information is included; outputting a subscription if more than 20% or more of the paid-up information is included;

if the rating exceeds 20% of the contract standard, then carrying out subsequent judgment, outputting three options for user selection, if the user selects rating or booking, directly outputting the result, and if the user selects unaware, carrying out further judgment according to the contracted information.

2. The large data warehouse storage, analysis, extraction system for legal information of claim 1, wherein: the operation module is also used for acquiring the instant utilization rate of the server CPU, and if the acquired instant utilization rate of the server CPU is greater than a first threshold value, the data request is operated in batches; and ensuring that the instant utilization rate of the CPU of the server is smaller than or equal to a first threshold value when each batch of data requests run.

3. The large data warehouse storage, analysis, extraction system for legal information of claim 2, wherein: the operation module is also used for acquiring the instantaneity data in the data request of the legal requirement of the user and sequencing the data request according to the instantaneity requirement.

4. A large data warehouse storage, analysis, extraction system for legal information as defined in claim 3, wherein: the operation module is also used for identifying the legal field and the specific category corresponding to the legal requirement of the user.

5. The large data warehouse storage, analysis, extraction system for legal information of claim 4, wherein: the operation module is also used for clustering the data requests of legal requirements of the users.

6. The large data warehouse storage, analysis, extraction system for legal information of claim 5, wherein: the operation module is also used for detecting the included behavior targets of the legal requirements of the user.

7. The large data warehouse storage, analysis, extraction system for legal information of claim 6, wherein: the system also comprises a storage module for storing the data request and the operation result data of legal requirements of the user.

8. The large data warehouse storage, analysis, extraction system for legal information of claim 7, wherein: the system also comprises a learning module for optimizing the algorithm of the system.

9. The large data warehouse storage, analysis, extraction system for legal information of claim 8, wherein: the system also comprises a feedback module which is used for inquiring the user and feeding back according to the answer of the user.

10. The large data warehouse storage, analysis, extraction system for legal information of claim 9, wherein: and the evaluation module is used for collecting the evaluation, suggestion and opinion of the system by the user.