CN109741818A - Resource allocation management method and device are intervened in medical inferior health based on artificial intelligence - Google Patents
Resource allocation management method and device are intervened in medical inferior health based on artificial intelligence Download PDFInfo
- Publication number
- CN109741818A CN109741818A CN201910024268.1A CN201910024268A CN109741818A CN 109741818 A CN109741818 A CN 109741818A CN 201910024268 A CN201910024268 A CN 201910024268A CN 109741818 A CN109741818 A CN 109741818A
- Authority
- CN
- China
- Prior art keywords
- training sample
- model
- accuracy rate
- demand model
- demand
- 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
- 230000036541 health Effects 0.000 title claims abstract description 34
- 238000013468 resource allocation Methods 0.000 title claims abstract description 33
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 30
- 238000007726 management method Methods 0.000 title claims abstract description 23
- 238000012549 training Methods 0.000 claims abstract description 200
- 238000000034 method Methods 0.000 claims abstract description 40
- 238000005303 weighing Methods 0.000 claims abstract description 30
- 238000012360 testing method Methods 0.000 claims abstract description 19
- 210000002569 neuron Anatomy 0.000 claims description 22
- 238000013528 artificial neural network Methods 0.000 claims description 14
- 230000002068 genetic effect Effects 0.000 claims description 13
- 230000007246 mechanism Effects 0.000 claims description 13
- 230000015572 biosynthetic process Effects 0.000 claims description 11
- 238000005259 measurement Methods 0.000 claims description 11
- 238000003786 synthesis reaction Methods 0.000 claims description 10
- 230000035772 mutation Effects 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 4
- 231100000219 mutagenic Toxicity 0.000 claims description 3
- 230000003505 mutagenic effect Effects 0.000 claims description 3
- 210000005036 nerve Anatomy 0.000 claims description 2
- 230000006870 function Effects 0.000 description 14
- 230000008569 process Effects 0.000 description 14
- 230000002354 daily effect Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 6
- 238000004590 computer program Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000003750 conditioning effect Effects 0.000 description 3
- 239000003814 drug Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000000554 physical therapy Methods 0.000 description 3
- 208000017667 Chronic Disease Diseases 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000018199 S phase Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000001684 chronic effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 230000005180 public health Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000001755 vocal effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The embodiment of the present application provides a kind of medical inferior health intervention resource allocation management method and device based on artificial intelligence, by obtained from database with the matched universal model of Target Enterprise, be trained according to universal model of the training sample of Target Enterprise to acquisition to obtain demand model corresponding with Target Enterprise.Demand model is tested according to the test sample of Target Enterprise, to obtain the accuracy rate of demand model, it modifies further according to the accuracy rate of demand model to the weighing factor and training sample of impact factor each in training sample, continue to be trained demand model using modified training sample, until the accuracy rate of obtained demand model meets default accuracy rate requirement.Pass through the resource allocation method, it can be on the basis of existing universal model, it carries out personalized training in conjunction with the feature of itself enterprise must to reach expected requirement and the matched demand model of itself enterprise, deploys for the medical resource of the subsequent enterprise and provide effective foundation.
Description
Technical field
The present invention relates to big data processing technology fields, sub- strong in particular to a kind of medical treatment based on artificial intelligence
Health intervenes resource allocation management method and device.
Background technique
Planning commission's statistical data is defended according to country to show, it is Chinese at present because chronic disease causes death toll to account about total dead people
Several 85%, chronic disease in Disease Spectrum proportion up to 70%, it has also become great public health problem and social concern, sternly
Ghost image rings the health of the our people masses.Especially workplace crowd, it is chronic due to the factor of vocational need and personal habits
Disease is high-incidence, and a large amount of personnel are in sub-health state.For this demand, many large and medium-sized enterprise are provided with employee in working place
Rehabilitation physical therapy room introduces experienced conditioning teacher, and employee is helped to improve.
Since the hardware condition of employee's rehabilitation physical therapy room of enterprise can not be mentioned in the same breath with medical institutions, determines and do not need
The Chinese medicine conditioning of a large amount of biochemical diagnosis becomes the first choice of enterprise.However, Chinese medicine conditioning need experience old docter of TCM more abundant and
Physical Therapist, and veteran old docter of TCM and Physical Therapist are scarce talents, can not largely be matched in employee's rehabilitation physical therapy room of enterprise
It sets.
This is easy for causing a problem, and supply falls short of demand for busy, and drug on the market for idle.Cause loss and the medical treatment of client sub-
The waste of Health intervention resource, to influence economic benefit and customer satisfaction rate.
Summary of the invention
In view of this, the purpose of the application is, a kind of medical inferior health intervention resource tune based on artificial intelligence is provided
With management method and device to improve the above problem.
The embodiment of the present application provides a kind of medical inferior health intervention resource allocation management method based on artificial intelligence, described
Method includes:
It is obtained from multiple universal models of database and the matched universal model of Target Enterprise；
The training sample of Target Enterprise is obtained, the training sample carries multiple and different impact factors, each influence
The factor carries corresponding weighing factor, and the training sample includes the employee's characteristic and medical resource allotment of the Target Enterprise
Information；
The training sample is pre-processed, pretreated training sample is directed into the universal model to described
Universal model is trained, to obtain the corresponding demand model of the Target Enterprise；
The test sample for obtaining Target Enterprise, the test sample is directed into the demand model, according to the need
The output result of modulus type obtains the accuracy rate of the demand model；
When the accuracy rate is unsatisfactory for default accuracy rate requirement, according to the accuracy rate of the demand model to the training
It the weighing factor of each impact factor and modifies to the training sample in sample；
The demand model is trained using modified training sample, until the accuracy rate of obtained demand model
Until meeting the default accuracy rate requirement.
Optionally, the step with the matched universal model of Target Enterprise is obtained in multiple universal models from database
Suddenly, comprising:
It is found from multiple universal models of the database according to the type of business of Target Enterprise and is looked forward to the target
The matched universal model of the type of business of industry；
Situation is constituted according to the department in the Target Enterprise to select from the universal model found and the target
The department of enterprise constitutes the consistent universal model of situation.
Optionally, each impact factor includes multiple features, described to carry out pretreated step to the training sample,
Include:
Whether the synthesis output amplitude for detecting the multiple impact factor in the training sample is in preset range；
If being not in the preset range, each impact factor is calculated separately relative in the universal model
The variance yields of the synthesis output amplitude of impact factor；
Detect whether the corresponding variance yields of each impact factor is greater than preset threshold, if more than the preset threshold, then
The impact factor that preset quantity is randomly selected from the multiple impact factor, to each shadow in the impact factor extracted
The feature for ringing the factor is normalized with the data volume of each impact factor of simplification.
Optionally, the universal model be based on neural network constructed by, the neural network include input layer, output layer and
Hidden layer, the input layer, output layer and hidden layer respectively include multiple neurons, the input layer, output layer and hidden layer
Between neuron there are connection weight weight values, it is described that pretreated training sample is directed into the universal model to described logical
The step of being trained with model, comprising:
Pretreated training sample is directed into the input layer of the neural network, by the training of the hidden layer
Afterwards, it is exported from the output layer；
Whether the result for detecting the output layer output reaches expected results, if not up to expected results, according to
The result of output and the expected results obtain error signal, and enter back-propagation phase；
Using the error signal as the input signal of back-propagation phase with anti-from the output layer to the input layer
It returns back, during reversed passback, corrects the connection weight of the neuron between the input layer, output layer and hidden layer
Value is to gradually decrease the error signal of final output.
Optionally, described during reversed passback, correct the nerve between the input layer, output layer and hidden layer
The step of error signal of the connection weight weight values to gradually decrease final output of member, comprising:
During reversed passback, the mind between the input layer, output layer and hidden layer is modified using following formula
The error signal of final output is gradually decreased through first connection weight weight values:
Wherein, WijIndicate i-th of neuron of input layer to the connection weight weight values between j-th of hidden layer upgrading, XPIt indicates
The P training sample input layer i-th of input value,Indicate the threshold value of j-th of neuron of hidden layer.
Optionally, shadow of the accuracy rate according to the demand model to the impact factor each in the training sample
The step of ringing weight and modifying to the training sample, comprising:
The accuracy rate of obtained demand model in the accuracy rate of currently available demand model and history number is carried out
It compares；
If the accuracy rate of currently available demand model is higher than the demand model of more than half obtained in history number
Accuracy rate then retains current training sample, and modifies to the weighing factor of each impact factor；
If the accuracy rate of currently available demand model is lower than the demand model of more than half obtained in history number
Part training sample in the training sample is then deleted, and adds new training sample by accuracy rate.
Optionally, shadow of the accuracy rate according to the demand model to the impact factor each in the training sample
After the step of ringing weight and modifying to the training sample, the method also includes:
Fitness function is established to assess the fitness value of each training sample in modified training sample, utilizes something lost
The selection mechanism of propagation algorithm selects the highest training sample of fitness value；
Any two training sample is randomly choosed from multiple training samples using the crossover mechanism of genetic algorithm to be handed over
Fork, to obtain follow-on training sample；
The highest training sample of fitness value in the follow-on training sample is calculated using the fitness function；
Detect the follow-on training sample fitness value whether be lower than previous generation training sample fitness value,
If being lower than, mutagenic factor is introduced to carry out variation behaviour to the follow-on training sample using the Variation mechanism of genetic algorithm
Make, then calculates the fitness value of the training sample after mutation operation；
Training sample is modified again according to the fitness value of training sample.
Optionally, described that the demand model is trained using modified training sample, until obtained demand
After the accuracy rate of model meets the step of until the default accuracy rate requires, the method also includes:
Receive it is currently entered to measurement information, it is described that multiple and different impact factors, each shadow are carried to measurement information
It rings the factor and carries corresponding weighing factor；
The demand model is directed into measurement information predicts, by described to obtain prediction result；
Receive the feedback information of the prediction result to the demand model of user's input；
It is adjusted according to model parameter of the feedback information to the demand model.
Optionally, described after the step of until the accuracy rate until obtained demand model meets preset requirement
Method further include:
The demand model that the default accuracy rate of obtained satisfaction requires is stored in the database, in the database
Data be updated.
The embodiment of the present application also provides a kind of medical inferior health intervention resource allocation managing device based on artificial intelligence, institute
Stating device includes:
Universal model obtains module, matched general with Target Enterprise for obtaining from multiple universal models of database
Model；
Training sample obtains module, and for obtaining the training sample of Target Enterprise, the training sample carries multiple and different
Impact factor, each impact factor carries corresponding weighing factor, and the training sample includes the Target Enterprise
Employee's characteristic and medical resource deployment information；
Demand model obtains module and leads pretreated training sample for pre-processing to the training sample
Enter to the universal model and the universal model is trained, to obtain the corresponding demand model of the Target Enterprise；
Accuracy rate obtains module and the test sample is directed into the need for obtaining the test sample of Target Enterprise
In modulus type, the accuracy rate of the demand model is obtained according to the output result of the demand model；
Modified module, for when the accuracy rate is unsatisfactory for default accuracy rate and requires, according to the standard of the demand model
True rate to the weighing factor of the impact factor each in the training sample and modifies to the training sample；
Training module, for being trained using modified training sample to the demand model, until obtained need
Until the accuracy rate of modulus type meets the default accuracy rate requirement.
The embodiment of the present application also provides a kind of electronic equipment, including memory, processor and is stored on the memory
And the computer program that can be run on the processor, the processor realize above-mentioned method step when executing described program
Suddenly.
Resource allocation management method and device are intervened in medical inferior health provided by the embodiments of the present application based on artificial intelligence,
First by obtained from database with the matched universal model of Target Enterprise, further according to the training sample of the Target Enterprise of acquisition
The universal model of acquisition is trained to obtain demand model corresponding with Target Enterprise.And then according to the Target Enterprise of acquisition
Test sample obtained demand model is tested, to obtain the accuracy rate of demand model, further according to the standard of demand model
True rate modifies to the weighing factor and training sample of impact factor each in training sample, using modified training sample after
It is continuous that demand model is trained, until the accuracy rate of obtained demand model meets default accuracy rate requirement.By this
Resource allocation method, can on the basis of existing universal model, in conjunction with itself enterprise feature carry out personalized training with
Must reach expected requirement with the matched demand model of itself enterprise, for the enterprise carry out medical resource and deploy to provide to be subsequent
Effective foundation.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 is the structural block diagram of electronic equipment provided by the embodiments of the present application.
Fig. 2 is that resource allocation management method is intervened in the medical inferior health provided by the embodiments of the present application based on artificial intelligence
Flow chart.
Fig. 3 is the flow chart of the sub-step of step S110 in Fig. 2.
Fig. 4 is the flow chart of the sub-step of step S130 in Fig. 2.
Fig. 5 is another flow chart of the sub-step of step S130 in Fig. 2.
Fig. 6 is the flow chart of the sub-step of step S150 in Fig. 2.
Fig. 7 is that resource allocation management method is intervened in the medical inferior health provided by the embodiments of the present application based on artificial intelligence
Another flow chart.
Fig. 8 is that resource allocation management method is intervened in the medical inferior health provided by the embodiments of the present application based on artificial intelligence
Another flow chart.
Fig. 9 is that resource allocation managing device is intervened in the medical inferior health provided by the embodiments of the present application based on artificial intelligence
Functional block diagram.
Icon: 100- electronic equipment；110- intervenes resource allocation managing device based on the medical inferior health of artificial intelligence；
111- universal model obtains module；112- training sample obtains module；113- demand model obtains module；114- accuracy rate obtains
Module；115- modified module；116- training module；120- processor；130- memory.
Specific embodiment
Below in conjunction with attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Usually exist
The component of the embodiment of the present invention described and illustrated in attached drawing can be arranged and be designed with a variety of different configurations herein.
Therefore, the detailed description of the embodiment of the present invention provided in the accompanying drawings is not intended to limit below claimed
The scope of the present invention, but be merely representative of selected embodiment of the invention.Based on the embodiment of the present invention, those skilled in the art
Member's every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.
As shown in Figure 1, the embodiment of the present invention, which is based on the studies above discovery, provides a kind of electronic equipment 100, the electronics
Equipment 100 includes that resource allocation managing device is intervened in memory 130, processor 120 and the medical inferior health based on artificial intelligence
110。
It is directly or indirectly electrically connected between the memory 130 and processor 120, to realize the transmission or friendship of data
Mutually.It is electrically connected for example, these elements can be realized between each other by one or more communication bus or signal wire.It is described to be based on
Resource allocation managing device 110 is intervened in the medical inferior health of artificial intelligence can be with software or firmware including at least one
(firmware) form is stored in the software function module in the memory 130.The processor 120 is described for executing
The executable computer program stored in memory 130, for example, resource is intervened in the medical inferior health based on artificial intelligence
Software function module included by adjustmenting management device 110 and computer program etc., to realize that the medical treatment based on artificial intelligence is sub-
Health intervention resource allocation management method.
Wherein, the memory 130 may be, but not limited to, random access memory (Random Access
Memory, RAM), read-only memory (Read Only Memory, ROM), programmable read only memory (Programmable
Read-Only Memory, PROM), erasable read-only memory (Erasable Programmable Read-Only
Memory, EPROM), electricallyerasable ROM (EEROM) (Electric Erasable Programmable Read-Only
Memory, EEPROM) etc..Wherein, memory 130 is for storing program, the processor 120 after receiving and executing instruction,
Execute described program.
The processor 120 may be a kind of IC chip, the processing capacity with signal.Above-mentioned processor
120 can be general processor, including central processing unit (Central Processing Unit, CPU), network processing unit
(Network Processor, NP) etc.；It can also be digital signal processor (DSP), specific integrated circuit (ASIC), scene
Programmable gate array (FPGA) either other programmable logic device, discrete gate or transistor logic, discrete hardware group
Part.It may be implemented or execute disclosed each method, step and the logic diagram in the embodiment of the present invention.General processor can be with
It is that microprocessor or the processor 120 are also possible to any conventional processor etc..
It is appreciated that structure shown in FIG. 1 is only to illustrate, the electronic equipment 100 may also include more than shown in Fig. 1
Perhaps less component or with the configuration different from shown in Fig. 1.Each component shown in Fig. 1 can use hardware, software
Or combinations thereof realize.
Optionally, the concrete type of the electronic equipment 100 is unrestricted, for example, it may be, but be not limited to, personal electricity
Brain (personal computer, PC), tablet computer, personal digital assistant (personal digital assistant,
PDA), mobile internet surfing equipment (mobile Internet device, MID), web (website) server, data server etc. have
There is the equipment of processing function.
In conjunction with Fig. 2, the embodiment of the present invention also provide it is a kind of can be applied to above-mentioned electronic equipment 100 based on artificial intelligence
Resource allocation management method is intervened in medical inferior health.Wherein, method and step defined in the related process of the method can be by
The processor 120 is realized.Detailed process shown in Fig. 2 will be described in detail below.
Step S110 is obtained and the matched universal model of Target Enterprise from multiple universal models of database.
In the present embodiment, electronic equipment 100 can be communicated to connect with cloud server, can be from the database of cloud server
Data are obtained, or data are stored into the database of cloud server.Storage has multiple logical in the database of cloud server
It with model, is needing to carry out model training to some enterprise, to obtain the characteristic and the enterprise for the employee of the enterprise
When the corresponding relationship of the allotment of medical resource, based on can obtaining wherein matched universal model from multiple universal model
It is trained to obtain.Wherein, the characteristic of employee can be such as the demand characteristics of the working characteristics, employee of employee to medical resource.
Wherein, the medical resource can distribute time etc. for medical teacher's quantity, medical teacher.
Referring to Fig. 3, in the present embodiment, step S110 may include following sub-step:
Step S111, finds from multiple universal models of the database and institute according to the type of business of Target Enterprise
State the matched universal model of the type of business of Target Enterprise.
Step S112, according in the Target Enterprise department constitute situation selected from the universal model found with
The department of the Target Enterprise constitutes the consistent universal model of situation.
Since the Department formation of the enterprise of different types, employee constitute the differences such as situation, for different types of
Enterprise, may be different for the demand of medical resource.Therefore, it is selected in multiple universal models from database matched logical
When with model, it can be selected from multiple universal models and Target Enterprise according to the type of business of Target Enterprise to be modeled
The consistent universal model of the type of business.For example, employee may go out in this enterprise if Target Enterprise is the enterprise of communication scheme class
Difference is more outside, and the daily schedule is not particularly stable, it is thus possible to which the demand to medical resource is not very big.It therefore, can be from
It is found in universal model and the consistent universal model of the type of business of the enterprise of the communication scheme class.And if Target Enterprise is member
Section chief's phase, its daily schedule of the employee of this kind of enterprise was more regular in the enterprise of office indoor office, and since sitting is handled official business,
Employee may be larger to the demand of medical resource.Therefore, for this kind of enterprise, it can select and be somebody's turn to do from multiple universal models
The consistent universal model of the type of business of class enterprise.
On the basis of the above, targeted model if desired is formed to the different departments in some enterprise, it can also be
In the universal model identical with the type of business of enterprise of above-mentioned acquisition, situation is constituted from lookup according to the department in Target Enterprise
To universal model in select and the department of Target Enterprise constitute the consistent universal model of situation.For example, if desired to manpower
Resources Department is individually modeled, if in the universal model selected including the universal model of Human Resource Department, from obtaining
The universal model is selected in the universal model obtained.
In this way, through the above steps, then the type of business phase with Target Enterprise can be obtained from the universal model in cloud
Together, and with the department of Target Enterprise the consistent universal model of situation is constituted.The basis of the subsequent universal model in above-mentioned acquisition
On be trained, can significantly save training burden, improve training effectiveness.
Step S120 obtains the training sample of Target Enterprise, and the training sample carries multiple and different impact factors, respectively
The impact factor carries corresponding weighing factor, and the training sample includes the employee's characteristic and medical treatment of the Target Enterprise
Resource allocation information.
Step S130 pre-processes the training sample, pretreated training sample is directed into described general
Model is trained the universal model, to obtain the corresponding demand model of the Target Enterprise.
In the present embodiment, it can be seen from the above, the training sample carries multiple and different impact factors, each shadow
It rings the factor and carries corresponding weighing factor.Wherein, the impact factor can be week information, daily different time sections letter
Breath, season etc. may generate the impact factor of Different Effects to the demand of employee.For example, such as existing in different week informations
Mon-Fri and in Saturday and weekend, there may be differences for the demand of medical resource by employee.Further, in week
In one specific every day into Friday, demand of the employee to medical resource is also likely to be present difference.And in daily difference
Between in section, demand of the employee to medical resource equally exists difference.And for different seasons, since different seasons can able person
The health degree of body has differences, and therefore, season information can also be included in influence factor.
In the present embodiment, for different impact factors be provided with different weighing factors, if such as time segment information to knot
The influence degree of the accuracy rate of fruit is larger, then can set the larger value for the weighing factor of time segment information.And if week information
It is larger to the influence degree of the accuracy rate of result, then the larger value can be set by week information.It should be noted that respectively influence because
The initial value of the weighing factor of son can be rule of thumb configured, and subsequent needs are according to the accuracy rate of trained result to each shadow
The weighing factor for ringing the factor is continuously adjusted.
Wherein, the value of the weighing factor of each impact factor is decimal greater than 0 less than 1, also, the shadow of all impact factors
The summation for ringing weight is 1.
In the present embodiment, due to having selected multiple impact factors, have one between information contained by each impact factor
Fixed plyability and correlation may will increase operand if directly modeled using the initial data of impact factor,
Expend a large amount of computing resource.Therefore, training sample can be pre-processed, to achieve the purpose that hierarchical layered cools down.
Referring to Fig. 4, in the present embodiment, when being pre-processed to training sample, can be realized by following steps:
Whether step S131, the synthesis output amplitude for detecting the multiple impact factor in the training sample are in pre-
If in range.
Step S132 calculates separately each impact factor relative to described logical if being not in the preset range
With the variance yields of the synthesis output amplitude of the impact factor in model.
Step S133, detects whether the corresponding variance yields of each impact factor is greater than preset threshold, if more than described pre-
If threshold value, then the impact factor of preset quantity is randomly selected from the multiple impact factor, in the impact factor extracted
The feature of each impact factor be normalized with the data volume of each impact factor of simplification.
In the present embodiment, each impact factor includes multiple features, such as above-mentioned time segment information, can be incited somebody to action
It is divided into multiple periods daily, if the daily working time is 8 hours, 8 periods can be divided into, or consider
Employee is either After Hours also likely to be present demand to medical resource before working, therefore can be forward and backward each in the work hours
Increase a period, that is, 10 periods will be divided into daily.
Optionally, multiple impact factors in training sample can be handled first as a whole, can detect training
Whether the synthesis output amplitude of multiple impact factors in sample is in preset range.If being not in preset range, showing can
The data volume of the comprehensive output amplitude of energy is larger, then needs individually to handle multiple impact factors, to carry out training sample
Simplify.
When the synthesis output amplitude of multiple impact factors is not in preset range, then calculate separately each influence because
Variance yields of the son relative to the synthesis output amplitude of the impact factor in the universal model detects each impact factor phase with this
For the extent of deviation of comprehensive output amplitude.
It can detect whether the corresponding variance yields of each impact factor is greater than preset threshold, if more than preset threshold, then table
The synthesis output amplitude that the bright impact factor deviates universal model is larger, can carry out subsequent simplification.
On the basis of the above, it if the variance yields of impact factor is greater than preset threshold, is taken out at random from multiple impact factors
Place is normalized to the feature of each impact factor in the impact factor extracted in the impact factor for taking preset quantity
Reason so achievees the purpose that the data volume for simplifying each impact factor.
In the present embodiment, after pre-processing to training sample, the training sample after needing to pre-process is imported
Extremely to be trained to universal model in the above-mentioned universal model selected.Referring to Fig. 5, the process can pass through following steps reality
It is existing:
Pretreated training sample is directed into the input layer of the neural network by step S134, by described implicit
After the training of layer, exported from the output layer.
Step S135, whether the result for detecting the output layer output reaches expected results, if not up to expected results,
Error signal is obtained according to the result of the output and the expected results, and enters back-propagation phase.
Step S136, using the error signal as the input signal of back-propagation phase from the output layer to described
Input layer reversely returns, and during reversed passback, corrects the neuron between the input layer, output layer and hidden layer
Connection weight weight values are to gradually decrease the error signal of final output.
In the present embodiment, the universal model is constructed based on neural network, which can be BP neural network.It is described
Neural network includes input layer, output layer and hidden layer, wherein the hidden layer can be one or more layers.It is the input layer, defeated
Layer and hidden layer respectively include multiple neurons out, and the neuron between the input layer, output layer and hidden layer has connection
Weighted value.
Optionally, training sample being directed into neural network, training sample enters network from the input layer of neural network,
After one or more layers hidden layer, exported from output layer.In the forward-propagating process of neural network, input layer, hidden layer and
The connection weight weight values of each neuron between output layer and the threshold value of neuron remain unchanged.Each layer of neuron is according to its company
Connect the input and state of one layer of neuron under the influence of weighted value.
In the present embodiment, whether the result that detection is exported from output layer can reach expected results, which is
The pre-set numerical value of user.It, can be according to output result and expection if the output result of output layer is not up to expected results
As a result error signal is obtained, and enters back-propagation phase, to be modified.
In back-propagation phase, obtained error signal will be exported as the input signal in the stage with from neural network
Output layer reversely returned to input layer.Specifically, from output layer by reversely being returned to input layer after each layer hidden layer.
During reversed passback, correct the connection weight weight values of the neuron between the input layer, output layer and hidden layer with by
Decrescence lack the error signal of final output.
In the present embodiment, during reversed passback, using following formula modify the input layer, output layer and
The connection weight weight values of neuron between hidden layer are to gradually decrease the error signal of final output:
Wherein, WijIndicate i-th of neuron of input layer to the connection weight weight values between j-th of hidden layer upgrading, XPIt indicates
The P training sample input layer i-th of input value,Indicate the threshold value of j-th of neuron of hidden layer.
Step S140 obtains the test sample of Target Enterprise, the test sample is directed into the demand model, root
The accuracy rate of the demand model is obtained according to the output result of the demand model.
Step S150, when the accuracy rate is unsatisfactory for default accuracy rate requirement, according to the accuracy rate of the demand model
It weighing factor to the impact factor each in the training sample and modifies to the training sample.
In the present embodiment, training is obtained and demand corresponding to Target Enterprise on the basis of obtaining universal model from cloud
After model, the accuracy rate to obtained demand model is also needed to verify.Optionally, the test specimens of Target Enterprise be can get
This, includes the actual demand amount of the employee of the enterprise to medical resource in varied situations in the test sample.
Test sample can be directed into demand model, model is to the output of test sample as a result, surveying with original according to demand
The actual demand amount of medical resource is compared to obtain the accuracy rate of the demand model in employee in sample sheet.
Whether the accuracy rate of detectable obtained demand model meets preset accuracy rate requirement, if meet it is preset accurate
Rate requirement, shows that the accuracy of the demand model is higher, can be used as the adjustmenting management model of the medical resource of the enterprise.
And if the accuracy rate of the demand model is unsatisfactory for the requirement of preset accuracy rate, can be according to the demand model
Accuracy rate is modified to the weighing factor of impact factor each in training sample and to the training sample.Please refer to figure
6, which can be realized by following steps:
Step S151, by the standard of obtained demand model in the accuracy rate of currently available demand model and history number
True rate is compared.
Step S152, if the accuracy rate of currently available demand model is higher than the need of more than half obtained in history number
The accuracy rate of modulus type then retains current training sample, and modifies to the weighing factor of each impact factor.
Step S153, if the accuracy rate of currently available demand model is lower than the need of more than half obtained in history number
Part training sample in the training sample is then deleted, and adds new training sample by the accuracy rate of modulus type.
In the present embodiment, can by the accuracy rate of the demand model of the accuracy rate of currently available demand model and history into
Row compares, to decide whether that demand is modified training sample.For example, if the accuracy rate of currently available demand model is higher than
The accuracy rate of the demand model of more than half obtained in history number then shows the accuracy rate of the currently available demand model
Although not up to default accuracy rate requirement, still preferable compared to for the accuracy rate of the demand model previously obtained, because
This, can not be replaced training sample, can retain current training sample, only the influence to impact factor each in training sample
Weight is modified.
And if the accuracy rate of currently available demand model is lower than the demand model of more than half obtained in history number
Accuracy rate, then it is unsatisfactory on prediction effect for showing that current demand model compares the demand model that had previously obtained,
Therefore, existing training sample can be modified.Part training sample in the training sample can be deleted, and added new
Training sample to constitute updated training sample, it is subsequent that universal model is trained with updated training sample.
Step S160 is trained the demand model using modified training sample, until obtained demand mould
Until the accuracy rate of type meets default accuracy rate requirement.
Before carrying out continuing training to demand model using updated training sample, also introduce in the present embodiment
The optimizing strategy of genetic algorithm is with to the optimizing again of modified training sample, to reach more preferably training effect.
Fig. 7 is please referred to, in the present embodiment, resource allocation pipe is intervened in the medical inferior health based on artificial intelligence
Reason method is further comprising the steps of:
Step S210 establishes fitness function to assess the fitness of each training sample in modified training sample
Value, selects the highest training sample of fitness value using the selection mechanism of genetic algorithm.
Step S220 randomly chooses any two training sample using the crossover mechanism of genetic algorithm from multiple training samples
This is intersected, to obtain follow-on training sample.
It is highest to calculate fitness value in the follow-on training sample using the fitness function by step S230
Training sample.
Whether step S240, the fitness value of the detection follow-on training sample are lower than the training sample of previous generation
Fitness value introduces mutagenic factor to the follow-on training sample using the Variation mechanism of genetic algorithm if being lower than
Mutation operation is carried out, then calculates the fitness value of the training sample after mutation operation.
Step S250 modifies training sample according to the fitness value of training sample again.
In genetic algorithm, the superiority and inferiority of fitness function evaluation training sample can be established, wherein the higher training of fitness value
A possibility that sample, fitness is better, is retained is higher.And the training sample that fitness value is smaller, then fitness is got over
Difference, it may be necessary to eliminate.
In the present embodiment, genetic algorithm can search out the training sample that fitness is high in training sample by successive ignition,
Also, in the process, in order to avoid falling into local optimum, and in order to increase the diversity of training sample, friendship can be used
Fork mechanism, Variation mechanism etc. improve optimizing accuracy.
Optionally, using the fitness function of foundation to assess the suitable of each training sample in modified training sample
Angle value is answered, and selects the highest training sample of fitness value using selection mechanism.
The crossover mechanism of genetic algorithm is recycled to randomly choose the progress of any two training sample from multiple training samples
Intersect, sample diversity is increased with this.Training sample after intersection is follow-on training sample.
Genetic algorithm can be searched out preferably during successive ignition using the manner of comparison of the result of iteration each time
Training sample.Above-mentioned modified training sample modified again with this.
Optionally, the demand model will be carried out continuing to instruct using the training sample obtained by genetic algorithm optimizing
Practice, until the accuracy rate of obtained demand model meets default accuracy rate requirement.Finally obtained demand model, it is as suitable
Together in the demand model of Target Enterprise, it is formulated to using the demand model according to the characteristic of employee, the demand of employee suitable
Medical resource, to medical resource carry out rationally, effectively utilize.
In the present embodiment, optionally, after the demand model for obtaining being adapted with Target Enterprise, the satisfaction can also be preset
The demand model that accuracy rate requires is sent to cloud, to store in database beyond the clouds, to the data in the database
It is updated.Increase the type of the model in database, and constantly the model in database is optimized.
The above process is in the present embodiment can also be according to actual to the optimization process of demand model in training
Prediction case is modified demand model to further increase the predictablity rate of demand model, preferably meets the reality of user
Border demand.
Optionally, referring to Fig. 8, method provided in this embodiment is further comprising the steps of:
Step S310, receive it is currently entered to measurement information, it is described that multiple and different impact factors is carried to measurement information,
Each impact factor carries corresponding weighing factor.
Step S320 is directed into described the demand model to measurement information and predicts, to obtain prediction result.
Step S330 receives the feedback information of the prediction result to the demand model of user's input.
Step S340 is adjusted according to model parameter of the feedback information to the demand model.
In the present embodiment, when formally carrying out medical resource prediction, it is above-mentioned can will to be used for being directed into measurement information for prediction
The demand model of foundation.Wherein, multiple and different impact factors and the different shadow of each impact factor should be carried to measurement information
Ring weight.Wherein, what information to be measured embodied is the information such as employee's characteristic and weather, week, season.Demand model can be with base
The prediction of the medical resource needed for these information carry out, and export prediction result.
User can feed back prediction result based on the actual demand of prediction result and combination itself, such as fully meet
Self-demand and self-demand are not inconsistent slightly either and self-demand grave fault etc..Also, it can also feed back and self-demand
Specific the case where not being consistent when not being inconsistent, such as can be when itself has resource requirement, but not for the medical treatment of service
Resource, be also possible to medical resource largely leave unused lead to wasting of resources etc..
The feedback information that can be inputted based on user is adjusted the model parameter of demand model, with the demand after adjustment
Model continues subsequent prediction, continues to optimize model with this, constantly promotes the satisfaction of user.
Referring to Fig. 9, the embodiment of the present application also provides a kind of medical inferior health intervention resource allocation based on artificial intelligence
Managing device 110, described device include:
Universal model obtains module 111, matched with Target Enterprise for obtaining from multiple universal models of database
Universal model.It is appreciated that the universal model, which obtains module 111, can be used for executing above-mentioned steps S110, about the Universal Die
The detailed implementation that type obtains module 111 is referred to above-mentioned to the related content of step S110.
Training sample obtains module 112, and for obtaining the training sample of Target Enterprise, the training sample carrying is multiple not
Corresponding weighing factor is arranged for each impact factor in same impact factor.It is appreciated that the training sample obtains module
112 can be used for executing above-mentioned steps S120, and the detailed implementation for obtaining module 112 about the training sample is referred to
It states to the related content of step S120.
Demand model obtains module 113, for pre-processing to the training sample, by pretreated training sample
It is directed into the universal model to be trained the universal model, to obtain the corresponding demand model of the Target Enterprise.It can
To understand, which, which obtains module 113, can be used for executing above-mentioned steps S130, obtain module about the demand model
113 detailed implementation is referred to above-mentioned to the related content of step S130.
Accuracy rate obtains module 114, for obtaining the test sample of Target Enterprise, the test sample is directed into described
In demand model, the accuracy rate of the demand model is obtained according to the output result of the demand model.It is appreciated that this is accurate
Rate, which obtains module 114, can be used for executing above-mentioned steps S140, and the detailed implementation for obtaining module 114 about the accuracy rate can
With referring to above-mentioned to the related content of step S140.
Modified module 115, for when the accuracy rate is unsatisfactory for default accuracy rate and requires, according to the demand model
Accuracy rate to the weighing factor of the impact factor each in the training sample and modifies to the training sample.It can be with
Understand, which can be used for executing above-mentioned steps S150, and the detailed implementation about the modified module 115 can
With referring to above-mentioned to the related content of step S150.
Training module 116, for being trained using modified training sample to the demand model, until obtain
Until the accuracy rate of demand model meets the default accuracy rate requirement.It is appreciated that the training module 116 can be used for executing
Above-mentioned steps S160, the detailed implementation about the training module 116 are referred to above-mentioned to the related content of step S160.
Above-mentioned module can be connected to each other or communicate via wired connection or wireless connection.Wired connection may include metal
Cable, optical cable, mixing cable etc., or any combination thereof.Wireless connection may include by LAN, WAN, bluetooth, ZigBee or
The connection of the forms such as NFC, or any combination thereof.Two or more modules can be combined into individual module, and any one
Module is segmented into two or more units.
It is apparent to those skilled in the art that for convenience and simplicity of description, the device of foregoing description
Specific work process, no longer can excessively be repeated herein with reference to the corresponding process in preceding method.
In conclusion resource allocation manager is intervened in the medical inferior health provided by the embodiments of the present application based on artificial intelligence
Method and device, first by obtained from database with the matched universal model of Target Enterprise, further according to the Target Enterprise of acquisition
Training sample the universal model of acquisition is trained to obtain demand model corresponding with Target Enterprise.And then according to acquisition
The test sample of Target Enterprise obtained demand model is tested, to obtain the accuracy rate of demand model, further according to needing
The accuracy rate of modulus type modifies to the weighing factor and training sample of impact factor each in training sample, and utilization is modified
Training sample continues to be trained demand model, until the default accuracy rate requirement of the accuracy rate satisfaction of obtained demand model is
Only.By the resource allocation method, personalization can be carried out in conjunction with the feature of itself enterprise on the basis of existing universal model
Training expected requirement must be reached and the matched demand model of itself enterprise, for it is subsequent be that the enterprise carries out medical resource
Allotment provides effective foundation.
In embodiment provided herein, it should be understood that disclosed device and method, it can also be by other
Mode realize.The apparatus embodiments described above are merely exemplary, for example, the flow chart and block diagram in attached drawing are shown
Architectural framework in the cards, function and the behaviour of devices in accordance with embodiments of the present invention, method and computer program product
Make.In this regard, each box in flowchart or block diagram can represent a part of a module, section or code, institute
The a part for stating module, section or code includes one or more executable instructions for implementing the specified logical function.
It should also be noted that function marked in the box can also be to be different from attached drawing in some implementations as replacement
The sequence marked occurs.For example, two continuous boxes can actually be basically executed in parallel, they sometimes can also be by
Opposite sequence executes, and this depends on the function involved.It is also noted that each box in block diagram and or flow chart,
And the combination of the box in block diagram and or flow chart, hardware can be based on the defined function of execution or the dedicated of movement
System realize, or can realize using a combination of dedicated hardware and computer instructions.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or equipment for including a series of elements not only includes those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or equipment institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including institute
State in the process, method, article or equipment of element that there is also other identical elements.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.It should also be noted that similar label and letter exist
Similar terms are indicated in following attached drawing, therefore, once being defined in a certain Xiang Yi attached drawing, are then not required in subsequent attached drawing
It is further defined and explained.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (10)
1. resource allocation management method is intervened in a kind of medical inferior health based on artificial intelligence, which is characterized in that the method packet
It includes:
It is obtained from multiple universal models of database and the matched universal model of Target Enterprise；
The training sample of Target Enterprise is obtained, the training sample carries multiple and different impact factors, each impact factor
Corresponding weighing factor is carried, the training sample includes the employee's characteristic and medical resource allotment letter of the Target Enterprise
Breath；
The training sample is pre-processed, pretreated training sample is directed into the universal model to described general
Model is trained, to obtain the corresponding demand model of the Target Enterprise；
The test sample for obtaining Target Enterprise, the test sample is directed into the demand model, according to the demand mould
The output result of type obtains the accuracy rate of the demand model；
When the accuracy rate is unsatisfactory for default accuracy rate requirement, according to the accuracy rate of the demand model to the training sample
In each impact factor weighing factor and modify to the training sample；
The demand model is trained using modified training sample, until the accuracy rate of obtained demand model meets
Until the default accuracy rate requires.
2. resource allocation management method, feature are intervened in the medical inferior health according to claim 1 based on artificial intelligence
The step of being, obtaining universal model matched with Target Enterprise in multiple universal models from database, comprising:
It is found and the Target Enterprise from multiple universal models of the database according to the type of business of Target Enterprise
The matched universal model of the type of business；
Situation is constituted according to the department in the Target Enterprise to select from the universal model found and the Target Enterprise
Department constitute the consistent universal model of situation.
3. resource allocation management method, feature are intervened in the medical inferior health according to claim 1 based on artificial intelligence
It is, each impact factor includes multiple features, described to carry out pretreated step to the training sample, comprising:
Whether the synthesis output amplitude for detecting multiple impact factors in the training sample is in preset range；
If being not in the preset range, each impact factor is calculated separately relative to the influence in the universal model
The variance yields of the synthesis output amplitude of the factor；
Detect whether the corresponding variance yields of each impact factor is greater than preset threshold, if more than the preset threshold, then from institute
State the impact factor that preset quantity is randomly selected in multiple impact factors, on each influence in the impact factor extracted because
The feature of son is normalized with the data volume of each impact factor of simplification.
4. resource allocation management method, feature are intervened in the medical inferior health according to claim 1 based on artificial intelligence
It is, the universal model is based on constructed by neural network, and the neural network includes input layer, output layer and hidden layer, institute
It states input layer, output layer and hidden layer and respectively includes multiple neurons, the nerve between the input layer, output layer and hidden layer
Member has connection weight weight values, described that pretreated training sample is directed into the universal model to universal model progress
Trained step, comprising:
Pretreated training sample is directed into the input layer of the neural network, after the training of the hidden layer, from
The output layer output；
Whether the result for detecting the output layer output reaches expected results, if not up to expected results, according to the output
Result and the expected results obtain error signal, and enter back-propagation phase；
Using the error signal as the input signal of back-propagation phase reversely to be returned from the output layer to the input layer
Pass, during reversed passback, correct the connection weight weight values of the neuron between the input layer, output layer and hidden layer with
Gradually decrease the error signal of final output.
5. resource allocation management method, feature are intervened in the medical inferior health according to claim 4 based on artificial intelligence
It is, it is described during reversed passback, correct the connection weight of the neuron between the input layer, output layer and hidden layer
The step of error signal of the weight values to gradually decrease final output, comprising:
During reversed passback, the neuron between the input layer, output layer and hidden layer is modified using following formula
Connection weight weight values to gradually decrease the error signal of final output:
Wherein, WijIndicate i-th of neuron of input layer to the connection weight weight values between j-th of hidden layer upgrading, XPIndicate P
Training sample input layer i-th of input value,Indicate the threshold value of j-th of neuron of hidden layer.
6. resource allocation management method, feature are intervened in the medical inferior health according to claim 1 based on artificial intelligence
Be, the accuracy rate according to the demand model to the weighing factor of the impact factor each in the training sample and
The step of modifying to the training sample, comprising:
The accuracy rate of currently available demand model is compared with the accuracy rate of obtained demand model in history number；
If the accuracy rate of currently available demand model is higher than the accurate of the demand model of more than half obtained in history number
Rate then retains current training sample, and modifies to the weighing factor of each impact factor；
If the accuracy rate of currently available demand model is accurate lower than the demand model of more than half obtained in history number
Part training sample in the training sample is then deleted, and adds new training sample by rate.
7. resource allocation management method, feature are intervened in the medical inferior health according to claim 1 based on artificial intelligence
Be, the accuracy rate according to the demand model to the weighing factor of the impact factor each in the training sample and
After the step of modifying to the training sample, the method also includes:
Fitness function is established to assess the fitness value of each training sample in modified training sample, is calculated using heredity
The selection mechanism of method selects the highest training sample of fitness value；
Any two training sample is randomly choosed from multiple training samples using the crossover mechanism of genetic algorithm to be intersected, with
Obtain follow-on training sample；
The highest training sample of fitness value in the follow-on training sample is calculated using the fitness function；
Detect the follow-on training sample fitness value whether be lower than previous generation training sample fitness value, if low
In, then using the Variation mechanism introducing mutagenic factor of genetic algorithm to carry out mutation operation to the follow-on training sample,
The fitness value of training sample after calculating mutation operation again；
Training sample is modified again according to the fitness value of training sample.
8. resource allocation management method, feature are intervened in the medical inferior health according to claim 1 based on artificial intelligence
It is, it is described that the demand model is trained using modified training sample, until obtained demand model is accurate
After rate meets the step of until the default accuracy rate requires, the method also includes:
Receive it is currently entered to measurement information, it is described to carry multiple and different impact factors to measurement information, each influence because
Son carries corresponding weighing factor；
The demand model is directed into measurement information predicts, by described to obtain prediction result；
Receive the feedback information of the prediction result to the demand model of user's input；
It is adjusted according to model parameter of the feedback information to the demand model.
9. resource allocation management method, feature are intervened in the medical inferior health according to claim 1 based on artificial intelligence
Be, the accuracy rate until obtained demand model meet preset requirement until the step of after, the method also includes:
The demand model that the default accuracy rate of obtained satisfaction requires is stored in the database, to the number in the database
According to being updated.
10. resource allocation managing device is intervened in a kind of medical inferior health based on artificial intelligence, which is characterized in that described device packet
It includes:
Universal model obtains module, for obtaining from multiple universal models of database and the matched Universal Die of Target Enterprise
Type；
Training sample obtains module, and for obtaining the training sample of Target Enterprise, the training sample carries multiple and different shadows
The factor is rung, corresponding weighing factor is set for each impact factor；
Demand model obtains module and is directed into pretreated training sample for pre-processing to the training sample
The universal model is trained the universal model, to obtain the corresponding demand model of the Target Enterprise；
Accuracy rate obtains module and the test sample is directed into the demand mould for obtaining the test sample of Target Enterprise
In type, the accuracy rate of the demand model is obtained according to the output result of the demand model；
Modified module, for when the accuracy rate is unsatisfactory for default accuracy rate and requires, according to the accuracy rate of the demand model
It weighing factor to the impact factor each in the training sample and modifies to the training sample；
Training module, for being trained using modified training sample to the demand model, until obtained demand mould
Until the accuracy rate of type meets the default accuracy rate requirement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910024268.1A CN109741818A (en) | 2019-01-10 | 2019-01-10 | Resource allocation management method and device are intervened in medical inferior health based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910024268.1A CN109741818A (en) | 2019-01-10 | 2019-01-10 | Resource allocation management method and device are intervened in medical inferior health based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109741818A true CN109741818A (en) | 2019-05-10 |
Family
ID=66364368
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910024268.1A Pending CN109741818A (en) | 2019-01-10 | 2019-01-10 | Resource allocation management method and device are intervened in medical inferior health based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109741818A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458383A (en) * | 2019-06-24 | 2019-11-15 | 平安国际智慧城市科技股份有限公司 | Demand handles implementation method, device and the computer equipment of serviceization, storage medium |
| CN110706225A (en) * | 2019-10-14 | 2020-01-17 | 山东省肿瘤防治研究院（山东省肿瘤医院） | Tumor identification system based on artificial intelligence |
| CN111950574A (en) * | 2019-05-15 | 2020-11-17 | 北京荣森利泰科贸有限公司 | Ritual training information pushing method and device based on wearable equipment |
| CN112700008A (en) * | 2021-01-06 | 2021-04-23 | 青岛弯弓信息技术有限公司 | Model matching processing method and system for cloud configuration platform |
| CN113705849A (en) * | 2020-05-21 | 2021-11-26 | 富士通株式会社 | Information processing apparatus, information processing method, and computer program |
| CN113724845A (en) * | 2021-08-30 | 2021-11-30 | 平安国际智慧城市科技股份有限公司 | Medical staff ratio detection method based on artificial intelligence and related device |
-
2019
- 2019-01-10 CN CN201910024268.1A patent/CN109741818A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111950574A (en) * | 2019-05-15 | 2020-11-17 | 北京荣森利泰科贸有限公司 | Ritual training information pushing method and device based on wearable equipment |
| CN110458383A (en) * | 2019-06-24 | 2019-11-15 | 平安国际智慧城市科技股份有限公司 | Demand handles implementation method, device and the computer equipment of serviceization, storage medium |
| CN110706225A (en) * | 2019-10-14 | 2020-01-17 | 山东省肿瘤防治研究院（山东省肿瘤医院） | Tumor identification system based on artificial intelligence |
| CN113705849A (en) * | 2020-05-21 | 2021-11-26 | 富士通株式会社 | Information processing apparatus, information processing method, and computer program |
| CN113705849B (en) * | 2020-05-21 | 2024-05-17 | 富士通株式会社 | Information processing apparatus and information processing method |
| CN112700008A (en) * | 2021-01-06 | 2021-04-23 | 青岛弯弓信息技术有限公司 | Model matching processing method and system for cloud configuration platform |
| CN112700008B (en) * | 2021-01-06 | 2022-06-28 | 青岛弯弓信息技术有限公司 | Model matching processing method and system for cloud configuration platform |
| CN113724845A (en) * | 2021-08-30 | 2021-11-30 | 平安国际智慧城市科技股份有限公司 | Medical staff ratio detection method based on artificial intelligence and related device |
| CN113724845B (en) * | 2021-08-30 | 2023-06-23 | 深圳平安智慧医健科技有限公司 | Medical personnel proportioning detection method based on artificial intelligence and related device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109741818A (en) | Resource allocation management method and device are intervened in medical inferior health based on artificial intelligence | |
| US10846620B2 (en) | Machine learning-based patent quality metric | |
| US8099311B2 (en) | System and method for routing tasks to a user in a workforce | |
| CN109902708A (en) | A kind of recommended models training method and relevant apparatus | |
| CN110390548A (en) | The selection method and device of coupon distribution strategy | |
| Nearchou | Multi-objective balancing of assembly lines by population heuristics | |
| CN110462636A (en) | The system and method for black box optimization | |
| US20100185572A1 (en) | Methods and systems for multi-participant interactive evolutionary computing | |
| CN107633323A (en) | Core protects method, apparatus, computer equipment and the storage medium of difficulty prediction | |
| CN109766454A (en) | A kind of investor's classification method, device, equipment and medium | |
| CN112669084B (en) | Policy determination method, device and computer readable storage medium | |
| CN107798027A (en) | A kind of heatrate Forecasting Methodology, information recommendation method and device | |
| US20160125297A1 (en) | System and method for solving spatiotemporal-based problems | |
| CN109918574A (en) | Item recommendation method, device, equipment and storage medium | |
| CN113256345A (en) | Self-defining method and device of advertisement putting strategy and computer equipment | |
| AU2022291858A1 (en) | Land management and restoration | |
| US11381635B2 (en) | Method of operating a server apparatus for delivering website content, server apparatus and device in communication with server apparatus | |
| Kayalica et al. | Electricity consumption analysis based on Turkish household budget surveys | |
| Hamza et al. | A constraint consensus memetic algorithm for solving constrained optimization problems | |
| Dodangeh et al. | Best project selection by using of Group TOPSIS method | |
| CN113344415A (en) | Deep neural network-based service distribution method, device, equipment and medium | |
| Akbarian | Ranking All DEA‐Efficient DMUs Based on Cross Efficiency and Analytic Hierarchy Process Methods | |
| CN104850306B (en) | A kind of complete icon aligning method of intelligent terminal and device | |
| CN110427234A (en) | The methods of exhibiting and device of the page | |
| Tavakkoli-Moghaddam et al. | Accessing feasible space in a generalized job shop scheduling problem with the fuzzy processing times: a fuzzy-neural approach |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190510 |
|
| RJ01 | Rejection of invention patent application after publication |