CN111461932B - Administrative punishment free-cutting right rationality evaluation method and device based on big data - Google Patents
Administrative punishment free-cutting right rationality evaluation method and device based on big data Download PDFInfo
- Publication number
- CN111461932B CN111461932B CN202010273409.6A CN202010273409A CN111461932B CN 111461932 B CN111461932 B CN 111461932B CN 202010273409 A CN202010273409 A CN 202010273409A CN 111461932 B CN111461932 B CN 111461932B
- Authority
- CN
- China
- Prior art keywords
- case
- punishment
- amount
- data
- model
- 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
- 238000011156 evaluation Methods 0.000 title abstract description 11
- 238000005520 cutting process Methods 0.000 title description 3
- 238000000034 method Methods 0.000 claims abstract description 37
- 239000013598 vector Substances 0.000 claims description 43
- 238000012549 training Methods 0.000 claims description 34
- 238000013210 evaluation model Methods 0.000 claims description 19
- 238000012360 testing method Methods 0.000 claims description 18
- 238000000605 extraction Methods 0.000 claims description 13
- 230000002457 bidirectional effect Effects 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 11
- 230000007246 mechanism Effects 0.000 claims description 9
- 238000005315 distribution function Methods 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 8
- 238000009826 distribution Methods 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 10
- 238000003860 storage Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Abstract
The application relates to a administrative punishment free-cut-volume right rationality assessment method and device based on big data, comprising the steps of obtaining basic data; extracting characteristics of the basic data, determining a case reason of a case according to the extracted characteristic information, determining a punishment amount range according to the case reason, judging whether the punishment amount of the case is in the punishment amount range, constructing a punishment amount rationality assessment model after judging that the punishment amount of the case is in the punishment amount range, and judging the rationality of the punishment amount of the case according to the punishment amount rationality assessment model; the application establishes a rationality evaluation method and a rationality evaluation device for the free right of judgment, can assist law enforcement personnel to accurately exercise the administrative punishment free right of judgment, scientifically and scientifically make a law enforcement scientific decision, and provides a scientific and efficient solution for standardizing the application of the administrative punishment free right of judgment and solving the problem of abusing the free right of judgment in the administrative law enforcement.
Description
Technical Field
The invention belongs to the technical field of big data, and particularly relates to a method and a device for evaluating administrative punishment free-cut right rationality based on big data.
Background
The rationality assessment of administrative penalty amounts is becoming more and more important in the face of the unlimited possibilities of social development and the complexity of administrative matters, which is a very important ring in the administrative penalty process. At present, the problem that the judgment range of the punishment amount of the same illegal action is larger exists in laws and regulations related to the administrative punishment of China, the punishment amount cannot be quantitatively explained, and the problem that the free right is abused in the administrative law enforcement process easily occurs.
In the administrative punishment process, due to the influence of certain factors, the problems of excessive punishment, excessive punishment and different punishments in the same case can occur in some areas and units, and law enforcement personnel cannot accurately exercise the administrative punishment free right of judgment by means of personal subjective consciousness and experience in the administrative law enforcement process, so that serious adverse effects are easy to generate. Therefore, the above problems are very necessary to develop corresponding techniques to solve, and the evaluation of the rationality of the administrative penalty amount of the discretion has become a problem that needs to be solved urgently at present.
Disclosure of Invention
In view of the above, the invention aims to overcome the defects of the prior art, and provides a administrative punishment free-cut-volume right rationality evaluation method and device based on big data, so as to solve the problem that law enforcement personnel cannot accurately exercise the administrative punishment free-cut-volume right by means of personal subjective consciousness and experience in the administrative law enforcement process in the prior art.
In order to achieve the above purpose, the invention adopts the following technical scheme:
A big data-based administrative penalty discretionary right rationality assessment method comprises the following steps:
acquiring basic data;
Extracting features from the basic data, determining a case counter according to the extracted feature information, determining a punishment amount range according to the case counter, and judging whether the punishment amount of the case is within the punishment amount range;
after the fact that the punishment amount of the case is within the punishment amount range is judged, a punishment amount rationality evaluation model is built, and the rationality of the punishment amount of the case is judged according to the punishment amount rationality evaluation model.
