CN112749826A - Criminal period prediction method, device, storage medium and equipment - Google Patents
Criminal period prediction method, device, storage medium and equipment Download PDFInfo
- Publication number
- CN112749826A CN112749826A CN201911053891.6A CN201911053891A CN112749826A CN 112749826 A CN112749826 A CN 112749826A CN 201911053891 A CN201911053891 A CN 201911053891A CN 112749826 A CN112749826 A CN 112749826A
- Authority
- CN
- China
- Prior art keywords
- criminal
- law
- target
- money
- case information
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 71
- 238000003860 storage Methods 0.000 title claims abstract description 20
- 230000006978 adaptation Effects 0.000 claims abstract description 140
- 238000009826 distribution Methods 0.000 claims description 62
- 238000004891 communication Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 12
- 238000013528 artificial neural network Methods 0.000 abstract description 5
- 238000004088 simulation Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 15
- 238000004590 computer program Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 8
- 238000012545 processing Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 238000007476 Maximum Likelihood Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000001364 causal effect Effects 0.000 description 2
- 238000009795 derivation Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/18—Legal services
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Marketing (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Technology Law (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a criminal period prediction method, a criminal period prediction device, a storage medium and equipment, wherein target case information is obtained, a target money is determined according to the adaptation degree of a law clause and the target case information, and then a target criminal period is determined according to the adaptation degree of the criminal period and the determined target money. The adaptation degree of the law clause and the target case information represents the degree of the law clause applicable to the target case information, the adaptation degree of the law clause and the determined target money represents the degree of the law clause applicable to the target case information, and the adaptation degree of the law clause and the determined target case is equivalent to the simulation of judicial judgment logic, so that the law clause prediction method has judicial interpretability, the accuracy of the law clause prediction method is equivalent to the accuracy of a neural network-based law clause prediction method, namely, the basic application scheme realizes the law interpretability of the law clause prediction process on the basis of ensuring the prediction accuracy, and improves the practicability of the law clause prediction.
Description
Technical Field
The present invention relates to the field of information processing technology, and more particularly, to a criminal term prediction method, apparatus, storage medium, and device.
Background
The certainty of law, the predictability of the consequences of behavior, are fundamental features of the society of law enforcement. Therefore, the criminal period prediction problem gradually becomes an important link for judging document information mining and analysis, the criminal period prediction aims at predicting the criminal period to be treated based on criminal fact description information, and the criminal period prediction method has great development space in auxiliary judicial judgment and legal consultation.
The current criminal phase prediction method is realized based on the natural language processing technology of the neural network, but the neural network model lacks certain interpretability due to the inherent black box modeling form of the neural network model. Only by using criminal facts as input and criminal period as output, the judging logic from criminal facts to criminal periods cannot be shown, and thus the method cannot be widely applied.
Disclosure of Invention
In view of the above, the present invention provides a criminal prediction method, apparatus, storage medium and device that overcomes or at least partially solves the above mentioned problems.
The embodiment of the invention provides the following scheme:
a criminal phase prediction method comprising:
acquiring target case information, wherein the target case information represents whether each judicial element related to the criminal phase exists in the case of the criminal phase to be predicted or not;
for each law related to the criminal period, determining a target money in the law according to the adaptation degree of each money in the law and the target case information;
and determining the target criminal stage according to the adaptation degree of each criminal stage and all the determined target money in the criminal stage value range.
Preferably, the determining the target money in the law sentence according to the adaptation degree of each money in the law sentence and the target case information includes:
determining the adaptation degree of each item in the law and the target case information according to the pre-learned adaptation degree distribution condition of the law items and the case information;
and determining the money with the maximum adaptation degree with the target case information in the law as the target money in the law.
In the method, preferably, the distribution of the degree of adaptation between the previously learned law clause and case information is as follows:
multiplying the probability distribution of the law and the case information learned according to the historical judicial cases by the continuous multiplication result of the probabilities of all judicial elements related to the criminal period existing in the case to obtain the adaptation degree distribution condition of the law and the case information;
wherein the probability of each judicial element existing in the case is as follows: the ratio of the number of historical judicial cases for which the judicial element exists to the total number of historical judicial cases.
Preferably, the determining the target criminal stage according to the matching degree of each criminal stage and all the determined target money in the criminal stage value range includes:
determining the adaptation degree of each criminal phase and all determined target money in the criminal phase value range according to the pre-learned adaptation degree distribution condition of the criminal phase and the legal fund;
and determining the penalty period with the maximum fitting degree with all the determined target money as the target penalty period.
In the method, preferably, the distribution of the fitting degree of the forelearned criminal phase and the law is as follows:
multiplying the probability distribution of the criminal period learned according to the historical judicial cases under the condition of the law clause by the continuous multiplication result of the probability of each item in the law clause related to the criminal period existing in the historical judicial cases to obtain the distribution condition of the adaptation degree of the criminal period and the law clause;
wherein, the probability that each style exists in the historical judicial case is as follows: the frequency with which the money occurs in historical jurisdictions.
The above method, preferably, further comprises:
outputting the adaptation degree of each target money and the target case information at least;
and/or the presence of a gas in the gas,
and outputting the adaptation degree of each criminal period and all determined target money in the criminal period value range.
The above method, preferably, said outputting at least a degree of adaptation of each of the individual laws related to criminal phase to said target case information; and/or outputting the adaptation degree of each criminal phase and all determined target money in the criminal phase value range, wherein the adaptation degree comprises the following steps:
outputting the adaptation degree of each money in each law related to the criminal phase and the target case information in the form of a node map;
and/or the presence of a gas in the gas,
and outputting the adaptation degree of each criminal term and all determined target money in the criminal term value range in a node map mode.
A criminal phase prediction device comprising:
the system comprises an acquisition module, a prediction module and a prediction module, wherein the acquisition module is used for acquiring target case information, and the target case information represents whether each judicial element related to the criminal phase exists in the case of the criminal phase to be predicted or not;
the first prediction module is used for determining a target money in each law related to the criminal period according to the adaptation degree of each money in the law and the target case information;
and the second prediction module is used for determining the target criminal stage according to the adaptation degree of each criminal stage and all the determined target money in the criminal stage value range.
A storage medium comprising a stored program, wherein said program performs a criminal term prediction method as claimed in any one of the preceding claims.
An apparatus comprising at least one processor, and at least one memory connected to the processor, a bus; the processor and the memory complete mutual communication through the bus; said processor is adapted to invoke program instructions in said memory to perform a penalty prediction method as described in any of the above.
By means of the technical scheme, the criminal period prediction method, the device, the storage medium and the equipment provided by the invention are used for acquiring the target case information, determining the target money according to the adaptation degree of the law clause and the target case information, and then determining the target criminal period according to the criminal period and the determined adaptation degree of the target money. The adaptation degree of the law clause and the target case information represents the degree of the law clause applicable to the target case information, the adaptation degree of the law clause and the determined target money represents the degree of the law clause applicable to the target case information, and the adaptation degree of the law clause and the determined target case is equivalent to the simulation of judicial judgment logic, so that the law clause prediction method has judicial interpretability, the accuracy of the law clause prediction method is equivalent to the accuracy of a neural network-based law clause prediction method, namely, the basic application scheme realizes the law interpretability of the law clause prediction process on the basis of ensuring the prediction accuracy, and improves the practicability of the law clause prediction.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 shows a flow chart of one implementation of the criminal phase prediction method provided by the embodiment of the invention.
Fig. 2 illustrates a network structure of a bayesian network provided by an embodiment of the present invention;
FIG. 3 illustrates another network structure of a Bayesian network provided by an embodiment of the present invention;
FIG. 4 illustrates another network structure of a Bayesian network provided by an embodiment of the present invention;
FIG. 5 illustrates an exemplary diagram of a node map provided by embodiments of the invention;
fig. 6 shows a schematic structural diagram of a criminal phase prediction device provided by an embodiment of the invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
An implementation flow chart of the criminal phase prediction method provided by the embodiment of the invention is shown in fig. 1, and may include:
step S11: and acquiring target case information, wherein the target case information represents whether each judicial element related to the criminal phase exists in the case of the criminal phase to be predicted.
In the embodiment of the invention, a plurality of judicial factors related to the criminal phase are predefined, when the criminal phase adapted to a certain case (for convenience of description, referred to as case a) needs to be predicted, whether each judicial factor in the plurality of judicial factors exists in the case a or not can be judged, and the judgment result is the target case information of the case a. For example, if the plurality of judicial factors related to the criminal period are less than 80mg/ml, and the vehicle runs at a speed seriously exceeding a specified speed, and drives three cases after pursuing (it should be noted that this is only an exemplary description here, and does not constitute a limitation to the present invention, and there are still more judicial factors in the actual situation, which is not illustrated here), the target case information of case a may be one of the following eight case information based on the three judicial factors related to the criminal period (less than 80mg/ml, runs at a speed seriously exceeding a specified speed, and drives after): the case a is characterized by having a judicial element of 80mg/ml or less, having no judicial element of traveling at a speed of more than a predetermined speed, and having no judicial element of traveling at a speed of more than a predetermined speed, or having no judicial element of traveling at a speed of more than a predetermined speed.
Wherein various judicial elements related to the criminal phase can be determined by:
for each kind of criminal name needing to be forecasted in a criminal period, case data related to the criminal name is selected from historical cases, namely cases finally critted with the criminal name are selected from the historical cases. Taking dangerous driving guilty as an example, all cases taking the dangerous driving guilty as a final trial name are extracted from historical cases. Other case data related to the crime name are selected in the same manner as described above, and are not listed here.
All the laws applicable to the criminal name are found in the extracted cases, laws irrelevant to the criminal period (such as general laws of the criminal law, such as penalty or the behavior ability of persons responsible for accidents) are removed from the cases, and the laws relevant to the criminal period are only reserved. For example, the law associated with dangerous driving guilt includes one of 133 th, 52 th, 53 th, and 66 th, 67 th in criminal law, wherein 52 th, 53 th are rejected because they are associated with penalties and are not associated with criminal periods, and the law associated with dangerous driving guilt only retains one of 133 th and 66 th, 67 th associated with criminal periods.
Extracting judicial elements according to the act related to the criminal phase, wherein the extracted judicial elements are independent from each other. This step may be implemented in conjunction with the knowledge of a judicial expert. Taking one of the 133 th article of criminal law as an example, the following four judicial factors can be extracted according to one of the 133 th article of criminal law:
pursuing and competing for driving, and the plot is severe;
drunk driving of a motor vehicle;
the vehicle is engaged in school bus business or passenger transportation, and the vehicle can carry passengers seriously exceeding the rated speed or can run at a speed seriously exceeding the specified speed per hour;
dangerous chemicals are transported in violation of dangerous chemical safety management regulations, and public safety is endangered.
With the assistance of the expert in the judicial field, each judicial element can be further refined, for example, drunk driving can be further subdivided into the following three specific elements: 80mg/100ml below; 80-200mg/100 ml; 200mg/100ml or more. In the embodiment of the present application, the judicial factors related to the criminal phase refer to the refined judicial factors.
Step S12: for each law related to the criminal period (for convenience of description, denoted as law F), the target money in the law F is determined according to the adaptation degree of each money in the law F and the target case information.
In the criminal law act, each act usually contains a plurality of terms, one term being a money. The adaptation degree of the money and the case information represents the degree of the money suitable for the target case information, so the adaptation degree of the money and the target case information represents the degree of the money suitable for the target case information. The target money is the money of which the adaptation degree with the target case information in the law F meets the condition. Optionally, the target money may be the money with the largest adaptation degree with the target case information in the law clause F.
Optionally, if there are at least two items whose adaptation degrees with the target case information satisfy the condition, a target item may be selected from the at least two items whose adaptation degrees with the target case information satisfy the condition according to a preset rule.
Since each criminal term-related law determines a target money, the number of target money is equal to the number of criminal term-related laws when the target money is determined in all criminal term-related laws.
In addition, considering that in practical application, case a may only relate to a part of law related to criminal period, and for the law related to criminal period which case a does not relate to, each item in the law related to criminal period and case a should not be applicable, in the embodiment of the present application, the item of each law includes a custom item besides the item specified in the criminal law, and is marked as "none" for convenience of description. If the target money is 'none', the case A is not related to the law, namely the law is not applicable to the case A.
Step S12: and determining the target criminal stage according to the adaptation degree of each criminal stage and all the determined target money in the criminal stage value range.
The criminal period value range is all possible criminal periods of all criminal names needing to be predicted in the criminal period, the criminal period in the criminal period value range is a discrete value, and usually a natural number taking a month as a unit is used.
The degree of adaptation of the criminal period to all target money determined refers to the degree of adaptation of the combination of the criminal period and all target money determined, which characterizes the degree to which the criminal period applies to the combination of all target money. The target criminal phase is the criminal phase in which the adaptation degrees with all target money in the criminal phase value range meet the conditions. Alternatively, the target penalty period may be the penalty period with the greatest fit to the set of all target funds.
The target criminal phase is the criminal phase for which the case for which the criminal phase is to be predicted is predicted.
According to the criminal period prediction method provided by the embodiment of the application, the target case information is obtained, the target money is determined according to the adaptation degree of the law clause and the target case information, and then the target criminal period is determined according to the criminal period and the determined adaptation degree of the target money. The adaptation degree of the law clause and the target case information represents the degree of the law clause applicable to the target case information, the adaptation degree of the law clause and the determined target money represents the degree of the law clause applicable to the target case information, and the adaptation degree of the law clause and the determined target case is equivalent to the simulation of judicial judgment logic, so that the law clause prediction method has judicial interpretability, the accuracy of the law clause prediction method is equivalent to the accuracy of a neural network-based law clause prediction method, namely, the basic application scheme realizes the law interpretability of the law clause prediction process on the basis of ensuring the prediction accuracy, and improves the practicability of the law clause prediction.
In an optional embodiment, the adaptation degree of each item in the law F and the target case information may be determined according to the pre-learned adaptation degree distribution of the law item and the case information.
The fit degree distribution of the style and case information in the law statement F can be obtained based on bayesian network (for convenience of description, it is recorded as the first bayesian network) learning.
As shown in fig. 2, a schematic structural diagram of a first bayesian network provided for an embodiment of the present application is shown, where a parent node in the first bayesian network is each judicial element related to the criminal phase, each judicial element related to the criminal phase is a parent node, child nodes in the first bayesian network are laws related to the criminal phase, each law related to the criminal phase is a child node, each judicial element related to the criminal phase is a parent node of all laws related to the criminal phase, and the first bayesian network embodies a causal relationship between a combination of the judicial elements and a money.
For a case, the combination of judicial factors characterizes whether or not each judicial factor related to a criminal phase appears in the case. As shown in fig. 2, there are n judicial factors related to the criminal phase, and each judicial factor has two possible values: and if the combination of the judicial factors related to the criminal period is 2 in total, namely, if the value of the judicial factor is yes, the situation that the judicial factor appears is shown, and if the value of the judicial factor is no, the situation that the judicial factor does not appear is shownnAnd (3) a situation. Each case corresponds to one case information, and the adaptation degree of each case information and each law statement form the distribution condition of the adaptation degree of the law statement and the case information.
In the foregoing embodiment, the extraction process of judicial factors related to criminal periods is described, and the learning process of the adaptation degree distribution of the law clauses and case information is described below.
In the embodiment of the application, the adaptation degree of the law clause and the case information is represented by the joint probability distribution of the law clause and the case information. According to the incidence relation between the parent nodes and the child nodes in the Bayesian network, the following can be obtained:
wherein, P (T ═ a, X)1,...,Xn) Shows the money a and case information (X) in the French T1,...,Xn) P (T ═ a | X) is used as a joint probability1,...,Xn) Shows that the money a in the French T is known (X)1,...,Xn) Probability under the value of (A), P (X)i) Represents judicial element XiThe probability of (c).
The maximum likelihood method is then used to estimate each parent node (i.e., the judicial element X) in the first Bayesian networki) Probability distribution of (2). In particular, the method comprises the following steps of,
firstly, obtaining each father node (marked as X) through the extracted history casesi) Likelihood function of (d):
wherein D is the set of the extracted whole history cases, and the number of cases in the set is h, DjDenotes the jth case in the set, Xi(dj) Shows the case of the jth case at the judicial element XiValue of (Y or N), P (X)i(dj) Theta) represents the parameter theta in the probability distribution PiCase j, in the known casejMiddle judicial element XiValue of Xi(dj) Is a parameter θ relating to the distribution P of the judicial elementsiThe probability of the next occurrence of a certain set of history cases D. Since D is known, it is necessary to find some θiMaximizing the probability of D occurring.
Since the judicial elements serving as the father nodes in the first bayesian network are all binary values, the probability distributions of the judicial elements can be represented by bernoulli distributions, and therefore, the parameter θiRepresenting judicial elements X under Bernoulli distributioniProbability of taking the value "yes". Thus, the likelihood function can be simplified to
Wherein h is the total number of the history cases in the set D, h1Is at judicial element XiThe value of (a) is the number of cases of (b).
And (3) carrying out derivation on the likelihood function, and making a derivation result equal to zero to obtain:
it is known from this that BernoulliUnder the distribution of profit, judicial factors XiThe probability of "yes" being the judicial element XiThe frequency of yes, so the following equation is:
and since the sub-node (i.e. the law) does not directly assume which probability distribution each term follows because the terms of part of the law are more than two, it is not suitable to use the maximum likelihood algorithm, but because the number of judicial cases is huge, the probability can be directly represented by the frequency of each term appearing in the history case under the condition that each judicial element is known by the central limit theorem, so that the sub-node (i.e. the law) can be obtained:
P(T＝a,X1,...,Xn)＝frequency(a|X1,...,Xn)
frequency(a|X1,...,Xn) Indicating that money a is known (X)1,...,Xn) The frequency at which the value of (a) is obtained.
In an optional embodiment, the fitting degree of each criminal term and all determined target money in the criminal term value range can be determined according to the pre-learned matching degree distribution condition of the criminal term and the legal fund.
The fitting degree distribution of the criminal period and the law fund can be obtained based on the Bayesian network (for convenience of description, the Bayesian network is recorded as a second Bayesian network) learning.
Fig. 3 is a schematic structural diagram of a second bayesian network provided in the embodiment of the present application. Father nodes in the second Bayesian network are law bars related to the criminal phase, each law bar related to the criminal phase is a father node, child nodes in the second Bayesian network are each criminal phase in the criminal phase value range, each criminal phase is a child node, and each law bar related to the criminal phase is all criminal phases in the criminal phase value rangeAnd (4) a parent node. The possible value of each law is the money included in the law, that is, each money included in the law constitutes one possible value of the law. This second bayesian network embodies the causal relationship between the combination of statutory terms and the criminal phase. Assuming that there are m criminal law rules, where the number of law rules contained in the ith (i ═ 1, 2, …, m) law Fi is Ki, the combination of law clauses shares the same total
And (3) a situation.
The learning process of the distribution condition of the adaptation degree of the criminal term and the law statement is the same as the learning process of the distribution condition of the adaptation degree of the law statement and the case information, and the details are not repeated here.
Specifically, the distribution situation of the fitting degree of the forelearned criminal phase and the law is as follows:
multiplying the probability distribution of the criminal period learned according to the historical judicial cases under the condition of the law clause by the continuous multiplication result of the probability of each item in the law clause related to the criminal period existing in the historical judicial cases to obtain the distribution condition of the adaptation degree of the criminal period and the law clause; the specific formula can be expressed as:
wherein, P (Term, T)1,...,Tm) Expressing the criminal Term and m law articles (T)1,...,Tm) Degree of adaptation of, TiRepresents the value of the ith law, P (T)i) And the probability of the ith law bar under the value is represented.
Wherein, the probability that each style exists in the historical judicial case is as follows: the frequency with which the money occurs in historical jurisdictions.
In the process of implementing the present application, if the penalty period is predicted directly based on a bayesian network, the network structure shown in fig. 4 needs to be considered for prediction, and in this case, all nodes (judicial elements and law rules) in the network need to be known, which requires both the judicial element nodes and the law rule nodes as model inputs, however, in a real scene, most of the time we can only obtain case descriptions, and do not know the specific law rules to which the case relates (although the judgment can be made by judicial experts through case descriptions, when the number of cases is large, only an approximate range can be determined, and the accuracy cannot be precise to which law rules), so the prediction accuracy is low. In addition, because the number of nodes is large, a large memory is needed when the joint probability distribution is stored, and a long time is needed when the distribution is called, so that the everywhere prediction efficiency is low. Based on this, the criminal phase prediction method provided by the embodiment of the application predicts the criminal phase based on two independent Bayesian networks, and compared with the method for predicting the criminal phase by using one Bayesian network, the method has the advantages of small memory requirement, high prediction efficiency, low requirement on users and no need of high judicial knowledge.
In an optional embodiment, in order to intuitively embody the judging logic, the adaptation degree of each target money and the target case information may be output, or the adaptation degree of each criminal term and all the determined target money within the criminal term value range may be output, or the adaptation degree of each target money and the target case information and the adaptation degree of each criminal term and all the determined target money within the criminal term value range are output.
In an optional embodiment, besides outputting the degree of adaptation between the target money and the target case information, the method can also output the degree of adaptation between other money and the target case information, so that more detailed trial logic can be embodied.
Alternatively, the above-mentioned suitability may be output in a predetermined format, for example, in a table form.
Furthermore, in order to embody the judging logic more intuitively, the adaptation degree of each money in each law related to the criminal period and the target case information can be output in a node diagram form;
and/or the presence of a gas in the gas,
and outputting the adaptation degree of each criminal term and all determined target money in the criminal term value range in a node map mode.
The nodes in the node map may be distributed hierarchically, wherein judicial elements are located in the same layer, french items are located in the same layer, and criminal periods are located in another layer. As shown in fig. 5, an exemplary diagram of a node map provided in the embodiment of the present application is shown, where judicial elements are located in a first layer, french articles are located in a second layer, and criminal periods are located in a third layer.
The judicial element nodes of the first layer are marked with the values (yes or no) of the judicial elements, the law node of the second layer is marked with a target law (such as one, three or none in fig. 5), and the criminal period nodes of the third layer are marked as target criminal periods.
Corresponding to the embodiment of the method, the embodiment of the present application further provides a penalty period prediction device, and a schematic structural diagram of the penalty period prediction device provided by the embodiment of the present application is shown in fig. 6, and the penalty period prediction device may include:
an acquisition module 61, a first prediction module 62 and a second prediction module 63; wherein the content of the first and second substances,
the acquisition module 61 is configured to acquire target case information, where the target case information represents whether each judicial element related to a criminal phase exists in a case of the criminal phase to be predicted;
the first prediction module 62 is configured to, for each law related to the criminal phase, determine a target money in the law according to the degree of adaptation of each money in the law to the target case information;
the second prediction module 63 is configured to determine a target criminal stage according to the adaptation degree of each criminal stage and all determined target money in the criminal stage value range.
The criminal phase prediction device obtains target case information, determines the target case according to the adaptation degree of the law clause and the target case information, and then determines the target criminal phase according to the criminal phase and the determined adaptation degree of the target case. The adaptation degree of the law clause and the target case information represents the degree of the law clause applicable to the target case information, the adaptation degree of the law clause and the determined target money represents the degree of the law clause applicable to the target case information, and the adaptation degree of the law clause and the determined target case is equivalent to the simulation of judicial judgment logic, so that the law clause prediction method has judicial interpretability, the accuracy of the law clause prediction method is equivalent to the accuracy of a neural network-based law clause prediction method, namely, the basic application scheme realizes the law interpretability of the law clause prediction process on the basis of ensuring the prediction accuracy, and improves the practicability of the law clause prediction.
In an alternative embodiment, the first prediction module 62 includes:
the first determining unit is used for determining the adaptation degree of each item in the law and the target case information according to the pre-learned adaptation degree distribution condition of the law items and the case information;
and the second determining unit is used for determining the money with the maximum adaptation degree with the target case information in the law as the target money in the law.
In an optional embodiment, the distribution of the degree of adaptation between the pre-learned law clause and case information is as follows:
multiplying the probability distribution of the law and the case information learned according to the historical judicial cases by the continuous multiplication result of the probabilities of all judicial elements related to the criminal period existing in the case to obtain the adaptation degree distribution condition of the law and the case information;
wherein the probability of each judicial element existing in the case is as follows: the ratio of the number of historical judicial cases for which the judicial element exists to the total number of historical judicial cases.
In an alternative embodiment, the second prediction module 63 may include:
the third determining unit is used for determining the adaptation degree of each criminal phase and all determined target money in the criminal phase value range according to the pre-learned adaptation degree distribution condition of the criminal phase and the legal fund;
and a fourth determination unit for determining a penalty period having the largest degree of fit with all the determined target money as a target penalty period.
In an alternative embodiment, the distribution of the fitting degrees of the forelearned criminal period and the law is as follows:
multiplying the probability distribution of the criminal period learned according to the historical judicial cases under the condition of the law clause by the continuous multiplication result of the probability of each item in the law clause related to the criminal period existing in the historical judicial cases to obtain the distribution condition of the adaptation degree of the criminal period and the law clause;
wherein, the probability that each style exists in the historical judicial case is as follows: the frequency with which the money occurs in historical jurisdictions.
In an optional embodiment, the apparatus may further include an output module, configured to:
outputting the adaptation degree of each target money and the target case information at least;
and/or the presence of a gas in the gas,
and outputting the adaptation degree of each criminal period and all determined target money in the criminal period value range.
In an optional embodiment, the output module may be specifically configured to:
outputting the adaptation degree of each money in each law related to the criminal phase and the target case information in the form of a node map;
and/or the presence of a gas in the gas,
and outputting the adaptation degree of each criminal term and all determined target money in the criminal term value range in a node map mode.
The criminal period predicting device comprises a processor and a memory, wherein the acquiring module 61, the first predicting module 62, the second predicting module 63 and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to be one or more, and the criminal phase prediction is realized by adjusting the kernel parameters.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
An embodiment of the present invention provides a storage medium having a program stored thereon, which when executed by a processor implements the criminal term prediction method.
An embodiment of the present invention provides a processor for running a program, wherein the criminal phase prediction method is executed when the program runs.
The embodiment of the invention provides equipment, which comprises a processor, a memory and a program which is stored on the memory and can run on the processor, wherein the processor executes the program and realizes the following steps:
acquiring target case information, wherein the target case information represents whether each judicial element related to the criminal phase exists in the case of the criminal phase to be predicted or not;
for each law related to the criminal period, determining a target money in the law according to the adaptation degree of each money in the law and the target case information;
and determining the target criminal stage according to the adaptation degree of each criminal stage and all the determined target money in the criminal stage value range.
Optionally, the determining the target money in the law according to the adaptation degree of each money in the law and the target case information includes:
determining the adaptation degree of each item in the law and the target case information according to the pre-learned adaptation degree distribution condition of the law items and the case information;
and determining the money with the maximum adaptation degree with the target case information in the law as the target money in the law.
Optionally, the distribution of the degree of adaptation between the pre-learned law clauses and case information is as follows:
multiplying the probability distribution of the law and the case information learned according to the historical judicial cases by the continuous multiplication result of the probabilities of all judicial elements related to the criminal period existing in the case to obtain the adaptation degree distribution condition of the law and the case information;
wherein the probability of each judicial element existing in the case is as follows: the ratio of the number of historical judicial cases for which the judicial element exists to the total number of historical judicial cases.
Optionally, the determining the target criminal stage according to the matching degree of each criminal stage and all determined target money in the criminal stage value range includes:
determining the adaptation degree of each criminal phase and all determined target money in the criminal phase value range according to the pre-learned adaptation degree distribution condition of the criminal phase and the legal fund;
and determining the penalty period with the maximum fitting degree with all the determined target money as the target penalty period.
Optionally, the distribution of the fitting degrees of the forelearned criminal phase and the law article is as follows:
multiplying the probability distribution of the criminal period learned according to the historical judicial cases under the condition of the law clause by the continuous multiplication result of the probability of each item in the law clause related to the criminal period existing in the historical judicial cases to obtain the distribution condition of the adaptation degree of the criminal period and the law clause;
wherein, the probability that each style exists in the historical judicial case is as follows: the frequency with which the money occurs in historical jurisdictions.
Optionally, the method further includes:
outputting the adaptation degree of each target money and the target case information at least;
and/or the presence of a gas in the gas,
and outputting the adaptation degree of each criminal period and all determined target money in the criminal period value range.
Optionally, the adaptation degree of each money in each law related to the criminal phase and the target case information is at least output; and/or outputting the adaptation degree of each criminal phase and all determined target money in the criminal phase value range, wherein the adaptation degree comprises the following steps:
outputting the adaptation degree of each money in each law related to the criminal phase and the target case information in the form of a node map;
and/or the presence of a gas in the gas,
and outputting the adaptation degree of each criminal term and all determined target money in the criminal term value range in a node map mode.
The device herein may be a server, a PC, a PAD, a mobile phone, etc.
The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device:
acquiring target case information, wherein the target case information represents whether each judicial element related to the criminal phase exists in the case of the criminal phase to be predicted or not;
for each law related to the criminal period, determining a target money in the law according to the adaptation degree of each money in the law and the target case information;
and determining the target criminal stage according to the adaptation degree of each criminal stage and all the determined target money in the criminal stage value range.
Optionally, the determining the target money in the law according to the adaptation degree of each money in the law and the target case information includes:
determining the adaptation degree of each item in the law and the target case information according to the pre-learned adaptation degree distribution condition of the law items and the case information;
and determining the money with the maximum adaptation degree with the target case information in the law as the target money in the law.
Optionally, the distribution of the degree of adaptation between the pre-learned law clauses and case information is as follows:
multiplying the probability distribution of the law and the case information learned according to the historical judicial cases by the continuous multiplication result of the probabilities of all judicial elements related to the criminal period existing in the case to obtain the adaptation degree distribution condition of the law and the case information;
wherein the probability of each judicial element existing in the case is as follows: the ratio of the number of historical judicial cases for which the judicial element exists to the total number of historical judicial cases.
Optionally, the determining the target criminal stage according to the matching degree of each criminal stage and all determined target money in the criminal stage value range includes:
determining the adaptation degree of each criminal phase and all determined target money in the criminal phase value range according to the pre-learned adaptation degree distribution condition of the criminal phase and the legal fund;
and determining the penalty period with the maximum fitting degree with all the determined target money as the target penalty period.
Optionally, the distribution of the fitting degrees of the forelearned criminal phase and the law article is as follows:
multiplying the probability distribution of the criminal period learned according to the historical judicial cases under the condition of the law clause by the continuous multiplication result of the probability of each item in the law clause related to the criminal period existing in the historical judicial cases to obtain the distribution condition of the adaptation degree of the criminal period and the law clause;
wherein, the probability that each style exists in the historical judicial case is as follows: the frequency with which the money occurs in historical jurisdictions.
Optionally, the method further includes:
outputting the adaptation degree of each target money and the target case information at least;
and/or the presence of a gas in the gas,
and outputting the adaptation degree of each criminal period and all determined target money in the criminal period value range.
Optionally, the adaptation degree of each money in each law related to the criminal phase and the target case information is at least output; and/or outputting the adaptation degree of each criminal phase and all determined target money in the criminal phase value range, wherein the adaptation degree comprises the following steps:
outputting the adaptation degree of each money in each law related to the criminal phase and the target case information in the form of a node map;
and/or the presence of a gas in the gas,
and outputting the adaptation degree of each criminal term and all determined target money in the criminal term value range in a node map mode.
As will be appreciated by one skilled in the art, 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, disk storage, CD-ROM, 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 flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams 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.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media 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 cassettes, magnetic tape magnetic disk 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. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, 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, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A criminal phase prediction method, comprising:
acquiring target case information, wherein the target case information represents whether each judicial element related to the criminal phase exists in the case of the criminal phase to be predicted or not;
for each law related to the criminal period, determining a target money in the law according to the adaptation degree of each money in the law and the target case information;
and determining the target criminal stage according to the adaptation degree of each criminal stage and all the determined target money in the criminal stage value range.
2. The method according to claim 1, wherein the determining the target money in the law sentence according to the degree of adaptation of each money in the law sentence to the target case information comprises:
determining the adaptation degree of each item in the law and the target case information according to the pre-learned adaptation degree distribution condition of the law items and the case information;
and determining the money with the maximum adaptation degree with the target case information in the law as the target money in the law.
3. The method according to claim 2, wherein the pre-learned degree distribution of adaptation between the law clause and the case information is as follows:
multiplying the probability distribution of the law and the case information learned according to the historical judicial cases by the continuous multiplication result of the probabilities of all judicial elements related to the criminal period existing in the case to obtain the adaptation degree distribution condition of the law and the case information;
wherein the probability of each judicial element existing in the case is as follows: the ratio of the number of historical judicial cases for which the judicial element exists to the total number of historical judicial cases.
4. The method according to claim 1, wherein the determining the target penalty periods according to the degree of adaptation of each penalty period to all the determined target money in the penalty period value range comprises:
determining the adaptation degree of each criminal phase and all determined target money in the criminal phase value range according to the pre-learned adaptation degree distribution condition of the criminal phase and the legal fund;
and determining the penalty period with the maximum fitting degree with all the determined target money as the target penalty period.
5. The method according to claim 4, wherein the pre-learned criminal phase and law fit degree distribution is:
multiplying the probability distribution of the criminal period learned according to the historical judicial cases under the condition of the law clause by the continuous multiplication result of the probability of each item in the law clause related to the criminal period existing in the historical judicial cases to obtain the distribution condition of the adaptation degree of the criminal period and the law clause;
wherein, the probability that each style exists in the historical judicial case is as follows: the frequency with which the money occurs in historical jurisdictions.
6. The method of claim 1, further comprising:
outputting the adaptation degree of each target money and the target case information at least;
and/or the presence of a gas in the gas,
and outputting the adaptation degree of each criminal period and all determined target money in the criminal period value range.
7. The method according to claim 6, wherein said outputting at least a degree of adaptation of each of the respective laws related to criminal phase to said target case information; and/or outputting the adaptation degree of each criminal phase and all determined target money in the criminal phase value range, wherein the adaptation degree comprises the following steps:
outputting the adaptation degree of each money in each law related to the criminal phase and the target case information in the form of a node map;
and/or the presence of a gas in the gas,
and outputting the adaptation degree of each criminal term and all determined target money in the criminal term value range in a node map mode.
8. A criminal phase prediction device, comprising:
the system comprises an acquisition module, a prediction module and a prediction module, wherein the acquisition module is used for acquiring target case information, and the target case information represents whether each judicial element related to the criminal phase exists in the case of the criminal phase to be predicted or not;
the first prediction module is used for determining a target money in each law related to the criminal period according to the adaptation degree of each money in the law and the target case information;
and the second prediction module is used for determining the target criminal stage according to the adaptation degree of each criminal stage and all the determined target money in the criminal stage value range.
9. A storage medium, characterized in that it comprises a stored program, wherein said program executes the penalty prediction method of any of claims 1 to 7.
10. An apparatus comprising at least one processor, and at least one memory connected to the processor, a bus; the processor and the memory complete mutual communication through the bus; the processor is adapted to invoke program instructions in the memory to perform a criminal prediction method according to any of the claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911053891.6A CN112749826A (en) | 2019-10-31 | 2019-10-31 | Criminal period prediction method, device, storage medium and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911053891.6A CN112749826A (en) | 2019-10-31 | 2019-10-31 | Criminal period prediction method, device, storage medium and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112749826A true CN112749826A (en) | 2021-05-04 |
Family
ID=75645611
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911053891.6A Pending CN112749826A (en) | 2019-10-31 | 2019-10-31 | Criminal period prediction method, device, storage medium and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112749826A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130297540A1 (en) * | 2012-05-01 | 2013-11-07 | Robert Hickok | Systems, methods and computer-readable media for generating judicial prediction information |
| CN106934483A (en) * | 2016-11-18 | 2017-07-07 | 北京工业大学 | A kind of criminal justice reasoning by cases method based on body by linear programming |
| CN109376227A (en) * | 2018-10-29 | 2019-02-22 | 山东大学 | A kind of prison term prediction technique based on multitask artificial neural network |
| CN109472424A (en) * | 2018-12-18 | 2019-03-15 | 广东博维创远科技有限公司 | Prediction technique, device, storage medium and the server of crime practical prison term |
| CN109558993A (en) * | 2018-12-18 | 2019-04-02 | 华南师范大学 | Prediction technique, device, storage medium and the server of theory of crime prison term |
-
2019
- 2019-10-31 CN CN201911053891.6A patent/CN112749826A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130297540A1 (en) * | 2012-05-01 | 2013-11-07 | Robert Hickok | Systems, methods and computer-readable media for generating judicial prediction information |
| CN106934483A (en) * | 2016-11-18 | 2017-07-07 | 北京工业大学 | A kind of criminal justice reasoning by cases method based on body by linear programming |
| CN109376227A (en) * | 2018-10-29 | 2019-02-22 | 山东大学 | A kind of prison term prediction technique based on multitask artificial neural network |
| CN109472424A (en) * | 2018-12-18 | 2019-03-15 | 广东博维创远科技有限公司 | Prediction technique, device, storage medium and the server of crime practical prison term |
| CN109558993A (en) * | 2018-12-18 | 2019-04-02 | 华南师范大学 | Prediction technique, device, storage medium and the server of theory of crime prison term |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20180349514A1 (en) | Domain-oriented predictive model feature recommendation system | |
| US20190072968A1 (en) | Generating driving behavior models | |
| CN110781294A (en) | Training corpus refinement and incremental update | |
| Tselentis et al. | Improving short‐term traffic forecasts: to combine models or not to combine? | |
| CN110516910A (en) | Declaration form core based on big data protects model training method and core protects methods of risk assessment | |
| CN110458324B (en) | Method and device for calculating risk probability and computer equipment | |
| Hao et al. | Data‐driven car‐following model based on rough set theory | |
| CN111538794B (en) | Data fusion method, device and equipment | |
| Hernández-Nieves et al. | A machine learning platform for stock investment recommendation systems | |
| CN107133238A (en) | A kind of text message clustering method and text message clustering system | |
| CN112069795A (en) | Corpus detection method, apparatus, device and medium based on mask language model | |
| US11562373B2 (en) | Utilizing machine learning models, predictive analytics, and data mining to identify a vehicle insurance fraud ring | |
| Lemmer | Confidence factors, empiricism and the Dempster-Shafer theory of evidence | |
| Shaaban et al. | Deep convolutional forest: a dynamic deep ensemble approach for spam detection in text | |
| US20180357564A1 (en) | Cognitive flow prediction | |
| CN111882426A (en) | Business risk classifier training method, device, equipment and storage medium | |
| US11131789B2 (en) | Ice formation and ice depth estimation system | |
| CN111160959A (en) | User click conversion estimation method and device | |
| CN109597982B (en) | Abstract text recognition method and device | |
| CN116635872A (en) | Antagonistic semi-supervised one-time learning | |
| CN107861945A (en) | Finance data analysis method, application server and computer-readable recording medium | |
| CN110532773B (en) | Malicious access behavior identification method, data processing method, device and equipment | |
| Costescu | Keeping the autonomous vehicles accountable: Legal and Logic Analysis on Traffic Code | |
| Oprea et al. | Data mining and ANFIS application to particulate matter air pollutant prediction. A comparative study | |
| CN113434857A (en) | User behavior safety analysis method and system applying deep learning |
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 |