CN113408263A - Criminal period prediction method and device, storage medium and electronic device - Google Patents
Criminal period prediction method and device, storage medium and electronic device Download PDFInfo
- Publication number
- CN113408263A CN113408263A CN202010182002.2A CN202010182002A CN113408263A CN 113408263 A CN113408263 A CN 113408263A CN 202010182002 A CN202010182002 A CN 202010182002A CN 113408263 A CN113408263 A CN 113408263A
- Authority
- CN
- China
- Prior art keywords
- case
- criminal
- document
- key element
- key
- 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 50
- 238000012549 training Methods 0.000 claims description 23
- 238000004891 communication Methods 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 description 19
- 238000010586 diagram Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 10
- 238000004590 computer program Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- 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
- G06Q50/26—Government or public services
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Biophysics (AREA)
- Educational Administration (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Primary Health Care (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a criminal period prediction method, a criminal period prediction device, a storage medium and electronic equipment. The method is based on each key element contained in the case of the case, and predicts the criminal phase of the case by using the trained criminal phase prediction model, so that a basis is provided for a judge to determine the criminal phase of the case, the complexity of determining the criminal phase of the case is reduced, and a large amount of human resources consumed for searching similar cases to determine the criminal phase are reduced.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a criminal phase prediction method, a criminal phase prediction device, a storage medium and electronic equipment.
Background
The judicial prisoner is the final link of the criminal case. After the criminal name applicable to the criminal case is determined, the criminal usually refers to a related law, and determines the criminal period of the criminal case by referring to the criminal period judgment result of a similar case within the range specified by the law.
However, since the complexity of searching similar cases according to the names of the criminals applicable to the criminals is high, more human resources are consumed, so that the complexity of determining the criminal period of the criminals is high and more human resources are consumed.
Disclosure of Invention
In view of the above, the present invention provides a criminal prediction method, apparatus, storage medium and electronic device that overcome or at least partially solve the above problems.
In a first aspect, the present application provides a penalty period prediction method comprising:
obtaining case situation documents of cases to be predicted in a criminal period;
determining a set of key elements contained in the case scenario, wherein the set of key elements comprises at least one key element influencing the criminal phase;
and predicting the criminal phase of the case by utilizing a trained criminal phase prediction model based on at least one key element contained in the key element set, wherein the criminal phase prediction model is obtained by utilizing a plurality of case condition document samples marked with actual criminal phases and at least one key element for training.
In a second aspect, the present application provides a criminal phase prediction device comprising:
the system comprises a document acquisition module, a document prediction module and a document prediction module, wherein the document acquisition module is used for acquiring case situation documents of cases to be predicted in a criminal period;
the element determining module is used for determining a key element set contained in the case scenario document, wherein the key element set comprises at least one key element influencing the criminal phase;
and the prediction module is used for predicting the criminal phase of the case by utilizing a trained criminal phase prediction model based on at least one key element contained in the key element set, wherein the criminal phase prediction model is obtained by utilizing a plurality of case condition document samples marked with actual criminal phases and the at least one key element.
In a third aspect, the present application provides a storage medium comprising a stored program, wherein said program performs the criminal period prediction method of the first aspect described above.
In a fourth aspect, the present application provides an electronic device comprising at least one processor, at least one memory connected to the processor, and a bus;
the processor and the memory complete mutual communication through a bus;
said processor is adapted to invoke program instructions in said memory to perform a penalty prediction method as described above in relation to the first aspect.
By means of the technical scheme, the criminal period of the case is predicted by using the trained criminal period prediction model based on all key elements contained in the case, so that a basis is provided for a judge to determine the criminal period of the case, the complexity of determining the criminal period of the case is reduced, and a large amount of human resources consumed for determining the criminal period by searching similar cases are reduced.
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 schematic flow diagram of one embodiment of a criminal phase prediction method provided by the present application;
FIG. 2 shows a schematic training flow of the criminal phase prediction model training provided by the present application;
FIG. 3 shows a schematic flow diagram of yet another embodiment of a penalty period prediction method provided by the present application;
fig. 4 shows a schematic view of a criminal phase prediction device provided by the present application;
fig. 5 shows a schematic view of a further construction of a criminal phase prediction device provided by the present application;
fig. 6 is a schematic diagram illustrating a structure of an electronic device according to the present application.
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.
Referring to fig. 1, the present application provides a criminal phase prediction method, which can be applied to any electronic device with data processing capability. The method comprises steps S101-S103. Wherein:
s101: obtaining case situation documents of cases to be predicted in the criminal period.
A case document is a file that describes the detailed information of a case. For example, a case may include information on where the criminal case occurred, the time, and the name of the criminal to which the case belongs.
The content of the case will influence the judgment result of the criminal case, and for this reason, the content in the case is used as a reference for predicting the criminal period of the criminal case.
S102: determining a set of key elements contained in the case paper, wherein the set of key elements comprises at least one key element influencing the criminal phase.
As can be known from the introduction of the case, the case contains content information that affects the case criminal period, and the content information may contain a plurality of influencing factors that affect the criminal period from different dimensions, and each influencing factor belongs to a key element in the case.
Such as: the names of crimes, the means of doing cases, the times of occurrence of cases, the places of occurrence of cases, and the like of cases in case documents belong to key factors influencing the criminal period of cases. The method for determining the key elements in the case document may be various, for example, the case document may be searched based on a preset key element library to locate each key element included in the case document, and a key element set of the case document may be obtained.
In one example, determining the set of key elements contained in the case document may also be: and analyzing the key elements in the case document according to the set element matching rules of the plurality of key elements to obtain a key element set contained in the case document. Wherein, the key elements in the key element set belong to a plurality of set key elements.
The element matching rule of the key element is a rule which is constructed based on the characteristics satisfied by the key element and is used for retrieving the key element from the case document. For example, all key elements that may be involved in different case documents are determined by analyzing a plurality of historical case documents, and then, for each key element, the feature or composition rule of each key element is determined according to the description and different expression forms of the key element in different historical case documents, and the like, and the element matching rule of the key element is constructed based on the feature or composition rule of the key element.
For example, for the key element of "drunk driving", considering that the features of drunk driving generally must include two words of wine and driving, the element matching rule of the key element may be a matching rule for locating words or sentences including "wine" and "driving".
For example, considering that a regular expression can be used to retrieve and replace the text conforming to a certain pattern, for each key element, the element matching rule of the key element can be the regular expression of the key element.
In order to take account of the specific forms of the key elements in different case documents, each key element may have one or more element matching rules.
It will be appreciated that the influencing factors of different criminal names differ considerably, and therefore the key elements that need to be followed by different criminal names will also differ. Therefore, the application can set various key elements corresponding to the criminal names respectively aiming at each criminal name, namely each criminal name corresponds to a plurality of key elements, and the types of the key elements corresponding to different criminal names can be different.
Correspondingly, before the key elements of the case are determined, the criminal names of the case can be determined according to the case document. In this case, at least one key element included in the case document is determined based on a plurality of key elements under the name of the criminal act set.
If so, after the criminal name of the case is determined, the key elements contained in the case document of the case are analyzed according to the set element matching rules of the multiple key elements corresponding to the criminal name.
It is understood that the elements affecting criminal phase in case documents may be elements related to law statutes, and may also be elements related to case. Therefore, the types of the elements to which the plurality of key elements set in the present application belong may include: one type or two types of elements of a first type related to case and elements of a second type related to law.
Optionally, in consideration of the fact that the case of the case can more intuitively reflect the specific situation of the case for the case, and the criminal period can be more favorably determined, therefore, the preset multiple key elements in the application at least comprise key elements related to the case.
S103: and predicting the criminal phase of the case by using the trained criminal phase prediction model based on at least one key element contained in the key element set.
The criminal period prediction model is obtained by training a plurality of case condition document samples marked with actual criminal periods and at least one key element. The at least one key element marked on the case scenario document sample refers to a key element extracted from the case scenario document sample, and the way of extracting the key element from the case scenario document sample may be manual extraction, or refer to the aforementioned related introduction of extracting the key element from the case scenario document, and is not described herein again.
The criminal prediction model can be a pre-trained neural network model, a linear regression model or a regression tree model, and the like, and is not limited.
The criminal phase prediction model can analyze and process the key elements in the key element set, and finally determines the criminal phase matched with the key elements in the key element set.
For example, the key elements in the key element set analyzed from the case paper may be converted into vectors, and the vectors of the key elements in the key element set are input into the penalty prediction model, so as to obtain the penalty predicted by the penalty prediction model.
For another example, an element feature may be constructed according to at least one key element included in the key element set, where the element feature characterizes an element category of the key element included in the key element set; and then, inputting the essential features into the trained criminal phase prediction model to obtain the criminal phase of the case predicted by the criminal phase prediction model.
The key element features corresponding to the key element sets can reflect which kinds of key elements are included in the key element sets. For example, the element feature may be a group of element codes, the number of bits of the group of element codes is the same as the number of preset key elements, each bit in the group of element codes represents a key element, and the value of each bit in the element codes represents whether the key element corresponding to the bit exists or not. For example, the value of each bit in the element code is 0 or 1, wherein if the value of a bit is 1, it indicates that the key element represented by the bit exists in the case document; and if the value of the bit is 0, the key element represented by the bit does not exist in the case document.
For ease of understanding, the following are illustrated:
assuming that 10 kinds of key elements, key element 1 to key element 10, respectively, are set, the number of element encoding bits characterizing feature elements may be set to 10 bits, and these 10 bits sequentially characterize key elements 1 to 10.
Then, if the key element set parsed from the case document only includes key elements 1 and 3, when constructing an element code representing element features based on the key element set, a value at a position representing the key element 1 in the element code is 1, a value at a position representing the key element 3 in the element code is 1, values at other positions in the element code are 0, and accordingly, the constructed element code may be: 1,0,1,0,0,0,0,0,0,0.
Correspondingly, the element code is input into a criminal prediction model, and the criminal prediction model can determine which kinds of key elements are contained in the case of the case based on the element code, so that the criminal period of the case is predicted based on the kinds of the key elements contained in the case.
By means of the technical scheme, the criminal period of the case is predicted by utilizing the trained criminal period prediction model based on all key elements contained in the case, so that a basis is provided for a judge to determine the criminal period of the case, the complexity of determining the criminal period of the case is reduced, a large amount of human resources consumed for determining the criminal period by searching similar cases are reduced, the criminal period processing efficiency of the case with the criminal period to be predicted is improved, and the subconscious deviation existing in artificial judgment is avoided.
Because the criminal period is not directly predicted based on the case in the criminal period prediction model based criminal period prediction process, the criminal period is predicted by combining the key elements which influence the criminal period and are contained in the case, and because the key elements contained in the case reflect the cases, the law and other information of the cases more directly and accurately, the technical scheme provided by the application not only reduces the consumption of human resources, but also is beneficial to more accurately predicting the criminal period by combining the key elements contained in the case.
It can be understood that, in order to facilitate understanding of the process of training the criminal prediction model, the criminal prediction model is taken as an example of a trained neural network model. For example, referring to fig. 2, there is shown a schematic diagram of a training process for providing a criminal phase prediction model according to the present application, the process comprising:
s201, obtaining a plurality of historical case documents serving as training samples, wherein the historical case documents are marked with actual criminal periods.
The historical case documents are the case documents of the cases with the determined criminal periods, and therefore the criminal period of the case corresponding to each historical case document is determined.
S202, aiming at each historical case document, determining the criminal name corresponding to the historical case document, and extracting a key element set containing at least one key element from the historical case document according to the set element matching rule of various key elements related to the criminal name.
The at least one key element extracted from the historical case document is the at least one key element which needs to be marked for the historical case document.
Of course, the step S202 is only one way to determine at least one key element in the historical case document as the training sample, and in practical applications, at least one key element in the historical case document may be predetermined in other ways, for example, at least one key element in the historical case document is predetermined and labeled by a worker.
S203, aiming at each historical case document, constructing an element code according to the key element set extracted from the historical case document.
The element code represents the specific category of the key elements contained in the historical case document.
The specific process of encoding the building elements may refer to the related description of the foregoing embodiments, and is not described herein again.
S204, aiming at each historical case document, inputting the element codes corresponding to the historical case document into a neural network model needing to be trained to obtain the predicted criminal period corresponding to the historical case document predicted by the neural network model.
S205, detecting whether the training end condition of the neural network model is reached or not according to the actual criminal period and the predicted criminal period corresponding to each historical case document, and if so, ending the training; if not, adjusting the internal parameters of the neural network model, and returning to the step S204 for training.
The training end condition of the neural network may be: the times of the circular training reach the set times, or the convergence condition of the neural network model is met; the prediction accuracy of the neural network model may also meet the set requirement, and the like, which is not limited.
The adjusting of the internal parameters of the neural network model includes adjusting the weight of each key element in the neural network model, some intrinsic parameters of the neural network model, and the like.
It can be understood that, in practical applications, in order to further ensure the accuracy of predicting the criminal phase by the neural network model, the application may further obtain a plurality of historical case documents for verifying the sample, wherein each historical case document is labeled with the actual criminal phase and at least one key element.
Correspondingly, the trained neural network model can be verified by using a plurality of historical case documents, and if the prediction accuracy of the neural network model is verified to meet the requirement, the trained neural network model can be used as a criminal prediction model. If the prediction accuracy of the neural network model is verified to be not in accordance with the requirement, the training sample can be obtained again and the neural network model can be retrained; or reconstructing various key elements related to different criminal names, or submitting to an expert for rechecking to determine the cause of the problem.
It is understood that the training method of the present embodiment is described by taking the training of the neural network model as the criminal prediction model as an example, but the neural network model is replaced by other machine learning models and is also applicable to the training method.
In the embodiment, the predicted criminal period corresponding to the case is obtained through the criminal period prediction model, so that the basis for case criminal period judgment can be provided for the judge, but the specificity of some cases is considered, and the accurate criminal period of the case cannot be determined only according to the predicted criminal period.
For example, referring to fig. 3, there is shown a schematic flow chart of yet another embodiment of the criminal phase prediction method provided by the present application, the method comprising:
s301: obtaining case situation documents of cases to be predicted in the criminal period.
S302: determining a set of key elements contained in the case paper, wherein the set of key elements comprises at least one key element influencing the criminal phase.
S303: based on at least one key element contained in the key element set, the criminal phase of the case is predicted by using a trained criminal phase prediction model, and the criminal phase prediction model is obtained by training at least one key element contained in a plurality of case condition document samples marked with actual criminal phases.
The above steps can refer to the related descriptions in the previous embodiments, and are not described herein again.
S304: and obtaining the weight of each of the set key elements.
Wherein, the respective weight of the key elements is the weight corresponding to each key element set in the trained criminal stage prediction model.
It can be understood that, after the training of the criminal phase prediction model is completed, when the criminal phase prediction model performs the criminal phase prediction, the weight of each key element applied by the criminal phase prediction model is fixed, and therefore, the respective weights of the various key elements set in the trained criminal phase prediction model can be obtained.
It can be understood that, considering that under different criminal names, the influence degree of the same key element on the criminal names can be different, under the condition of aiming at multiple key elements corresponding to different criminal names, a criminal period prediction model suitable for the criminal names can be trained independently aiming at each criminal name.
Accordingly, after determining the criminal name of the case based on the case' S case document, in the step S303, the predicted criminal period of the case can be determined by using the criminal period prediction model matched with the criminal name based on the key elements extracted from the case document. In this case, the weight of the plurality of kinds of key elements set in step S304 is also the weight of each of the plurality of kinds of key elements related to the criminal name.
The sequence of step S304 and step S303 is not limited to that shown in fig. 3, and the sequence of these two steps may be interchanged or performed simultaneously.
S305: at least one historical case document satisfying the conditions is searched from the document library according to the key elements contained in the key element set of the case document.
The historical case document satisfying the condition is a historical case document having at least one key element identical to the case document.
If the key element set in the case document includes the key element 1, the key element 2, and the key element 3, the history case document is a history case document that satisfies the condition if any one or more of the key element 1, the key element 2, and the key element 3 is also included in the history case document.
In the document library, a plurality of history documents are stored. Wherein, each historical case document corresponds to a key element which is analyzed from the historical case document in advance.
For example, case documents of some more classical cases can be screened out to be used as historical case documents stored in the document library.
Alternatively, the historical case documents in the document library may be a plurality of historical case documents used for training a criminal prediction model. In the process of training the criminal prediction model, the extraction of the key elements corresponding to the historical case document is completed, so that the historical case document and the key elements contained in the historical case document are stored in the document library.
S306, aiming at each historical case document meeting the conditions, determining the similarity between the historical case document and the case document according to the same at least one key element in the historical case document and the respective weight of the same at least one key element.
For example, the similarity between the historical case document and the case document is the weighted sum of the same key elements in the historical case document and the case document.
For example, if the historical case document and the case document have the same key element 1 and key element 3, where the weight of the key element 1 is 0.6 and the weight of the key element 3 is 0.2, the similarity is 1 × 0.6+1 × 0.2 — 0.8.
S307, at least one historical case document with the similarity meeting the requirement is determined from the at least one historical case document meeting the condition.
For example, the similarity conformity requirement may be sorted from high to low, where the sorting belongs to a preset position, for example, the top 5 history cases with higher similarity are determined as the history cases with the similarity conformity requirement.
For another example, the similarity meets the requirement, that is, the similarity exceeds a set threshold, for example, if the similarity exceeds 0.8, the similarity is determined to meet the requirement.
S308: and outputting a key element set, at least one historical case document with the similarity meeting the requirement and the criminal period of the case predicted by a criminal period prediction model.
The historical case information documents with higher similarity to the case information documents of the case are output, so that similar case information documents can be prevented from being manually searched by a judge, a reference basis with another dimension can be provided for the judge to determine the final criminal period of the case by combining the predicted criminal period, and the case information document is more accurately and reliably determined by the judge.
It should be noted that the key element set extracted from the case document outputting the case is only an optional manner, and the purpose of the key element set is to provide search bases for the searched historical case documents for the judge, so that the judge assists in judging whether the searched historical case document is the required case document by combining the key elements in the key element set.
In yet another aspect, the present application also provides a criminal phase prediction device. For example, referring to fig. 4, a criminal phase predicting device of the present application embodiment includes:
the document acquiring module 401 is used for acquiring case situation documents of cases to be predicted in a criminal period;
an element determining module 402, configured to determine a set of key elements included in the case document, where the set of key elements includes at least one key element affecting criminal phase;
the prediction module 403 is configured to predict the criminal phase of a case by using a trained criminal phase prediction model based on at least one key element included in the key element set, where the criminal phase prediction model is obtained by training a plurality of case context document samples labeled with an actual criminal phase and at least one key element.
By means of the technical scheme, the criminal period of the case is predicted by using the trained criminal period prediction model based on all key elements contained in the case, so that a basis is provided for a judge to determine the criminal period of the case, the complexity of determining the criminal period of the case is reduced, and a large amount of human resources consumed for determining the criminal period by searching similar cases are reduced.
In one example, the element determining module, when determining the key element set included in the case scenario document, is specifically configured to:
and analyzing the key elements in the case document according to the set element matching rules of the plurality of key elements to obtain the key element set contained in the case document.
In one example, the criminal phase prediction apparatus may further include:
and the criminal name determining module is used for determining the criminal name related to the case based on the case document.
The element determining module is specifically configured to, when analyzing the key elements in the case document according to a set element matching rule of multiple key elements to obtain the key element set included in the case document:
and analyzing the key elements in the case document according to the set element matching rules of a plurality of key elements corresponding to criminal names to obtain the key element set contained in the case document.
In one example, the prediction module, when predicting the criminal phase of a case using a trained criminal phase prediction model based on at least one key element contained in the set of key elements, is specifically configured to:
and constructing element characteristics according to at least one key element contained in the key element set, wherein the element characteristics characterize the element types of the key elements contained in the key element set.
Inputting the element characteristics into a trained criminal phase prediction model to obtain the criminal phase of the case predicted by the criminal phase prediction model.
Referring to fig. 5, there is shown another schematic structural diagram of the criminal phase prediction device provided in the present application, specifically, the device includes, in addition to a document acquisition module 401, an element determination module 402 and a prediction module 403:
a weight obtaining module 404, configured to obtain respective weights of the set multiple key elements, where the respective weights of the multiple key elements are weights corresponding to the key elements set in the trained criminal period prediction model. A document primary selection module 405, configured to search, according to key elements included in the key element set of the case scenario document, at least one historical case scenario document that satisfies a condition from a document library, where the historical case scenario document that satisfies the condition refers to a historical case scenario document that includes at least one same key element as the case scenario document;
a similarity calculation module 406, configured to determine, for each historical case document that meets a condition, a similarity between the historical case document and the case document according to at least one identical key element in the historical case document and the case document and a respective weight of the at least one identical key element;
and the document checking module 407 is configured to determine at least one historical case document with the similarity meeting the requirement from the at least one historical case document meeting the condition.
Optionally, the apparatus may further include an output module 408, configured to output the set of key elements, at least one historical case document with satisfactory similarity, and the criminal phase of the case predicted by the criminal phase prediction model.
In one example, the types of elements to which the plurality of elements belong include: one type or two types of elements of a first type related to case and elements of a second type related to law.
The criminal prediction device comprises a processor and a memory, wherein the document acquisition module, the element determination module, the prediction module 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 than one, and criminal prediction is carried out on the case based on the case situation document of the case by adjusting the kernel parameters, so that a judge is guided to make criminal judgment on the case document.
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.
An embodiment of the present invention provides an electronic device, as shown in fig. 6, the electronic device includes at least one processor 601, at least one memory 602 connected to the processor 601, and a bus 603; the processor and the memory complete mutual communication through a bus; the processor is arranged to call program instructions in the memory to perform the above-mentioned penalty prediction method. 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:
obtaining case situation documents of cases to be predicted in a criminal period;
determining a set of key elements contained in the case scenario, wherein the set of key elements comprises at least one key element influencing the criminal phase;
and predicting the criminal phase of the case by utilizing a trained criminal phase prediction model based on at least one key element contained in the key element set, wherein the criminal phase prediction model is obtained by utilizing a plurality of case condition document samples marked with actual criminal phases and at least one key element.
In one example, the determining the set of key elements contained in the case scenario document includes: and analyzing the key elements in the case document according to the set element matching rules of the plurality of key elements to obtain the key element set contained in the case document.
In one example, before the determining the set of key elements contained in the case scenario document, the method further includes:
determining the name of a criminal related to the case based on the case paper;
the analyzing the key elements in the case document according to the set element matching rules of the multiple key elements to obtain the key element set contained in the case document comprises:
and analyzing the key elements in the case document according to the set element matching rules of a plurality of key elements corresponding to the criminal names to obtain the key element set contained in the case document.
In one example, said predicting the criminal phase of said case using a trained criminal phase prediction model based on at least one key element contained in said set of key elements comprises:
constructing element characteristics according to at least one key element contained in the key element set, wherein the element characteristics represent the element types of the key elements contained in the key element set;
inputting the element characteristics into a trained criminal phase prediction model to obtain the criminal phase of the case predicted by the criminal phase prediction model.
In one example, the penalty period prediction method further comprises:
obtaining respective weights of multiple set key elements, wherein the respective weights of the multiple key elements are weights corresponding to the key elements set in the trained criminal stage prediction model;
searching at least one historical case document meeting the conditions from a document library according to the key elements contained in the key element set of the case document, wherein the historical case document meeting the conditions is the historical case document containing at least one same key element as the case document;
for each historical case document meeting the conditions, determining the similarity between the historical case document and the case document according to at least one identical key element in the historical case document and the respective weight of the at least one identical key element;
and determining at least one historical case document with the similarity meeting the requirement from the at least one historical case document meeting the condition.
In one example, the penalty period prediction method further comprises:
and outputting the key element set, information of at least one historical case document with the similarity meeting the requirement and the criminal period of the case predicted by the criminal period prediction model.
In one example, the element types to which the plurality of elements belong include: one type or two types of elements of a first type related to case and elements of a second type related to law.
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.
In a typical configuration, a device includes one or more processors (CPUs), memory, and a bus. The device may also include input/output interfaces, network interfaces, and the like.
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. 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:
obtaining case situation documents of cases to be predicted in a criminal period;
determining a set of key elements contained in the case scenario, wherein the set of key elements comprises at least one key element influencing the criminal phase;
and predicting the criminal phase of the case by utilizing a trained criminal phase prediction model based on at least one key element contained in the key element set, wherein the criminal phase prediction model is obtained by utilizing a plurality of case condition document samples marked with actual criminal phases and at least one key element for training.
2. The method of claim 1, wherein said determining a set of key elements contained in said case scenario document comprises:
and analyzing the key elements in the case document according to the set element matching rules of the plurality of key elements to obtain the key element set contained in the case document.
3. The method according to claim 2, further comprising, prior to said determining a set of key elements contained in said case scenario document:
determining the name of a criminal related to the case based on the case paper;
the analyzing the key elements in the case document according to the set element matching rules of the multiple key elements to obtain the key element set contained in the case document comprises:
and analyzing the key elements in the case document according to the set element matching rules of a plurality of key elements corresponding to the criminal names to obtain the key element set contained in the case document.
4. The method according to any one of claims 1 to 3, wherein said predicting the criminal phase of said case using a trained criminal phase prediction model based on at least one key element contained in said set of key elements comprises:
constructing element characteristics according to at least one key element contained in the key element set, wherein the element characteristics represent the element types of the key elements contained in the key element set;
inputting the element characteristics into a trained criminal phase prediction model to obtain the criminal phase of the case predicted by the criminal phase prediction model.
5. The method of any of claims 1 to 3, further comprising:
obtaining respective weights of multiple set key elements, wherein the respective weights of the multiple key elements are weights corresponding to the key elements set in the trained criminal stage prediction model;
searching at least one historical case document meeting the conditions from a document library according to the key elements contained in the key element set of the case document, wherein the historical case document meeting the conditions is the historical case document containing at least one same key element as the case document;
for each historical case document meeting the conditions, determining the similarity between the historical case document and the case document according to at least one identical key element in the historical case document and the respective weight of the at least one identical key element;
and determining at least one historical case document with the similarity meeting the requirement from the at least one historical case document meeting the condition.
6. The method of claim 5, further comprising:
and outputting the key element set, the information of at least one historical case document with the similarity meeting the requirement and the criminal period of the case predicted by the criminal period prediction model.
7. The method of claim 2, wherein the element types to which the plurality of elements belong comprise: one type or two types of elements of a first type related to case and elements of a second type related to law.
8. A criminal phase prediction device, comprising:
the system comprises a document acquisition module, a document prediction module and a document prediction module, wherein the document acquisition module is used for acquiring case situation documents of cases to be predicted in a criminal period;
the element determining module is used for determining a key element set contained in the case scenario document, wherein the key element set comprises at least one key element influencing the criminal phase;
and the prediction module is used for predicting the criminal phase of the case by utilizing a trained criminal phase prediction model based on at least one key element contained in the key element set, wherein the criminal phase prediction model is obtained by utilizing a plurality of case condition document samples marked with actual criminal phases and the at least one key element.
9. A storage medium, characterized in that the storage medium comprises a stored program, wherein the program performs the criminal prediction method of any one of claims 1 to 7.
10. An electronic device comprising at least one processor, at least one memory connected to the processor, and a bus;
the processor and the memory complete mutual communication through a 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 |
|---|---|---|---|
| CN202010182002.2A CN113408263A (en) | 2020-03-16 | 2020-03-16 | Criminal period prediction method and device, storage medium and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010182002.2A CN113408263A (en) | 2020-03-16 | 2020-03-16 | Criminal period prediction method and device, storage medium and electronic device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113408263A true CN113408263A (en) | 2021-09-17 |
Family
ID=77676401
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010182002.2A Pending CN113408263A (en) | 2020-03-16 | 2020-03-16 | Criminal period prediction method and device, storage medium and electronic device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113408263A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113888368A (en) * | 2021-11-10 | 2022-01-04 | 深圳航天科创实业有限公司 | Feature selection method for criminal case detention risk assessment based on image coding |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007004756A (en) * | 2005-06-27 | 2007-01-11 | Kikuo Akagane | Automatic retrieval for legal precedent |
| CN109241528A (en) * | 2018-08-24 | 2019-01-18 | 讯飞智元信息科技有限公司 | A kind of measurement of penalty prediction of result method, apparatus, equipment and storage medium |
| CN110119857A (en) * | 2019-07-01 | 2019-08-13 | 南京擎盾信息科技有限公司 | A kind of prison term automatic prediction method meeting plus-minus punishment expert common sense rule |
| CN110610005A (en) * | 2019-09-16 | 2019-12-24 | 哈尔滨工业大学 | Stealing crime auxiliary criminal investigation method based on deep learning |
| CN110647631A (en) * | 2018-06-25 | 2020-01-03 | 阿里巴巴集团控股有限公司 | Case recommendation method and device, storage medium and processor |
-
2020
- 2020-03-16 CN CN202010182002.2A patent/CN113408263A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007004756A (en) * | 2005-06-27 | 2007-01-11 | Kikuo Akagane | Automatic retrieval for legal precedent |
| CN110647631A (en) * | 2018-06-25 | 2020-01-03 | 阿里巴巴集团控股有限公司 | Case recommendation method and device, storage medium and processor |
| CN109241528A (en) * | 2018-08-24 | 2019-01-18 | 讯飞智元信息科技有限公司 | A kind of measurement of penalty prediction of result method, apparatus, equipment and storage medium |
| CN110119857A (en) * | 2019-07-01 | 2019-08-13 | 南京擎盾信息科技有限公司 | A kind of prison term automatic prediction method meeting plus-minus punishment expert common sense rule |
| CN110610005A (en) * | 2019-09-16 | 2019-12-24 | 哈尔滨工业大学 | Stealing crime auxiliary criminal investigation method based on deep learning |
Non-Patent Citations (2)
| Title |
|---|
| HUAJIE CHEN ET: "Charge-Based Prison Term Prediction with Deep Gating Network", CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING / 9TH INTERNATIONAL JOINT CONFERENCE ON NATURAL LANGUAGE PROCESSING, pages 6362 - 6367 * |
| 郭镔: "面向智慧检务的案件研判关键技术研究", 《中国优秀博硕士学位论文全文数据库（硕士）信息科技辑》, pages 19 - 40 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113888368A (en) * | 2021-11-10 | 2022-01-04 | 深圳航天科创实业有限公司 | Feature selection method for criminal case detention risk assessment based on image coding |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111222305B (en) | Information structuring method and device | |
| CN111144723A (en) | Method and system for recommending people's job matching and storage medium | |
| CN111125343A (en) | Text analysis method and device suitable for human-sentry matching recommendation system | |
| CN111105209A (en) | Job resume matching method and device suitable for post matching recommendation system | |
| CN110458324B (en) | Method and device for calculating risk probability and computer equipment | |
| CN107102993B (en) | User appeal analysis method and device | |
| CN109947902B (en) | Data query method and device and readable medium | |
| CN111767716A (en) | Method and device for determining enterprise multilevel industry information and computer equipment | |
| CN112199602B (en) | Post recommendation method, recommendation platform and server | |
| CN112487149A (en) | Text auditing method, model, equipment and storage medium | |
| CN110955766A (en) | Method and system for automatically expanding intelligent customer service standard problem pairs | |
| CN112182214A (en) | Data classification method, device, equipment and medium | |
| CN116501898A (en) | Financial text event extraction method and device suitable for few samples and biased data | |
| CN112966100A (en) | Training method and device for data classification and classification model and electronic equipment | |
| CN110837730B (en) | Method and device for determining unknown entity vocabulary | |
| CN110689211A (en) | Method and device for evaluating website service capability | |
| CN110532229B (en) | Evidence file retrieval method, device, computer equipment and storage medium | |
| CN116228383A (en) | Risk prediction method and device, storage medium and electronic equipment | |
| CN117520503A (en) | Financial customer service dialogue generation method, device, equipment and medium based on LLM model | |
| CN116304033B (en) | Complaint identification method based on semi-supervision and double-layer multi-classification | |
| CN113408263A (en) | Criminal period prediction method and device, storage medium and electronic device | |
| CN115982388B (en) | Case quality control map establishment method, case document quality inspection method, case quality control map establishment equipment and storage medium | |
| CN110703924B (en) | Cold start method and equipment of new user based on input method application | |
| CN114647739B (en) | Entity chain finger method, device, electronic equipment and storage medium | |
| CN116383710A (en) | Label determining method, device, electronic equipment and storage medium |
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 |