Further, the obtaining basic data includes:
acquiring initial data in a database;
and cleaning the initial data to obtain basic data.
Further, the feature extraction is performed on the basic data, and the case scheme is determined according to the extracted feature information, including:
Extracting basic characteristics of basic data by adopting an information extraction technology;
and determining the case law of the case according to the extracted basic characteristics.
Further, determining a penalty amount range according to the case, and judging whether the penalty amount of the case is within the penalty amount range, including:
a range of penalty amounts specified by the determined laws and regulations according to the scheme;
judging whether the penalty amount of the case is within the range of the penalty amount specified by the laws and regulations.
Further, the penalty amount range defined by the determined law and regulation according to the case includes:
Determining punishment basis according to the basic characteristics of the cases;
and determining a punishment amount range specified by laws and regulations according to the punishment basis.
Further, the penalty amount rationality evaluation model determines rationality of the penalty amount for the case, including:
model training is carried out on the class case;
judging whether free judge of punishment amount exists according to punishment basis of the case; if so, acquiring similar cases according to the case content through a case similarity judging model, and judging whether the punishment amount is reasonable according to the similarity degree of the similar cases; otherwise, calculating a distribution function of the administrative punishment amount, and judging whether the punishment amount of the case meets the distribution.
Further, the model training on the class case includes:
Obtaining case data, obtaining non-case data in the same proportion, formatting the case data and the non-case data, and constructing a formatted data set;
the formatted data set comprises a training data set and a test data set, and the training data set is divided into N parts;
word2vec is used for mapping word vectors, the word vector is input into BiLSTM models for training, and a result is output;
Obtaining an optimal model through N times of cross validation, and storing the optimal model;
Loading the optimal model for testing, inputting test data in a test data set into the optimal model, sequentially performing word segmentation, word vector mapping and class similarity calculation, and evaluating the performance of the model.
Further, the BiLSTM model comprises a word vector embedding layer, a bidirectional LSTM layer, an attention mechanism layer and a classifier; the word vector embedding layer is used for converting training data into word vectors; the bidirectional LSTM layer takes word vectors as input, converts the word vectors and outputs feature vectors with fixed lengths; the attention mechanism layer takes the feature vector output by the bidirectional LSTM layer as input, and adds different weights according to the importance degree of words in the text to obtain more accurate text representation; the classification layer normalizes attention the weights with softmax, and performs weighted summation and then performs similarity calculation of the cosine with the parameters.
The embodiment of the application provides a administrative punishment free-discretion right rationality assessment device based on big data, which comprises the following steps:
The acquisition module is used for acquiring the basic data;
the extraction module is used for carrying out feature extraction on the basic data, determining a case counter according to the extracted feature information, determining a punishment amount range according to the case counter, and judging whether the punishment amount of the case is in the punishment amount range or not;
And the judging module is used for constructing a punishment amount rationality evaluation model after judging that the punishment amount of the case is in the punishment amount range and judging the rationality of the punishment amount of the case according to the punishment amount rationality evaluation model.
Further, the judging module includes:
the training unit is used for carrying out model training on the class case;
The calculation module is used for judging whether free amount of punishment exists according to punishment basis of the cases; if so, acquiring similar cases according to the case content through a case similarity judging model, and judging whether the punishment amount is reasonable according to the similarity degree of the similar cases; otherwise, calculating a distribution function of the administrative punishment amount, and judging whether the punishment amount of the case meets the distribution.
By adopting the technical scheme, the invention has the following beneficial effects:
The application establishes a administrative punishment free-right rationality evaluation method based on big data, establishes a rationality evaluation function for the punishment amount of the free-right, and can assist law enforcement personnel to accurately exercise the administrative punishment free-right, and scientifically make a decision; the method provides a scientific and efficient solution for standardizing the application of the administrative punishment free right to prevent and solve the problem of misuse of the free right in administrative law enforcement.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of the steps of a method for rationality assessment of administrative penalty discretion based on big data;
FIG. 2 is a flow chart of a method for rationality assessment of administrative penalty discretion based on big data according to the present invention;
FIG. 3 is a schematic diagram of a device for evaluating rationality of administrative penalty discretion based on big data;
FIG. 4 is a schematic diagram of a hardware operating environment computer device according to the big data based administrative penalty discretionary rationality assessment method of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, based on the examples herein, which are within the scope of the invention as defined by the claims, will be within the scope of the invention as defined by the claims.
The following describes a specific big data-based administrative penalty discretionary right rationality assessment method and device provided in the embodiment of the present application with reference to the accompanying drawings.
As shown in fig. 1, the administrative penalty discretion right rationality assessment method based on big data provided in the embodiment of the application includes:
s101, acquiring basic data;
the database stores initial data, the initial data in the database is firstly obtained, and the initial data is preprocessed, namely the initial data is cleaned, so that basic data are obtained.
S102, extracting features from the basic data, determining a case basis according to the extracted feature information, determining a punishment amount range according to the case basis, and judging whether the punishment amount of the case is within the punishment amount range;
And extracting characteristics from the basic data, and determining the scheme of the case according to the characteristics. After determining the case clearance, determining the punishment amount range according to the relevant laws and regulations through the case clearance, judging whether the punishment amount of the case is in the punishment amount range, if not, directly judging that the case is unreasonable, and ending the whole process, and if yes, executing the next step S103.
S103, after the fact that the punishment amount of the case is within the punishment amount range is judged, a punishment amount rationality evaluation model is built, and the rationality of the punishment amount of the case is judged according to the punishment amount rationality evaluation model.
And outputting a result on the rationality of the case punishment amount through the punishment amount rationality evaluation model.
The administrative punishment free-cutting right rationality assessment method based on big data has the working principle that: acquiring basic data; extracting features from the basic data, determining a case counter according to the extracted feature information, determining a punishment amount range according to the case counter, and judging whether the punishment amount of the case is within the punishment amount range; after the fact that the punishment amount of the case is within the punishment amount range is judged, a punishment amount rationality evaluation model is built, and the rationality of the punishment amount of the case is judged according to the punishment amount rationality evaluation model.
In some embodiments, referring to fig. 2, obtaining the base data includes:
acquiring initial data in a database;
and cleaning the initial data to obtain basic data.
In some embodiments, feature extraction is performed on the basic data, and case scheme determination is performed according to the extracted feature information, including:
Extracting basic characteristics of basic data by adopting an information extraction technology;
and determining the case law of the case according to the extracted basic characteristics.
Preferably, the determining, according to the case, the range of the penalty amount, and judging whether the penalty amount of the case is within the range of the penalty amount, includes:
a range of penalty amounts specified by the determined laws and regulations according to the scheme;
judging whether the penalty amount of the case is within the range of the penalty amount specified by the laws and regulations.
Specifically, a case law of a case is found by extracting basic characteristics of basic data, and whether the basic data related to the case is in the range is judged according to the case law by determining the punishment amount range; specifically, the case penalty amount is preliminarily determined to be within the range of the penalty amount specified by the law and regulation. And preliminarily judging that the punishment amount of the case is not in the punishment amount range defined by the laws and regulations. If the amount is not within the range of the penalty amount, the judgment is unreasonable, and if the amount is within the range of the penalty amount, the next step is carried out.
Specifically, when judging the rationality of the penalty amount of the case, firstly, finding the range of the penalty amount specified in the law and regulation according to the law and regulation, judging whether the penalty amount of the case is within the range of the specified penalty amount, if not, directly judging as unreasonable, and if so, continuing to judge in the next step, namely judging whether the penalty amount is reasonable according to the rationality evaluation model.
In some embodiments, the range of penalty amounts specified by the determined laws and regulations according to the scheme includes:
Determining punishment basis according to the basic characteristics of the cases;
And determining a punishment amount range specified by laws and regulations according to punishment basis.
Specifically, basic characteristics of the case are extracted, a penalty basis is determined, and whether the penalty amount is reasonable or not is judged according to the amount range related to the penalty basis, wherein the penalty basis is related legal regulations.
In some embodiments, constructing a penalty amount rationality assessment model and determining rationality of the penalty amount for the case based on the penalty amount rationality assessment model comprises:
model training is carried out on the class case;
judging whether free judge of punishment amount exists according to punishment basis of the case; if so, acquiring similar cases according to the case content through a case similarity judging model, and judging whether the punishment amount is reasonable according to the similarity degree of the similar cases; otherwise, calculating a distribution function of the administrative punishment amount, and judging whether the punishment amount of the case meets the distribution.
Specifically, model training is performed on the class case, including:
(1) Obtaining case data, obtaining non-case data in the same proportion, formatting the case data and the non-case data, and constructing a formatted data set; the same ratio is 1:1, the application obtains case data according to information crawled on the Internet, and the case data is obtained according to 1:1, then carrying out formatting processing on the data to construct a formatted data set.
(2) The formatted data set comprises a training data set and a test data set, and the training data set is divided into N parts; specifically, the formatted data is randomly divided into two parts, namely a training data set and a test data set, wherein the ratio of the training data set to the test data set is 8:2, the training data set is divided into N parts after being disturbed, so that N times of cross verification can be conveniently carried out later, and the training data set is used for evaluating the performance of the model;
(3) Word2vec is used for mapping word vectors, training data is input into BiLSTM models for training, and results are output; the BiLSTM model comprises a word vector embedding layer, a bidirectional LSTM layer, an attention mechanism layer and a classifier; the word vector embedding layer is used for converting training data into word vectors; the bidirectional LSTM layer takes word vectors as input, converts the word vectors and outputs feature vectors with fixed lengths; the attention mechanism layer takes the feature vector output by the bidirectional LSTM layer as input, and adds different weights according to the importance degree of words in the text to obtain more accurate text representation; the classification layer normalizes attention the weights with softmax, and performs weighted summation and then performs similarity calculation of the cosine with the parameters. The word2vec is selected by two methods, namely Skip-gram and CBOW, and the technical scheme of the application adopts the Skip-gram method. Compared with the traditional one-hot vector, the Skip-gram method can enable the generated vector to have semantic information, and word context information is considered, so that accuracy of the method is improved greatly. The BiLSTM model considers that sentences should not only consider one direction, but also should include reverse information, which is more complete for the representation of text. The attention mechanism can focus attention on important points, while ignoring other unimportant factors, helping to extract similar important information in the case.
(4) Obtaining an optimal model through N times of cross verification, and storing the optimal model;
(5) And loading an optimal model for testing, inputting test data in a test data set into the optimal model, sequentially performing word segmentation, word vector mapping and class similarity calculation, and evaluating the performance of the model.
As shown in fig. 3, an embodiment of the present application provides a device for evaluating rationality of administrative penalty discretion based on big data, including:
an acquisition module 301, configured to acquire basic data;
the extracting module 302 is configured to perform feature extraction on the basic data, determine a case routing of a case according to the extracted feature information, and determine a penalty amount range according to the case routing, so as to determine whether the penalty amount of the case is within the penalty amount range;
And the judging module 303 is configured to construct a punishment amount rationality evaluation model after judging that the punishment amount of the case is within the punishment amount range, and judge the rationality of the punishment amount of the case according to the punishment amount rationality evaluation model.
Preferably, the judging module includes:
the training unit is used for carrying out model training on the class case;
The calculation module is used for judging whether free amount of punishment exists according to punishment basis of the cases; if so, acquiring similar cases according to the case content through a case similarity judging model, and judging whether the punishment amount is reasonable according to the similarity degree of the similar cases; otherwise, calculating a distribution function of the administrative punishment amount, and judging whether the punishment amount of the case meets the distribution.
The application provides a big data-based administrative punishment free-right rationality assessment device, which is based on the working principle that an acquisition module 301 is used for acquiring basic data; the extraction module 302 is configured to perform feature extraction on the basic data, determine a case routing of a case according to the extracted feature information, and determine a penalty amount range according to the case routing, so as to determine whether the penalty amount of the case is within the penalty amount range; the judging module 303 is configured to construct a penalty amount rationality evaluation model after judging that the penalty amount of the case is within the penalty amount range, and judge the rationality of the penalty amount of the case according to the penalty amount rationality evaluation model.
As shown in fig. 4, the present application provides a computer device including: the memory and processor may also include a network interface, the memory storing a computer program, the memory may include volatile memory, random Access Memory (RAM) and/or nonvolatile memory in a computer readable medium, such as Read Only Memory (ROM) or flash RAM. The computer device stores an operating system, with memory being an example of a computer-readable medium. The computer program, when executed by a processor, causes the processor to perform a method for discretionary evaluation of administrative penalties based on big data, the structure shown in fig. 4 is merely a block diagram of a part of the structure related to the inventive solution and does not constitute a limitation of the computer device to which the inventive solution is applied, and a specific computer device may include more or less components than those shown in the drawings, or may combine some components, or have a different arrangement of components.
In one embodiment, the big data based administrative penalty discretion rationality assessment method provided by the present application may be implemented in the form of a computer program that is executable on a computer device as shown in FIG. 4.
In some embodiments, the computer program, when executed by the processor, causes the processor to perform the steps of: acquiring basic data; extracting features of the basic data, and determining the scheme of the case according to the extracted feature information; judging whether the case is in accordance with the basic data or not, and judging the rationality of the punishment amount of the case after judging the case is in accordance.
The present application also provides a computer storage medium, examples of which include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassette storage or other magnetic storage devices, or any other non-transmission medium, that can be used to store information that can be accessed by a computing device.
In some embodiments, the present invention also proposes a computer readable storage medium storing a computer program, which when executed by a processor causes the processor to perform the steps of: acquiring basic data; extracting features of the basic data, and determining the scheme of the case according to the extracted feature information; judging whether the case is in accordance with the basic data or not, and judging the rationality of the punishment amount of the case after judging the case is in accordance.
In summary, the application provides a big data-based administrative penalty free-play right rationality assessment method, which comprises the steps of obtaining basic data; extracting features of the basic data, and determining the scheme of the case according to the extracted feature information; judging whether the case is met with the basic data or not, and judging the rationality of the punishment amount of the case after judging the meeting. The application establishes a rationality evaluation method and a rationality evaluation device for the free right of judgment, can assist law enforcement personnel to accurately exercise the administrative punishment free right of judgment, scientifically and scientifically make a decision, and provides a scientific and efficient solution for solving the problem of misuse of the free right of judgment in the administrative law enforcement for standardizing the application of the administrative punishment free right of judgment.
It can be understood that the above-provided method embodiments correspond to the above-described apparatus embodiments, and corresponding specific details may be referred to each other and will not be described herein.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, magnetic disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (6)
1. The administrative punishment discretionary right rationality assessment method based on big data is characterized by comprising the following steps:
acquiring basic data;
Extracting features from the basic data, determining a case counter according to the extracted feature information, determining a punishment amount range according to the case counter, and judging whether the punishment amount of the case is within the punishment amount range;
After judging that the punishment amount of the case is in the punishment amount range, constructing a punishment amount rationality evaluation model and judging the rationality of the punishment amount of the case according to the punishment amount rationality evaluation model;
the penalty amount rationality assessment model determines the rationality of the penalty amount for the case, including:
model training is carried out on the class case;
Judging whether free judge of punishment amount exists according to punishment basis of the case; if so, acquiring similar cases according to the case content through a case similarity judging model, and judging whether the punishment amount is reasonable according to the similarity degree of the similar cases; otherwise, calculating a distribution function of the administrative punishment amount, and judging whether the punishment amount of the case meets the distribution; the distribution function can assist law enforcement officers to accurately exercise administrative punishment free discretion rights;
the model training of the class case comprises the following steps:
Obtaining case data, obtaining non-case data in the same proportion, formatting the case data and the non-case data, and constructing a formatted data set;
the formatted data set comprises a training data set and a test data set, and the training data set is divided into N parts;
word2vec is used for mapping word vectors, the word vector is input into BiLSTM models for training, and a result is output; the word2vec selects a Skip-gram method, the generated vector can have semantic information through the Skip-gram method, and context information of words is considered;
Obtaining an optimal model through N times of cross validation, and storing the optimal model;
loading the optimal model for testing, inputting test data in a test data set into the optimal model, sequentially performing word segmentation, word vector mapping and class similarity calculation, and evaluating the performance of the model;
The BiLSTM model comprises a word vector embedding layer, a bidirectional LSTM layer, an attention mechanism layer and a classifier; the word vector embedding layer is used for converting training data into word vectors; the bidirectional LSTM layer takes word vectors as input, converts the word vectors and outputs feature vectors with fixed lengths; the attention mechanism layer takes the feature vector output by the bidirectional LSTM layer as input, and adds different weights according to the importance degree of words in the text to obtain more accurate text representation; the classifying layer normalizes attention the weight by softmax, and performs weighted summation and then performs similarity calculation of the cosine with the parameters; the sentences in the BiLSTM model can not only consider information of one direction, but also consider reverse information.
2. The method of claim 1, wherein the obtaining the base data comprises:
acquiring initial data in a database;
and cleaning the initial data to obtain basic data.
3. The method of claim 1, wherein the feature extraction is performed on the base data, and determining the case scheme based on the extracted feature information comprises:
Extracting basic characteristics of basic data by adopting an information extraction technology;
and determining the case law of the case according to the extracted basic characteristics.
4. The method of claim 1, wherein determining, based on the case routing, a range of penalty amounts, whether a penalty amount for a case is within the range of penalty amounts, comprises:
a range of penalty amounts specified by the determined laws and regulations according to the scheme;
judging whether the penalty amount of the case is within the range of the penalty amount specified by the laws and regulations.
5. The method of claim 4, wherein the range of penalty amounts specified by the determined legal regulations according to the proposal comprises:
Determining punishment basis according to the basic characteristics of the cases;
and determining a punishment amount range specified by laws and regulations according to the punishment basis.
6. An administrative penalty discretionary right rationality assessment device based on big data, comprising:
The acquisition module is used for acquiring the basic data;
the extraction module is used for carrying out feature extraction on the basic data, determining a case counter according to the extracted feature information, determining a punishment amount range according to the case counter, and judging whether the punishment amount of the case is in the punishment amount range or not;
The judging module is used for constructing a punishment amount rationality evaluation model after judging that the punishment amount of the case is in the punishment amount range and judging the rationality of the punishment amount of the case according to the punishment amount rationality evaluation model;
The judging module comprises:
the training unit is used for carrying out model training on the class case;
The calculation module is used for judging whether free amount of punishment exists according to punishment basis of the cases; if so, acquiring similar cases according to the case content through a case similarity judging model, and judging whether the punishment amount is reasonable according to the similarity degree of the similar cases; otherwise, calculating a distribution function of the administrative punishment amount, and judging whether the punishment amount of the case meets the distribution; the distribution function can assist law enforcement officers to accurately exercise administrative punishment free discretion rights;
the model training of the class case comprises the following steps:
Obtaining case data, obtaining non-case data in the same proportion, formatting the case data and the non-case data, and constructing a formatted data set;
the formatted data set comprises a training data set and a test data set, and the training data set is divided into N parts;
word2vec is used for mapping word vectors, the word vector is input into BiLSTM models for training, and a result is output; the word2vec selects a Skip-gram method, the generated vector can have semantic information through the Skip-gram method, and context information of words is considered;
Obtaining an optimal model through N times of cross validation, and storing the optimal model;
loading the optimal model for testing, inputting test data in a test data set into the optimal model, sequentially performing word segmentation, word vector mapping and class similarity calculation, and evaluating the performance of the model;
The BiLSTM model comprises a word vector embedding layer, a bidirectional LSTM layer, an attention mechanism layer and a classifier; the word vector embedding layer is used for converting training data into word vectors; the bidirectional LSTM layer takes word vectors as input, converts the word vectors and outputs feature vectors with fixed lengths; the attention mechanism layer takes the feature vector output by the bidirectional LSTM layer as input, and adds different weights according to the importance degree of words in the text to obtain more accurate text representation; the classifying layer normalizes attention the weight by softmax, and performs weighted summation and then performs similarity calculation of the cosine with the parameters; the sentences in the BiLSTM model can not only consider information of one direction, but also consider reverse information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010273409.6A CN111461932B (en) | 2020-04-09 | Administrative punishment free-cutting right rationality evaluation method and device based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010273409.6A CN111461932B (en) | 2020-04-09 | Administrative punishment free-cutting right rationality evaluation method and device based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111461932A CN111461932A (en) | 2020-07-28 |
| CN111461932B true CN111461932B (en) | 2024-05-14 |
Family
ID=
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7961956B1 (en) * | 2009-09-03 | 2011-06-14 | Thomas Cecil Minter | Adaptive fisher's linear discriminant |
| CN107818138A (en) * | 2017-09-28 | 2018-03-20 | 银江股份有限公司 | A kind of case legal regulation recommends method and system |
| CN107918921A (en) * | 2017-11-21 | 2018-04-17 | 南京擎盾信息科技有限公司 | Criminal case court verdict measure and system |
| CN109582950A (en) * | 2018-09-25 | 2019-04-05 | 南京大学 | A kind of judgement document argues appraisal procedure |
| WO2019069865A1 (en) * | 2017-10-03 | 2019-04-11 | 日本電気株式会社 | Parameter estimation system, parameter estimation method, and parameter estimation program recording medium |
| CN110335181A (en) * | 2019-07-05 | 2019-10-15 | 北京梦天门科技股份有限公司 | A kind of Health Inspection discretion algorithm based on artificial intelligence |
| CN110457479A (en) * | 2019-08-12 | 2019-11-15 | 贵州大学 | A kind of judgement document's analysis method based on criminal offence chain |
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7961956B1 (en) * | 2009-09-03 | 2011-06-14 | Thomas Cecil Minter | Adaptive fisher's linear discriminant |
| CN107818138A (en) * | 2017-09-28 | 2018-03-20 | 银江股份有限公司 | A kind of case legal regulation recommends method and system |
| WO2019069865A1 (en) * | 2017-10-03 | 2019-04-11 | 日本電気株式会社 | Parameter estimation system, parameter estimation method, and parameter estimation program recording medium |
| CN107918921A (en) * | 2017-11-21 | 2018-04-17 | 南京擎盾信息科技有限公司 | Criminal case court verdict measure and system |
| CN109582950A (en) * | 2018-09-25 | 2019-04-05 | 南京大学 | A kind of judgement document argues appraisal procedure |
| CN110335181A (en) * | 2019-07-05 | 2019-10-15 | 北京梦天门科技股份有限公司 | A kind of Health Inspection discretion algorithm based on artificial intelligence |
| CN110457479A (en) * | 2019-08-12 | 2019-11-15 | 贵州大学 | A kind of judgement document's analysis method based on criminal offence chain |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108874768B (en) | A kind of e-commerce falseness comment recognition methods based on theme emotion joint probability | |
| CN112860841B (en) | Text emotion analysis method, device, equipment and storage medium | |
| JP2011048822A5 (en) | ||
| CN110069545B (en) | Behavior data evaluation method and device | |
| CN112329816A (en) | Data classification method and device, electronic equipment and readable storage medium | |
| CA3104242C (en) | Systems and methods for determining structured proceeding outcomes | |
| CN111199157B (en) | Text data processing method and device | |
| KR20210092979A (en) | Fake news artificial intelligence classification system, and method thereof | |
| CN106649686A (en) | User interest group dividing method and system based on multilayer potential features | |
| Behera et al. | Machine learning approach for reliability assessment of open source software | |
| Masciari et al. | Detecting fake news by image analysis | |
| CN110990523A (en) | Legal document determining method and system | |
| CN117409419A (en) | Image detection method, device and storage medium | |
| CN109492401B (en) | Content carrier risk detection method, device, equipment and medium | |
| CN111461932B (en) | Administrative punishment free-cutting right rationality evaluation method and device based on big data | |
| CN116450848B (en) | Method, device and medium for evaluating computing thinking level based on event map | |
| CN112070559A (en) | State acquisition method and device, electronic equipment and storage medium | |
| Joshi et al. | Fraud App Detection of Google Play Store Apps Using Decision Tree | |
| Nam et al. | Effects of pre-trained word embeddings on text-based deception detection | |
| Singh et al. | Twitter sentiment analysis for stock prediction | |
| CN110738032A (en) | method and device for generating judgment sections of official paperwork institute | |
| CN113553063A (en) | Method and device for determining contribution degree of open source of code, computer equipment and medium | |
| CN111813964B (en) | Data processing method based on ecological environment and related equipment | |
| CN111461932A (en) | Administrative punishment discretion rationality assessment method and device based on big data | |
| CN111949783A (en) | Question and answer result generation method and device in knowledge base |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |