CN112734565A - Method and device for predicting mobile coverage rate - Google Patents
Method and device for predicting mobile coverage rate Download PDFInfo
- Publication number
- CN112734565A CN112734565A CN202110036773.5A CN202110036773A CN112734565A CN 112734565 A CN112734565 A CN 112734565A CN 202110036773 A CN202110036773 A CN 202110036773A CN 112734565 A CN112734565 A CN 112734565A
- Authority
- CN
- China
- Prior art keywords
- data
- training
- coverage rate
- service data
- mobility
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 143
- 238000012216 screening Methods 0.000 claims abstract description 25
- 238000010801 machine learning Methods 0.000 claims description 67
- 238000004422 calculation algorithm Methods 0.000 claims description 26
- 238000004364 calculation method Methods 0.000 claims description 25
- 238000012545 processing Methods 0.000 claims description 24
- 238000004590 computer program Methods 0.000 claims description 16
- 238000012795 verification Methods 0.000 claims description 16
- 230000002068 genetic effect Effects 0.000 claims description 11
- 238000003860 storage Methods 0.000 claims description 11
- 238000012706 support-vector machine Methods 0.000 claims description 9
- 238000012417 linear regression Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 230000009969 flowable effect Effects 0.000 claims description 5
- 238000005457 optimization Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 239000012530 fluid Substances 0.000 claims description 2
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 24
- 238000010586 diagram Methods 0.000 description 17
- 238000011156 evaluation Methods 0.000 description 15
- 230000002776 aggregation Effects 0.000 description 8
- 238000004220 aggregation Methods 0.000 description 8
- 230000005540 biological transmission Effects 0.000 description 7
- 238000004891 communication Methods 0.000 description 7
- 230000008676 import Effects 0.000 description 7
- 238000012360 testing method Methods 0.000 description 6
- 239000000872 buffer Substances 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008602 contraction Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000007614 genetic variation Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000004083 survival effect Effects 0.000 description 2
- 230000001413 cellular effect Effects 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Accounting & Taxation (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Finance (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Economics (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A method and a device for predicting a mobile coverage rate belong to the technical field of big data and artificial intelligence, and the method comprises the following steps: acquiring service data of each service application server system at a historical moment and mobility coverage rate data at the historical moment; training the first training model according to the service data at the historical moment and the mobility coverage rate data at the historical moment; screening the mobility coverage rate data and the service data at the corresponding moment according to a preset condition, and training a second training model according to a screening result; obtaining a mobility coverage rate predicted value according to the first training model and the predicted service data; comparing the predicted value with a preset threshold value, and substituting the mobility coverage rate predicted value into a second training model according to a comparison result to obtain a plurality of service data configuration scheme data; and obtaining a business data adjusting scheme according to the comparison result of the business data configuration scheme data and the predicted business data.
Description
Technical Field
The invention relates to the technical field of big data and artificial intelligence, which can be applied to the field of banking business data processing and the field of finance, in particular to a method and a device for predicting liquidity coverage rate.
Background
Mobility coverage is a regulatory core index that governs short-term mobility risks. Liquidity coverage is intended to ensure that a commercial bank has sufficient qualified good liquidity assets to meet future liquidity requirements by becoming available to those assets under liquidity pressure scenarios prescribed by the bank prison.
In the prior art, when the liquidity coverage rate calculation of the commercial bank is shown in a form of a business report, the timeliness and the automation degree are not high, and meanwhile, the future liquidity coverage rate of the commercial bank cannot be predicted, so that the risk level control capability of the bank is improved.
Disclosure of Invention
The invention aims to provide a method and a device for predicting the mobility coverage rate, and the method and the device can be used for early warning the mobility coverage rate risk and predicting to obtain the optimized early warning of service data configuration according to the predicted mobility coverage rate condition; the prediction result is effectively utilized, the risk level control capability is improved, and the resistance capability of the unexpected loss of the bank is enhanced; the accuracy of index measurement and calculation can be improved, and the foresight and intelligent degree of index management are improved.
To achieve the above object, the present invention provides a method for predicting a flowable coverage, comprising: acquiring service data of each service application server system at a historical moment and mobility coverage rate data at the historical moment; training a pre-established machine learning model according to the business data at the historical moment and the mobility coverage rate data at the historical moment to obtain a first training model; screening the mobility coverage rate data and the service data at the corresponding moment according to a preset condition, and training a pre-established machine learning model according to a screening result to obtain a second training model; obtaining a mobility coverage rate predicted value according to the first training model and the predicted service data; comparing the predicted value with a preset threshold value, and substituting the mobility coverage rate predicted value into the second training model according to a comparison result to obtain a plurality of service data configuration scheme data; and obtaining a service data adjustment scheme according to the comparison result of the service data configuration scheme data and the predicted service data.
The present invention also provides a mobile coverage prediction apparatus, comprising: the device comprises a data acquisition module, a training module, a prediction module, a configuration scheme calculation module and a matching module; the data acquisition module is used for acquiring service data of each service application server system at historical time and mobility coverage rate data at historical time; the training module is used for training a pre-established machine learning model according to the business data at the historical moment and the mobility coverage rate data at the historical moment to obtain a first training model; screening the mobility coverage rate data and the service data at the corresponding moment according to a preset condition, and training a pre-established machine learning model according to a screening result to obtain a second training model; the prediction module is used for obtaining a mobility coverage rate prediction value according to the first training model and the predicted service data; comparing the predicted value with a preset threshold value, and providing the mobility coverage rate predicted value to the second training module according to a comparison result; the configuration scheme calculation module is used for substituting the mobility coverage rate predicted value into the second training model to obtain a plurality of service data configuration scheme data; and the matching module is used for obtaining a business data adjusting scheme according to the comparison result of the business data configuration scheme data and the predicted business data.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method when executing the computer program.
The present invention also provides a computer-readable storage medium storing a computer program for executing the above method.
The invention has the beneficial technical effects that: on one hand, a mobility coverage rate target can be determined according to the self operation condition, and on the other hand, when the actual mobility coverage rate deviates from the target, an adjustment measure can be actively adopted to reduce the deviation degree, so that the resistance capability of the mobility coverage rate risk under external impact is improved; moreover, the probability that the mobility coverage rate does not reach the standard is reduced through the prediction of the mobility coverage rate; the mobility coverage rate prediction can also provide certain research reference for mobility coverage rates of other emerging markets, and is wide in application range and high in transportability.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention. In the drawings:
FIG. 1A is a schematic flow chart of a method for predicting flowable coverage according to the present invention;
FIG. 1B is a flow chart of a method for mobility coverage prediction according to the present invention;
FIG. 2 is a flow chart of a method for mobility coverage prediction according to the present invention;
FIG. 3 is a flow chart of a method for mobility coverage prediction according to the present invention;
FIG. 4 is a block diagram of a mobile coverage prediction apparatus provided in the present invention;
FIG. 5 is a block diagram of a data acquisition module provided by the present invention;
FIG. 6 is a block diagram of a machine learning 1 module provided by the present invention;
FIG. 7 is a block diagram of a machine learning 2 module provided by the present invention;
FIG. 8 is a block diagram of a liquidity coverage calculation and prediction module provided by embodiments of the present invention;
fig. 9 is a block diagram of a service data configuration scheme calculation module according to an embodiment of the present invention;
FIG. 10 is a block diagram of a component similarity matching module provided by an embodiment of the present invention;
FIG. 11 is a block diagram of a data presentation module provided by an embodiment of the present invention;
fig. 12 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The following detailed description of the embodiments of the present invention will be provided with reference to the drawings and examples, so that how to apply the technical means to solve the technical problems and achieve the technical effects can be fully understood and implemented. It should be noted that, unless otherwise specified, the embodiments and features of the embodiments of the present invention may be combined with each other, and the technical solutions formed are within the scope of the present invention.
Additionally, the steps illustrated in the flow charts of the figures may be performed in a computer system such as a set of computer-executable instructions and, although a logical order is illustrated in the flow charts, in some cases, the steps illustrated or described may be performed in an order different than here.
Referring to fig. 1A, a method for predicting a fluid coverage provided by the present invention includes:
s1011, acquiring service data of each service application server system at the historical moment and mobility coverage rate data at the historical moment;
s1012, training a pre-established machine learning model according to the service data at the historical moment and the mobility coverage rate data at the historical moment to obtain a first training model;
s1013, screening the mobility coverage rate data and the service data at the corresponding moment according to a preset condition, and training a pre-established machine learning model according to a screening result to obtain a second training model;
s1014, obtaining a mobility coverage rate predicted value according to the first training model and the predicted service data;
s1015 compares the predicted value with a preset threshold value, and brings the mobility coverage rate predicted value into the second training model according to the comparison result to obtain a plurality of service data configuration scheme data;
s1016, according to the result of comparing the service data configuration scheme data with the predicted service data, obtaining a service data adjustment scheme.
In practical work, the application principle of the mobility coverage rate prediction method is as follows: acquiring service data of each service application server system at a historical moment and mobility coverage rate data at the historical moment; training and calculating a pre-established machine learning model 1 according to the business data at the historical moment and the mobility coverage rate data at the historical moment to obtain a training model 1; according to the preset conditions, the history after screening; according to the set conditions, the screened mobility coverage rate data at the historical moment and the service data at the corresponding moment train the pre-established machine learning model 2 to obtain a training model 2; calculating and predicting mobility coverage rate data by using the trained machine learning model 1; analyzing the obtained mobility coverage rate predicted value, and forming mobility coverage rate early warning according to a preset threshold value; according to the mobility coverage rate prediction data, a plurality of service data configuration scheme data are obtained through calculation by using the trained machine learning model 2; respectively carrying out component proportion similarity matching on the plurality of service data configuration schemes and the predicted service data to obtain the similarity between the plurality of service data configuration schemes and the predicted service data; and obtaining a service data adjusting scheme according to the similarity and the user preference.
Referring to fig. 2, in an embodiment of the present invention, training a pre-established machine learning model according to the traffic data at the historical time and the mobility coverage data at the historical time, and obtaining a first training model includes:
s201, randomly dividing the characteristic value information and the fluidity coverage rate data at the historical moment into K non-coincident sub data sets, taking K-1 sub data sets as training sets, and taking the remaining 1 sub data sets as verification sets;
s202, respectively inputting the training set selected each time into a pre-established machine learning model in an iterative mode, training, and performing model learning optimization by using a corresponding verification set; the method comprises the steps of performing multivariate linear regression integrated learning training through a genetic algorithm and a support vector machine regression algorithm to obtain a first training model; k is a positive integer greater than 1.
Wherein obtaining a mobility coverage prediction value according to the first training model and the predicted service data comprises: and calculating to obtain a mobility coverage rate predicted value through the feature value information corresponding to the predicted service data and the first training model.
In actual work, the original amount of historical time service data and historical time mobility coverage rate data can be randomly divided into K non-coincident subdata sets, K-1 sets are selected as training sets each time, and the rest 1 sets are used as verification sets. Respectively inputting K-1 selected each time as a training set into a pre-established machine learning model in an iterative mode, and performing initial training; and (4) optimizing the initially trained machine learning model by taking the remaining 1 data set as a test sample, and generating the trained machine learning model in an iterative mode. Specifically, a genetic algorithm GA model and a support vector machine regression (SVR) model are adopted for the machine learning model. And the GA algorithm model is used for obtaining the optimal service data configuration scheme corresponding to the corresponding predicted value at the future moment. At a certain time in the GA algorithm model, a feature attribute set WP of banking data represents an individual, wherein P is feature data, W is a corresponding weight, and a certain value W in WPiPiRepresents a genetic element in an individual, i.gtoreq.1. The suitability evaluation function is Fn. The species are divided according to individual similarity (multiple individuals make up one species). And then the adaptive evaluation function carries out target function confidence evaluation on the individual to obtain confidence evaluation. The evolution process of the GA algorithm model is survival competition, the same species are superior and inferior, a certain percentage of individuals are reserved, then competition and selection are carried out, genetic variation among species is used for reproducing offspring individuals, and the individuals with high environmental adaptability survive. After multiple iterative evolutions, the confidence evaluation of the objective function is optimal, and the set WP at the moment is the optimal individual genetic combination at the current moment. The regression confidence of the SVR model is used as a convergence target in the GA model, namely, the regression equation f (x) of the SVR model is an objective function of the GA model. The SVR model is a multiple linear regression of a support vector machine regression (SVR) algorithm,by giving a new input sample x, it is deduced from the given data sample what its corresponding output Y is, which is a real number. The regression problem can be described in mathematical language as: given a set of data samples is { (x)i,yi)|xi∈Rn,yiE.g. R, i ═ 1,2,3. Finding RnThe above function f (x) yields a regression equation to infer the y value for any x input using y ═ f (x).
In an embodiment of the present invention, obtaining the service data adjustment scheme according to the comparison result between the service data configuration scheme data and the predicted service data includes: respectively carrying out component proportion similarity matching on the plurality of service data configuration scheme data and the predicted service data to obtain the similarity between each service data configuration scheme data and the predicted service data; and screening the service data configuration scheme data according to the similarity to obtain a service data adjustment scheme meeting preset conditions.
Referring to fig. 3, in an embodiment of the present invention, training a pre-established machine learning model according to a screening result to obtain a second training model includes:
s301, acquiring preference data of a user, screened service data and fluidity coverage data at a historical moment, randomly dividing the preference data, the screened service data and the fluidity coverage data into K non-coincident sub data sets, taking K-1 sub data sets as a training set, and taking the remaining 1 sub data sets as a verification set;
s302, respectively inputting the training set selected each time into a pre-established machine learning model in an iterative mode, training, and performing model learning optimization by using a corresponding verification set; training through a neural network clustering algorithm to obtain a second training model; k is a positive integer greater than 1.
Therefore, the step of substituting the mobility coverage rate predicted value into the second training model according to the comparison result to obtain a plurality of service data configuration scheme data further comprises the following steps: and substituting the mobility coverage rate predicted value and the preference degree data into the second training model according to a comparison result to obtain a plurality of service data configuration scheme data. Therefore, the preference of the user is brought into the proportion recommendation, and the friendliness and the applicability of the proportion adjustment are further improved.
In actual work, the original amount of historical time service data and historical time mobility coverage rate data can be randomly divided into K non-coincident subdata sets, K-1 sets are selected as training sets each time, and the rest 1 sets are used as verification sets. Respectively inputting K-1 selected each time as a training set into a pre-established machine learning model in an iterative mode, and performing initial training; and (4) optimizing the initially trained machine learning model by taking the remaining 1 data set as a test sample, and generating the trained machine learning model in an iterative mode. Specifically, a neural network clustering algorithm SOM is adopted for the machine learning model to realize data clustering. And forming a high-dimensional data set as an input sample and an output sample of the data sample by using the preference data of the user, the screened service data and the mobility coverage rate data at the historical moment due to the complex data types. Randomly selecting data from an output sample as an initial value, then selecting an input vector from the input sample according to random probability, finding a weight vector with the minimum distance from the input vector to define the weight vector as a local optimal set, and then adjusting the weight in a region near the local optimal set to approach the input vector. And taking factors such as the radius of a contraction domain, a learning rate, a repetition rate, user preference and the like as evaluation function factors. And (4) performing iterative training to enable the evaluation to be optimal, and performing optimal aggregation classification at the moment. The data aggregation classification is a data set. The historical mobility coverage data at each moment can obtain an optimal aggregation classification data set, namely an optimal service data configuration scheme corresponding to the mobility coverage at the time point. And judging and scoring according to the input liquidity coverage rate value and the user preference degree.
Referring to fig. 1B, in practical applications, the method for predicting the mobility coverage provided by the present invention can be implemented as follows:
step S1021: service data of each service application server system at the historical moment of the bank and mobility coverage rate data at the historical moment; and screening the service data of each service application server system at the historical moment of the bank, processing missing data and eliminating abnormal data to serve as source data.
Step S1022: and training a pre-established machine learning model 1 according to the service data of each service application server system at the historical moment of the bank and the mobility coverage rate data at the historical moment. The original amount of historical time service data and historical time mobility coverage rate data are randomly divided into K misaligned subdata sets, K-1 subdata sets are selected as training sets each time, and the remaining 1 subdata sets are used as verification sets. Respectively inputting K-1 selected each time as a training set into a pre-established machine learning model in an iterative mode, and performing initial training; and (5) optimizing the initially trained machine learning model by using the remaining 1 data set as a test sample. And obtaining the mobility coverage rate prediction model through the multiple linear regression integrated learning training of the genetic algorithm and the support vector machine regression algorithm.
Step S1023: and according to the set conditions, the screened mobile coverage rate data at the historical moment of the bank and the service data of each service application server system at the corresponding historical moment train the pre-established machine learning model 2. Randomly dividing the original amount of screened historical time service data and historical time mobility coverage rate data into K non-coincident subdata sets, selecting K-1 sets as training sets each time, and taking the remaining 1 sets as verification sets. Respectively inputting K-1 selected each time as a training set into a pre-established machine learning model in an iterative mode, and performing initial training; and (5) optimizing the initially trained machine learning model by using the remaining 1 data set as a test sample. And training through a neural network clustering algorithm to obtain a service data configuration scheme calculation model.
Step S1024: and calculating the mobility coverage rate of the predicted future time by using the trained machine learning model 1 mobility coverage rate prediction model. Processing data of service data of each service system of the bank, of which the mobility coverage is to be determined, and calculating and predicting the mobility coverage of the future time; and early warning the market risk according to the predicted liquidity coverage rate and a preset liquidity coverage rate early warning rule.
Step S1025: and early warning the mobility coverage rate risk according to the predicted mobility coverage rate and a preset mobility coverage rate early warning rule.
Step S1026: and calculating to obtain a plurality of business data configuration scheme data by using the trained machine learning model 2 according to the future time mobility coverage rate prediction data obtained in the step S1025.
Step S1027: and respectively carrying out component proportion similarity matching on the plurality of service data configuration schemes and the predicted service data to obtain the similarity between the plurality of service data configuration schemes and the predicted service data.
Step S1028: and obtaining a service data adjusting scheme according to the similarity and the user preference.
The present invention also provides a mobile coverage prediction apparatus, comprising: the device comprises a data acquisition module, a training module, a prediction module, a configuration scheme calculation module and a matching module; the data acquisition module is used for acquiring service data of each service application server system at historical time and mobility coverage rate data at historical time; the training module is used for training a pre-established machine learning model according to the business data at the historical moment and the mobility coverage rate data at the historical moment to obtain a first training model; screening the mobility coverage rate data and the service data at the corresponding moment according to a preset condition, and training a pre-established machine learning model according to a screening result to obtain a second training model; the prediction module is used for obtaining a mobility coverage rate prediction value according to the first training model and the predicted service data; comparing the predicted value with a preset threshold value, and providing the mobility coverage rate predicted value to the second training module according to a comparison result; the configuration scheme calculation module is used for substituting the mobility coverage rate predicted value into the second training model to obtain a plurality of service data configuration scheme data; and the matching module is used for obtaining a business data adjusting scheme according to the comparison result of the business data configuration scheme data and the predicted business data.
In the above embodiment, the data acquisition module includes a data processing unit and a feature value extraction unit, and the data processing unit is configured to integrate, screen, process missing data, and reject abnormal data for the business data of each business system at the historical time of the bank and the liquidity coverage data at the historical time. The characteristic value extraction unit is used for obtaining an initial characteristic value attribute of basic service data of the mobility coverage rate prediction data.
In an embodiment of the present invention, the training module may include a first model training unit, where the first model training unit is configured to randomly segment the eigenvalue information and the mobility coverage data at the historical time into K non-coincident sub data sets, use K-1 sub data sets as a training set, and use the remaining 1 as a verification set; respectively inputting the training set selected each time into a pre-established machine learning model for training in an iterative mode, and performing model optimization by using a corresponding verification set; obtaining a first training model through multiple linear regression training of a genetic algorithm and a support vector machine regression algorithm; k is a positive integer greater than 1. Wherein the training module may further comprise a second model training unit: and acquiring preference data of a user, training a pre-established machine learning model according to the preference data and the screening result, and training through a neural network clustering algorithm to obtain a second training model.
In an embodiment of the present invention, the configuration scheme calculating module further includes: and substituting the mobility coverage rate predicted value and the preference degree data into the second training model according to a comparison result to obtain a plurality of service data configuration scheme data.
For the sake of better clarity, the flowing coverage rate predicting device provided by the present invention is further described with reference to the accompanying drawings, and it should be understood by those skilled in the art that the example is only for the understanding of the flowing coverage rate predicting device provided by the present invention, and is not further limited thereto.
Referring to fig. 4, the mobile coverage prediction apparatus provided in the present invention may include a data obtaining module 201, a machine learning 1 module 202, a machine learning 2 module 203, a mobile coverage calculation module 204, a mobile coverage pre-warning module 205 of the prediction module, a service data configuration scheme calculation module 206, a component similarity matching module 207, and a data display module 208 in practical application. The machine learning 1 module 202 and the machine learning 2 module 203 are training modules, the mobility coverage calculation module 204 and the prediction module mobility coverage early warning module 205 are prediction modules, the service data configuration scheme calculation module 206 is a configuration scheme calculation module, and the component similarity matching module 207 and the data display module 208 are matching modules.
Referring to fig. 1B, the data obtaining module 201 is mainly configured to perform the step S1021, that is, obtain service data of each service application server system at the historical time of the bank and liquidity coverage data at the historical time.
Specifically, as shown in fig. 5, the device includes a data obtaining unit 11, a data processing unit 12, a feature value obtaining unit 13, and a data transmission unit 14.
The data acquisition unit 11: and receiving and loading service data of various service systems of the bank. Such as deposit system business data, loan system business data, wind control system business data, clearing system business data, financial market system business data and the like, form a big data set as source data of the machine learning 1 module.
The data processing unit 12: and integrating and screening the service data of each service system at the historical time of the bank and the liquidity coverage rate data at the historical time of the bank, processing missing data and eliminating abnormal data. And (4) combing the missing data of the system, and if the missing data exists, carrying out interpolation processing according to the date data before and after the missing data. The interpolation mode can select linear interpolation, that is, the feature value of the current day is the average value of the data of the previous and next two days, that is, the interpolation mode can select linear interpolation
xi-1、xi+1Data of two days before and after, xiThe data of the current day; if a certain date is characteristicIf the value is sharply increased or decreased, for example, if the difference is extremely large within a certain range before or after the occurrence, the data is considered to be caused by an emergency or an accidental event, and the data is discarded.
The eigenvalue acquisition unit 13: as shown in table 1 below, the initial eigenvalue is imported, and is an initial eigenvalue attribute and a corresponding attribute value table of the mobility coverage prediction data base service data in this embodiment.
TABLE 1
The characteristic data known by the fluidity coverage rate of qualified high-quality flowing assets/cash net outflow amount of 30 days in the future 100% comprises two types, wherein one type is qualified high-quality flowing asset service data, P1 is used for representing the characteristic value data of the qualified high-quality flowing asset service data, weight is represented by W1, and then the characteristic data set is W1P 1; the other is the cash net outflow traffic data of a future period of time, the cash net outflow is cash outflow-cash inflow, the characteristic value data as the cash net outflow traffic data of the future period of time is represented by P2, and the weight is represented by W2, so that the characteristic data set is W2P 2. The eigenvalue vector of the fluidity coverage is expressed as
The data transmission unit 14: and the characteristic value data obtained by processing is transmitted to the next module.
Referring to fig. 1B, the machine learning 1 module 202 is mainly configured to execute the step S1022, that is, train the machine learning model 1 established in advance according to the business data of each business application server system at the historical time of the bank and the liquidity coverage data at the historical time, and specifically may include a feature value importing unit, a historical data importing unit, and a model training unit to establish the learning model.
As shown in fig. 6, the characteristic value data importing unit 21: the feature value is imported, and the feature value data obtained by the device module 201 corresponding to step S1021 is imported.
History result data importing unit 22: and matching historical mobility coverage result data aiming at different time points of the characteristic value data by taking the historical mobility coverage data at the moment as a training set for machine learning.
Machine learning model 1 training unit 23: the machine learning model 1 is a mobility coverage prediction model, and in this case, a genetic algorithm GA model and a support vector machine regression (SVR) model are used as dual-function models. And the GA algorithm model is used for obtaining the optimal service data configuration scheme corresponding to the corresponding predicted value at the future moment. At a certain time in the GA algorithm model, a feature attribute set WP of banking data represents an individual, wherein P is feature data, W is a corresponding weight, and a certain value W in WPiPiRepresents one of the genetic factors, i.gtoreq.1. The suitability evaluation function is Fn. The species are divided according to individual similarity (multiple individuals make up one species). And then the adaptive evaluation function carries out target function confidence evaluation on the individual to obtain confidence evaluation. The evolution process of the GA algorithm model is survival competition, the same species are superior and inferior, a certain percentage of individuals are reserved, then competition and selection are carried out, genetic variation among species is used for reproducing offspring individuals, and the individuals with high environmental adaptability survive. After multiple iterative evolutions, the confidence evaluation of the objective function is optimal, and the set WP at the moment is the optimal individual genetic combination at the current moment. The regression confidence of the SVR model is used as a convergence target in the GA model. I.e., the regression equation f (x) of the SVR model, is the objective function of the GA model. The SVR model is a multiple linear regression of a support vector machine regression (SVR) algorithm, and by giving a new input sample x, the output Y corresponding to the input sample x is deduced according to the given data sample, and the output Y is a real number. The regression problem can be described in mathematical language as: given a set of data samples is { (x)i,yi)|xi∈Rn,yiE.g. R, i ═ 1,2,3. Finding RnThe above function f (x) yields a regression equation to infer the y value for any x input using y ═ f (x).
Prediction result analysis unit 24: the prediction data and the historical liquidity coverage data obtained by the machine learning model 1 training unit 23 are imported into the historical result data import unit 22 as samples. Randomly dividing all the historical time service data and the historical time mobility coverage rate data into K misaligned sub-data sets, selecting K-1 data sets as training samples each time, and taking the remaining 1 data set as a verification sample. And obtaining a regression equation by adjusting the iteration times and the test sample deviation.
The mobility coverage calculation module 204 and the prediction module mobility coverage early-warning module 205 correspond to steps S1024 and S1025 in fig. 1B, where the mobility coverage prediction module is configured to predict mobility coverage data by using the trained machine learning model 1, and includes a data importing unit, a mobility coverage calculation and prediction processing unit, a mobility coverage early-warning unit, and a data transmission unit, which are specifically shown in fig. 8.
And the data import unit 41 imports business data of all business systems of the banks, which are to be calculated and predicted to be the liquidity coverage rate.
The mobile coverage rate calculation and prediction processing unit 42 is used for loading data of the banking business data server by the system every day, forming a characteristic value sample, calculating and predicting mobile coverage rate by using the trained machine learning model 1, predicting the mobile coverage rate of each business system business data of the bank to be calculated and predicted the mobile coverage rate, and generating the mobile coverage rate of the future time;
the calculation and prediction management unit 43 for mobility coverage is used for configuring the early warning parameters and the early warning model. Calling a machine learning server through the mobility coverage calculation and prediction processing module 42 to generate future mobility coverage data, and performing early warning according to the future data; the early warning model is used for setting a liquidity coverage early warning threshold (for example, the liquidity coverage required by supervision cannot be lower than 100%) as a lower limit, and generating an early warning that the liquidity coverage is too low if the liquidity coverage is lower than the lower limit.
The user sets a mobility coverage early warning parameter, an early warning reminding mode and frequency through the mobility coverage early warning unit 44, and performs early warning (highlighting, short message or mail reminding and other modes) on the prediction data reaching the early warning threshold.
The flow coverage prediction data structure may be as follows:
the data transmission unit 45 transmits the prediction data of the future mobility coverage rate and the early warning information to show that the data and the early warning information are both displayed, and the information is transmitted to the next module for processing.
The machine learning 2 module 203 is similar to the machine learning 1 module 202, and is configured to train the pre-established machine learning model 2 according to the set conditions, the service data of each service application server system at the historical time of the screened bank and the mobility coverage data at the historical time, and includes a feature value importing unit, a historical data importing unit, a model training unit, and a transmission unit.
Specifically, as shown in fig. 7, the characteristic value data importing unit 31 imports the characteristic value data obtained as described above as a data characteristic value.
History result data importing unit 32: and matching the screened historical mobility coverage result data aiming at different time points of the characteristic value data by taking the screened historical mobility coverage data as a training set for machine learning.
Machine learning model 2 training unit 33: and (3) data classification training: the characteristic value data import 31 and the historical time fluidity coverage data import 32 are used as samples. And classifying the data by adopting a clustering mode to realize the optimal aggregation of the data of each service product. In the case, a neural network clustering algorithm SOM is adopted to realize product data clustering, and the aggregation process can be described as follows: the resulting large dataset in block 202 forms a high-dimensional dataset as input and output samples of data samples due to the complex data types. Randomly selecting data from an output sample as an initial value, then selecting an input vector from the input sample according to random probability, finding a weight vector with the minimum distance from the input vector to define the weight vector as a local optimal set, and then adjusting the weight in a region near the local optimal set to approach the input vector. And taking factors such as the radius of a contraction domain, a learning rate, a repetition rate, user preference and the like as evaluation function factors. Model training is performed by iteratively repeating the above process, so that the evaluation is optimal, and the optimal aggregation classification is performed at this time. And (3) data future prediction training: the data aggregation classification is a data set. The historical mobility coverage data at each moment can obtain an optimal aggregation classification data set, namely an optimal service data configuration scheme corresponding to the mobility coverage at the time point. An evaluation score is made based on the entered liquidity coverage value (i.e., corresponding to the predicted value of liquidity coverage in block 205) and the user preference. Generally, the liquidity coverage rate data at the historical moment is compared with the input liquidity coverage rate data, and the larger the value is, the higher the score is; the higher the user's preference for the optimal aggregate classification, the higher the score.
The result transmission unit 34: the prediction data and the historical mobility coverage data obtained as described above are imported as samples. Randomly dividing all the historical time service data and the historical time mobility coverage rate data into K misaligned sub-data sets, selecting K-1 data sets as training samples each time, and taking the remaining 1 data set as a verification sample. And obtaining an optimal configuration scheme by adjusting the calculation times and the test sample deviation.
The service data configuration scheme calculation module 206 is mainly configured to calculate and obtain a plurality of service data configuration scheme data by using the trained machine learning model 2 according to the obtained future time mobility coverage rate prediction data. Specifically, as shown in fig. 9, the data importing unit 51, the service data configuration scheme calculating unit 52, and the data transmitting unit 53 are included; the data importing unit 51 is configured to receive data output by the mobile coverage early warning module 205, deliver the data to the service data configuration scheme calculating unit 52 to calculate and obtain a plurality of service data configuration scheme data, and then provide the data to the data transmitting unit 53 and deliver the data to the component similarity matching module 207.
The component similarity matching module 207 is mainly used for performing component proportion similarity matching on the plurality of service data configuration schemes and the predicted service data respectively to obtain the similarity between the plurality of service data configuration schemes and the predicted service data; as shown in fig. 10, the method may include:
the data import unit 71: and receiving the data transmitted by the imported previous module. Component ratio similarity matching unit 72: and performing service component analysis on the configuration scheme to obtain a future configuration scheme, and performing similarity analysis on the future configuration scheme and a plurality of configuration schemes obtained by a previous module to obtain a similarity score value. The data transmission unit 73: and transmitting the obtained similarity data and the corresponding service configuration scheme to the next module.
Referring to fig. 11, the data display module 208 is mainly used for obtaining a service data adjustment scheme according to the similarity and the user preference, and specifically includes:
the data importing unit 81: and receiving the data transmitted by the previous module.
The user interaction unit 82: the method is used for interacting with a user, obtaining the preference of the user to a business data configuration scheme, and feeding back the preference to the machine learning 2 model for being used as a subsequent data training sample.
Service data configuration adjustment scheme presentation unit 83: and comparing the obtained optimal configuration scheme with the configuration scheme at the future moment to obtain a service data configuration adjustment scheme, and displaying the service data configuration adjustment scheme.
The invention has the beneficial technical effects that: on one hand, a mobility coverage rate target can be determined according to the self operation condition, and on the other hand, when the actual mobility coverage rate deviates from the target, an adjustment measure can be actively adopted to reduce the deviation degree, so that the resistance capability of the mobility coverage rate risk under external impact is improved; moreover, the probability that the mobility coverage rate does not reach the standard is reduced through the prediction of the mobility coverage rate; the mobility coverage rate prediction can also provide certain research reference for mobility coverage rates of other emerging markets, and is wide in application range and high in transportability.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method when executing the computer program.
The present invention also provides a computer-readable storage medium storing a computer program for executing the above method.
As shown in fig. 12, the electronic device 600 may further include: communication module 110, input unit 120, audio processing unit 130, display 160, power supply 170. It is noted that the electronic device 600 does not necessarily include all of the components shown in fig. 12; furthermore, the electronic device 600 may also comprise components not shown in fig. 12, which may be referred to in the prior art.
As shown in fig. 12, the central processor 100, sometimes referred to as a controller or operational control, may include a microprocessor or other processor device and/or logic device, the central processor 100 receiving input and controlling the operation of the various components of the electronic device 600.
The memory 140 may be, for example, one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. The information relating to the failure may be stored, and a program for executing the information may be stored. And the central processing unit 100 may execute the program stored in the memory 140 to realize information storage or processing, etc.
The input unit 120 provides input to the cpu 100. The input unit 120 is, for example, a key or a touch input device. The power supply 170 is used to provide power to the electronic device 600. The display 160 is used to display an object to be displayed, such as an image or a character. The display may be, for example, an LCD display, but is not limited thereto.
The memory 140 may be a solid state memory such as Read Only Memory (ROM), Random Access Memory (RAM), a SIM card, or the like. There may also be a memory that holds information even when power is off, can be selectively erased, and is provided with more data, an example of which is sometimes called an EPROM or the like. The memory 140 may also be some other type of device. Memory 140 includes buffer memory 141 (sometimes referred to as a buffer). The memory 140 may include an application/function storage section 142, and the application/function storage section 142 is used to store application programs and function programs or a flow for executing the operation of the electronic device 600 by the central processing unit 100.
The memory 140 may also include a data store 143, the data store 143 for storing data, such as contacts, digital data, pictures, sounds, and/or any other data used by the electronic device. The driver storage portion 144 of the memory 140 may include various drivers of the electronic device for communication functions and/or for performing other functions of the electronic device (e.g., messaging application, address book application, etc.).
The communication module 110 is a transmitter/receiver 110 that transmits and receives signals via an antenna 111. The communication module (transmitter/receiver) 110 is coupled to the central processor 100 to provide an input signal and receive an output signal, which may be the same as in the case of a conventional mobile communication terminal.
Based on different communication technologies, a plurality of communication modules 110, such as a cellular network module, a bluetooth module, and/or a wireless local area network module, may be provided in the same electronic device. The communication module (transmitter/receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and receive audio input from the microphone 132 to implement general telecommunications functions. Audio processor 130 may include any suitable buffers, decoders, amplifiers and so forth. In addition, an audio processor 130 is also coupled to the central processor 100, so that recording on the local can be enabled through a microphone 132, and so that sound stored on the local can be played through a speaker 131.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention 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 invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (14)
1. A method for mobility coverage prediction, the method comprising:
acquiring service data of each service application server system at a historical moment and mobility coverage rate data at the historical moment;
training a pre-established machine learning model according to the business data at the historical moment and the mobility coverage rate data at the historical moment to obtain a first training model;
screening the mobility coverage rate data and the service data at the corresponding moment according to a preset condition, and training a pre-established machine learning model according to a screening result to obtain a second training model;
obtaining a mobility coverage rate predicted value according to the first training model and the predicted service data;
comparing the predicted value with a preset threshold value, and substituting the mobility coverage rate predicted value into the second training model according to a comparison result to obtain a plurality of service data configuration scheme data;
and obtaining a service data adjustment scheme according to the comparison result of the service data configuration scheme data and the predicted service data.
2. The method of predicting flowable coverage as set forth in claim 1, wherein obtaining traffic data of each of the traffic application server systems at a historical time and flowable coverage data at the historical time comprises:
and screening abnormal data in the service data to obtain preprocessed service data serving as source data.
3. The mobility coverage prediction method of claim 1, wherein training a pre-established machine learning model according to the traffic data at the historical time and the mobility coverage data at the historical time to obtain a first training model comprises:
dividing the preprocessed service data and the mobility coverage rate data at the historical moment into K non-coincident sub data sets at random, taking K-1 sub data sets as training sets, and taking the remaining 1 sub data sets as verification sets;
respectively inputting the training set selected each time into a pre-established machine learning model in an iterative mode, training, and performing model learning optimization by using a corresponding verification set;
the method comprises the steps of performing multivariate linear regression integrated learning training through a genetic algorithm and a support vector machine regression algorithm to obtain a first training model;
k is a positive integer greater than 1.
4. The method of claim 1, wherein obtaining a mobility coverage prediction value based on the first training model and the predicted traffic data comprises:
and calculating to obtain a mobility coverage rate predicted value through the feature value information corresponding to the predicted service data and the first training model.
5. The method of claim 1, wherein obtaining a traffic data adjustment scheme according to the comparison of the traffic data configuration scheme data and the predicted traffic data comprises:
respectively carrying out component proportion similarity matching on the plurality of service data configuration scheme data and the predicted service data to obtain the similarity between each service data configuration scheme data and the predicted service data;
and screening the service data configuration scheme data according to the similarity to obtain a service data adjustment scheme meeting preset conditions.
6. The method of predicting flowable coverage as set forth in claim 1, wherein training the pre-established machine learning model according to the screening result further comprises: and acquiring preference data of a user, and training a pre-established machine learning model according to the preference data and the screening result to obtain a second training model.
7. The method of claim 6, wherein substituting the mobility coverage prediction value into the second training model according to the comparison result to obtain a plurality of service data configuration scenario data further comprises: and substituting the mobility coverage rate predicted value and the preference degree data into the second training model according to a comparison result to obtain a plurality of service data configuration scheme data.
8. An apparatus for predicting fluid coverage, the apparatus comprising: the device comprises a data acquisition module, a training module, a prediction module, a configuration scheme calculation module and a matching module;
the data acquisition module is used for acquiring service data of each service application server system at historical time and mobility coverage rate data at historical time;
the training module is used for training a pre-established machine learning model according to the business data at the historical moment and the mobility coverage rate data at the historical moment to obtain a first training model; screening the mobility coverage rate data and the service data at the corresponding moment according to a preset condition, and training a pre-established machine learning model according to a screening result to obtain a second training model;
the prediction module is used for obtaining a mobility coverage rate prediction value according to the first training model and the predicted service data; comparing the predicted value with a preset threshold value, and providing the mobility coverage rate predicted value to the second training module according to a comparison result;
the configuration scheme calculation module is used for substituting the mobility coverage rate predicted value into the second training model to obtain a plurality of service data configuration scheme data;
and the matching module is used for obtaining a business data adjusting scheme according to the comparison result of the business data configuration scheme data and the predicted business data.
9. The mobile coverage prediction device of claim 8, wherein the data acquisition module comprises a data processing unit and a feature value extraction unit, and the data processing unit is configured to integrate, screen, process missing data, and reject abnormal data for the service data of each service system at the historical time and the mobile coverage data at the historical time to obtain preprocessed service data; the characteristic value extraction unit is used for acquiring initial characteristic value attribute information of basic service data of the mobility coverage rate prediction data.
10. The mobile coverage prediction device of claim 9, wherein the training module comprises a first model training unit, and the first model training unit is configured to randomly divide the eigenvalue information and the mobile coverage data at the historical time into K disjoint sub data sets, take K-1 sub data sets as a training set, and take the remaining 1 sub data sets as a verification set; respectively inputting the training set selected each time into a pre-established machine learning model for training in an iterative mode, and performing model learning optimization by using a corresponding verification set; obtaining a first training model through multivariate linear regression ensemble learning training of a genetic algorithm and a support vector machine regression algorithm;
k is a positive integer greater than 1.
11. The mobile coverage prediction device of claim 8, wherein the training module further comprises:
and acquiring preference data of a user, training a pre-established machine learning model according to the preference data and the screening result, and learning and training through a neural network clustering algorithm to acquire a second training model.
12. The mobile coverage prediction device of claim 11, wherein the configuration scheme calculation module further comprises: and substituting the mobility coverage rate predicted value and the preference degree data into the second training model according to a comparison result to obtain a plurality of service data configuration scheme data.
13. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any of claims 1 to 7 when executing the computer program.
14. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program for executing the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110036773.5A CN112734565B (en) | 2021-01-12 | 2021-01-12 | Fluidity coverage prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110036773.5A CN112734565B (en) | 2021-01-12 | 2021-01-12 | Fluidity coverage prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112734565A true CN112734565A (en) | 2021-04-30 |
| CN112734565B CN112734565B (en) | 2024-03-05 |
Family
ID=75590556
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110036773.5A Active CN112734565B (en) | 2021-01-12 | 2021-01-12 | Fluidity coverage prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112734565B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113657535A (en) * | 2021-08-24 | 2021-11-16 | 北京奇艺世纪科技有限公司 | Model training method and device, electronic equipment and storage medium |
| CN114153399A (en) * | 2021-12-07 | 2022-03-08 | 四川云从天府人工智能科技有限公司 | Data storage method, device, control device and medium of storage system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10255550B1 (en) * | 2017-06-07 | 2019-04-09 | States Title, Inc. | Machine learning using multiple input data types |
| CN111882428A (en) * | 2020-07-27 | 2020-11-03 | 上海应用技术大学 | Commercial bank liquidity risk assessment method and device |
-
2021
- 2021-01-12 CN CN202110036773.5A patent/CN112734565B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10255550B1 (en) * | 2017-06-07 | 2019-04-09 | States Title, Inc. | Machine learning using multiple input data types |
| CN111882428A (en) * | 2020-07-27 | 2020-11-03 | 上海应用技术大学 | Commercial bank liquidity risk assessment method and device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113657535A (en) * | 2021-08-24 | 2021-11-16 | 北京奇艺世纪科技有限公司 | Model training method and device, electronic equipment and storage medium |
| CN114153399A (en) * | 2021-12-07 | 2022-03-08 | 四川云从天府人工智能科技有限公司 | Data storage method, device, control device and medium of storage system |
| CN114153399B (en) * | 2021-12-07 | 2023-10-20 | 四川云从天府人工智能科技有限公司 | Data storage method, device, control device and medium of storage system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112734565B (en) | 2024-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200401946A1 (en) | Management and Evaluation of Machine-Learned Models Based on Locally Logged Data | |
| CN111275546B (en) | Financial customer fraud risk identification method and device | |
| KR102299319B1 (en) | Computer program stored on computer-readable medium, method and server for classifying a client intent by a chat bot | |
| CN111951044A (en) | Bank terminal interaction method and system | |
| CN111783873B (en) | User portrait method and device based on increment naive Bayes model | |
| US20190220710A1 (en) | Data processing method and data processing device | |
| CN111582341B (en) | User abnormal operation prediction method and device | |
| CN112734565B (en) | Fluidity coverage prediction method and device | |
| CN106104427B (en) | The reformatting of the perceived content of input | |
| US20190362354A1 (en) | Real-time updating of predictive analytics engine | |
| CN112766825A (en) | Enterprise financial service risk prediction method and device | |
| US11151467B1 (en) | System, method, and computer program for generating intelligent automated adaptive decisions | |
| CN110716767B (en) | Model component calling and generating method, device and storage medium | |
| US20240046922A1 (en) | Systems and methods for dynamically updating machine learning models that provide conversational responses | |
| CN113191880A (en) | Bank teller terminal cash adding suggestion determination method and device | |
| CN115562960A (en) | Application optimization method and system of intelligent equipment | |
| CN111768306B (en) | Risk identification method and system based on intelligent data analysis | |
| KR102276391B1 (en) | Computer program for providing a way to respond to customers | |
| CN111401395B (en) | Data processing method, terminal equipment and storage medium | |
| US11928629B2 (en) | Graph encoders for business process anomaly detection | |
| CN116957646A (en) | Method and device for determining willingness to bind card | |
| US11985102B2 (en) | Processing clusters with mathematical models for message suggestion | |
| CN115713354A (en) | Mobile banking user loss early warning method and system | |
| CN117951547A (en) | Bid and tendered data processing method and device based on artificial intelligence | |
| CN112085495A (en) | Processing method and device of quota account data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |