CN110110235A - Method and apparatus for handling data - Google Patents
Method and apparatus for handling data Download PDFInfo
- Publication number
- CN110110235A CN110110235A CN201910401497.0A CN201910401497A CN110110235A CN 110110235 A CN110110235 A CN 110110235A CN 201910401497 A CN201910401497 A CN 201910401497A CN 110110235 A CN110110235 A CN 110110235A
- Authority
- CN
- China
- Prior art keywords
- keyword
- probability
- degree
- incidence
- degree sequence
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Embodiment of the disclosure discloses the method and apparatus for handling data, one specific embodiment of this method includes: to utilize pre-set at least two models based on probability graph, incidence degree sequence is determined respectively, wherein, incidence degree sequence is used to characterize the degree of association of pre-generated keyword and the object in the object set obtained in advance；Identified incidence degree sequence is merged, fusion incidence degree sequence is obtained；Based on fusion incidence degree sequence, the affiliated partner of keyword is determined from object set.The accuracy rate of the affiliated partner of determining keyword can be improved in the method for being used to handle data, the model merely with one based on probability graph is avoided to carry out data processing and unreasonable prediction result occur, further, the affiliated partner that can use carries out information push, to facilitate terminal user to obtain data processed result, user experience is improved.
Description
Technical field
Embodiment of the disclosure is related to field of computer technology, and in particular to the method and apparatus for handling data.
Background technique
Model based on probability graph carries out data processing and is suitable for multiple technologies field, influences the accuracy rate of data processing
Principal element includes: that the algorithm structure of the model based on probability graph and the type of input parameter are based on probability graph in the related technology
Model algorithm structure it is single and input that parameter is unfiled, cause the result accuracy rate of data processing lower.
Summary of the invention
Embodiment of the disclosure proposes the method and apparatus for handling data.
In a first aspect, the embodiment of the present disclosure provides a kind of method for handling data, this method comprises: using setting in advance
Model of at least two set based on probability graph determines incidence degree sequence respectively, wherein the incidence degree sequence is for characterizing in advance
The degree of association of the keyword of generation and the object in the object set obtained in advance；Identified incidence degree sequence is melted
It closes, obtains fusion incidence degree sequence；Based on fusion incidence degree sequence, the affiliated partner of keyword is determined from object set.
In some embodiments, pre-set at least two models based on probability graph are being utilized, is determining association respectively
Before degree series, method further include: to pre-generated key class, first kind keyword and the second class keyword are obtained,
So that determining incidence degree sequence with first kind keyword and the second class keyword based on the model of probability graph to input parameter.
In some embodiments, pre-set at least two models based on probability graph are being utilized, is determining association respectively
Before degree series, method further include: obtain object intersection in object prior probability so that based on the model of probability graph with
The prior probability of object is input parameter, determines incidence degree sequence.
In some embodiments, pre-set at least two models based on probability graph include: concatenated two probability
Graph model or two probability graph models of parallel connection；And pre-set at least two models based on probability graph are utilized, respectively
Determine incidence degree sequence, comprising: first kind keyword and prior probability are inputted into first in concatenated two probability graph models
It is a to be calculated, the output result of first probability graph model and the second class keyword are inputted into concatenated two probability artworks
Second in type is calculated, to obtain the first incidence degree sequence；By one in first kind keyword and the second class keyword,
One is calculated in prior probability input two probability graph models in parallel, will be in first kind keyword and the second class keyword
Another one is calculated in two in parallel probability graph models of another one, prior probability input, to the defeated of two probability graph models
Result is summed out, and normalizes summed result to obtain the second incidence degree sequence.
In some embodiments, identified incidence degree sequence is merged, obtains fusion incidence degree sequence, comprising:
Determine the average value or weighted average that the degree of association of same target is corresponded in the first incidence degree sequence and the second incidence degree sequence
Value obtains fusion incidence degree sequence using identified average value or weighted average as the fusion degree of association.
In some embodiments, identified incidence degree sequence is merged, obtains fusion incidence degree sequence, comprising:
The quantity that the degree of association of any object is indicated in the first incidence degree sequence and the second incidence degree sequence is counted, using quantity as fusion
The degree of association obtains fusion incidence degree sequence.
In some embodiments, based on fusion incidence degree sequence, the affiliated partner of keyword is determined from object set,
The corresponding object of maximum value for comprising determining that the fusion degree of association is the affiliated partner of keyword.
In some embodiments, after the affiliated partner for determining keyword in object set, method further include: base
It is filtered in affiliated partner of the preset rules to the keyword determined.
Second aspect, embodiment of the disclosure provide a kind of for handling the device of data, comprising: the degree of association determines single
Member is configured to determine incidence degree sequence respectively using pre-set at least two models based on probability graph, wherein close
Connection degree series are used to characterize the degree of association of pre-generated keyword and the object in the object set obtained in advance；Fusion is single
Member is configured to merge identified incidence degree sequence, obtains fusion incidence degree sequence；Affiliated partner determination unit,
It is configured to determine the affiliated partner of keyword from object set based on fusion incidence degree sequence.
In some embodiments, device further include: key class unit is configured to pre-generated keyword point
Class obtains first kind keyword and the second class keyword, so that based on the model of probability graph with first kind keyword and second
Class keyword is input parameter, determines incidence degree sequence.
In some embodiments, device further include: prior probability acquiring unit obtains the priori of the object in object intersection
Probability, so that determining incidence degree sequence with the prior probability of object based on the model of probability graph to input parameter.
In some embodiments, pre-set at least two models based on probability graph include: concatenated two probability
Graph model or two probability graph models of parallel connection；And degree of association determination unit is further configured to: by first kind keyword
First inputted in concatenated two probability graph models with prior probability is calculated, by the defeated of first probability graph model
Input in concatenated two probability graph models second of result and the second class keyword is calculated out, to obtain the first association
Degree series；It will be in two probability graph models in parallel of one, prior probability input in first kind keyword and the second class keyword
One is calculated, by two probability in parallel of another one, prior probability input in first kind keyword and the second class keyword
Another one is calculated in graph model, is summed to the output result of two probability graph models, and is normalized to summed result
To obtain the second incidence degree sequence.
In some embodiments, integrated unit is further configured to: calculating the first incidence degree sequence and second degree of association
Corresponded in sequence the average value of the degree of association of same target perhaps weighted average using average value or weighted average as melting
The degree of association is closed, fusion incidence degree sequence is obtained.
In some embodiments, integrated unit is further configured to: statistics first incidence degree sequence and described the
The quantity that the degree of association of any object is indicated in two incidence degree sequences obtains fusion and closes using the quantity as the fusion degree of association
Join degree series.
In some embodiments, affiliated partner determination unit is further configured to: determining the maximum value of the fusion degree of association
Corresponding object is the affiliated partner of keyword.
In some embodiments, device further include: filter element is configured to based on preset rules to the key determined
The affiliated partner of word is filtered.
The third aspect, embodiment of the disclosure provide a kind of server, comprising: one or more processors；Storage device,
It is stored thereon with one or more programs；When one or more programs are executed by one or more processors, so that one or more
A processor realizes the method such as first aspect.
Fourth aspect, embodiment of the disclosure provide a kind of computer-readable medium, are stored thereon with computer program, should
The method such as first aspect is realized when program is executed by processor.
Embodiment of the disclosure provide the method and apparatus for handling data, firstly, using it is pre-set at least
Two models based on probability graph determine incidence degree sequence respectively, can obtain multiple degrees of association based on identical initial data
Sequence prediction obtains fusion degree of association sequence as a result, then, merge to identified incidence degree sequence, finally, based on fusion
Incidence degree sequence determines the affiliated partner of keyword from object set, and the affiliated partner of determining keyword can be improved
Accuracy rate avoids the model merely with one based on probability graph from carrying out data processing and unreasonable prediction result occur, into one
Step ground, the affiliated partner that can use carry out information push, so that terminal user be facilitated to obtain data processed result, improve
User experience.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the disclosure is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is that some embodiments of the present disclosure can be applied to exemplary system architecture figure therein；
Fig. 2 is the flow chart according to one embodiment of the method for handling data of the disclosure；
Fig. 3 is the flow chart according to another embodiment of the method for handling data of the disclosure；
Fig. 4 is the flow chart according to another embodiment of the method for handling data of the disclosure；
Fig. 5 is according to an embodiment of the present disclosure for handling a reality of the model based on probability graph of the method for data
Apply the schematic diagram of example；
Fig. 6 is another of the model based on probability graph of the method according to an embodiment of the present disclosure for being used to handle data
The schematic diagram of embodiment；
Fig. 7 is according to an embodiment of the present disclosure for handling the flow chart of an application scenarios of the method for data；
Fig. 8 is the structural schematic diagram according to one embodiment of the device for handling data of the disclosure；
Fig. 9 is adapted for the structural schematic diagram for realizing the electronic equipment of embodiment of the disclosure.
Specific embodiment
The disclosure is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, part relevant to related invention is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the feature in embodiment and embodiment in the disclosure can phase
Mutually combination.The disclosure is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is shown can the method for handling data using embodiment of the disclosure or the dress for handling data
The exemplary system architecture 100 set.
As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105.
Network 104 between terminal device 101,102,103 and server 105 to provide the medium of communication link.Network 104 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User 110 can be used terminal device 101,102,103 and be interacted with server 105 by network 104, with reception or
Send message etc..Various telecommunication customer end applications can be installed, such as search engine class is answered on terminal device 101,102,103
With, shopping class application, instant messaging tools, mailbox client, social platform software, video playback class apply etc..
Terminal device 101,102,103 can be hardware, be also possible to software.When terminal device 101,102,103 is hard
When part, it can be the various electronic equipments with display screen, including but not limited to smart phone, tablet computer, e-book reading
Device, pocket computer on knee and desktop computer etc..When terminal device 101,102,103 is software, may be mounted at
In above-mentioned cited electronic equipment.Multiple softwares or software module may be implemented into (such as providing Distributed Services in it
Multiple softwares or software module), single software or software module also may be implemented into.It is not specifically limited herein.
Server 105 can be to provide the server of various services.Such as server 105 can be to terminal device 101,
102,103 the background server supported is provided.The data that background server can submit terminal are analyzed, stored or are calculated
Terminal device 101,102,103 is pushed to Deng processing, and by the data prediction result determined using the model based on probability graph.
Under normal conditions, the method provided by embodiment of the disclosure for handling data is generally held by server 105
Row, correspondingly, the device for handling data is generally positioned in server 105.
It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
With continued reference to Fig. 2, the process of one embodiment of the method for handling data according to the disclosure is shown
200.The method for being used to handle data, comprising the following steps:
In step 210, using pre-set at least two models based on probability graph, degree of association sequence is determined respectively
Column, wherein incidence degree sequence is used to characterize the association of pre-generated keyword and the object in the object set obtained in advance
Degree.
It in the present embodiment, include at least two moulds based on probability graph in the algorithm of the method 200 for handling data
Type (executing subject for running the algorithm can be server 105 as shown in Figure 1), each based on the model of probability graph with preparatory
Object in the keyword of generation and the object set obtained in advance is being closed as parameter, the object in computing object intersection is inputted
Conditional probability under the conditions of key word, conditional probability are the degree of association.
As an embodiment of the present embodiment, with no restrictions to the quantity of the object in object intersection, i.e., object can
Think one, or multiple.Correspondingly, also with no restrictions to the quantity of preset keyword, i.e., keyword can be one,
Or it is multiple.Based on above embodiment, conditional probability can be object under multiple keywords simultaneously occurrence condition
Joint probability distribution, also, the conditional probability that multiple objects are calculated forms incidence degree sequence.
Wherein, the Computing Principle of the model based on probability graph is Bayes formula, specifically, obtains keyword right
The prior probability of conditional probability and object as under the conditions of, being then based on Bayesian formula by one is the class derivation of equation, meter
The Joint Distribution for calculating conditional probability or conditional probability of the object under key condition, due to specific formulation process
The not emphasis of the disclosure, this will not be repeated here.
In a step 220, identified incidence degree sequence is merged, obtains fusion incidence degree sequence；
In the present embodiment, fusion for example can be the degree of association that will be calculated by the different models based on probability graph
In sequence, the degree of association for being used to indicate the same object is merged.
For example, it is assumed that there are two incidence degree sequence A={ a1, a2, a3}；B={ b1, b2, b3, wherein a1、b2For to reply
As c1The degree of association, a2、b1For corresponding objects c2The degree of association, a3、b3For corresponding objects c3The degree of association.
Since the degree of association is conditional probability, in some alternative embodiments, fusion can be probability fusion,
That is object c1The fusion degree of association can be (a1+b2)/2, object c2The fusion degree of association can be (a2+b1)/2, object c1's
Merging the degree of association can be (a3+b3)/2, then merging incidence degree sequence is { (a1+b2)/2, (a2+b1)/2, (a3+b3)/2}。
In other optional embodiments, with above-mentioned two incidence degree sequence A={ a1, a2, a3}；B={ b1, b2,
b3For, wherein a1、b1For corresponding objects c1The degree of association, a2For corresponding objects c2The degree of association, b2For corresponding objects c3Pass
Connection degree, a3For corresponding objects c4The degree of association, b3For corresponding objects c5The degree of association.Then fusion can be each object of statistics
Corresponding degree of association quantity, specifically, object c1The fusion degree of association be 2, object c2~c5The fusion degree of association be 1, then melt
Closing incidence degree sequence is { 2,1,1,1,1 }.
In step 230, it is based on the fusion incidence degree sequence, the keyword is determined from the object set
Affiliated partner.
In the present embodiment, the maximum value of the degree of association in fusion incidence degree sequence can be chosen as screening rule, really
Make the affiliated partner of keyword.
With reference to the embodiment in step 220, it is assumed that fusion incidence degree sequence is { (a1+b2)/2, (a2+b1)/2, (a3+
b3)/2 } in, merge the degree of association (a1+b2)/2 are maximum, then the corresponding fusion degree of association (a1+b2The object c of)/21It is determined as closing
The affiliated partner of key word.
Alternatively, the quantity of the degree of association of any object is indicated, by the most object of quantity in statistics fusion incidence degree sequence
It is determined as the affiliated partner of keyword.
With continued reference to the embodiment in step 220, it is assumed that fusion incidence degree sequence is { 2,1,1,1,1 }, then correspondence is melted
Close the object c of the degree of association 21It is confirmed as the affiliated partner of keyword.
In addition, in addition to above-mentioned steps 210 are to step 230, the disclosure is used to locate in some optional implementations
The method of reason data can also include: that the affiliated partner based on the keyword determined carries out information push.
In these optional implementations, executing subject can be pushed the affiliated partner of determining keyword with information
Form feed back to terminal device.Further, executing subject can also be according to the affiliated partner determined to terminal device
Push pushed information relevant to the affiliated partner.In turn, since the method using the present embodiment is capable of determining that and keyword
Associated affiliated partner so that based on affiliated partner generate pushed information specific aim it is stronger, thus be avoided as much as to
The problem of network resources waste that the not high information of the terminal push degree of association may cause.
Below by taking automobile failure diagnosis as an example, the method for handling data of embodiment of the disclosure is illustrated.
The application scenarios of automobile failure diagnosis are described as follows:
Pre-generated keyword is the failure that automobile occurs, for example, keyword k1: stop working in traveling；Keyword k2: again
Secondary starting is not got angry.The object obtained in advance is the reason of causing vehicle failure, for example, object ob1: water temperature sensing device temperature
Spend environmental abnormality；Object ob2: petrol pump is abnormal；Object ob3: the position sensor of air throttle is abnormal；Object ob4: Timing Belt
Fracture.Using the priori knowledge of existing automobile technical field, object set OB={ ob is determined1, ob2, ob3, ob4In it is any
The prior probability of one does not consider the probability of any condition, obtain object prior probability set Pr={ pr1, pr2, pr3, pr4,
Then failure cause is obtained to the induction probability P of failurein={ pin11, pin12, pin13, pin14, pin21, pin22, pin23, pin24,
Wherein, pin11=(k1|ob1), pin12=(k1|ob2), pin13=(k1|ob3), pin14=(k1|ob4)；pin21=(k2|ob1),
pin22=(k2|ob2), pin23=(k2|ob3), pin24=(k2|ob4)。
Executing subject executes step 210, by prior probability Pr、Induce probability PinFew two are separately input into based on probability graph
Model in calculated, get conditional probability p (ob of the object in object intersection under key conditioni|kj), wherein
obi∈ OB, i ∈ { 1,2,3,4 }, j ∈ { 1,2 }.
Further, executing subject is using the model based on probability graph with conditional probability p (obi|kj) it is that parameter is counted
It calculates, obtains conditional probability Joint Distribution p (ob of the object in object set under multiple key conditionsi|k1, k2), that is, it is associated with
Degree, multiple degrees of association constitute incidence degree sequence.
Executing subject successively executes step 220, step 230, the joint probability obtained to the model based on different probability figure
Distribution p (obi|k1, k2) merged, fusion incidence degree sequence is obtained, then, according to fusion incidence degree sequence, is determined from right
Affiliated partner as determining keyword in set, that is, determine the reason of leading to vehicle failure.
The method for handling data that embodiment of the disclosure provides, firstly, utilizing pre-set at least two base
In the model of probability graph, incidence degree sequence is determined respectively, it is pre- can to obtain multiple incidence degree sequences based on identical initial data
Survey result.Then, identified incidence degree sequence is merged, obtains fusion incidence degree sequence.Finally, based on fusion association
Degree series determine the affiliated partner of keyword from object set.The accurate of the affiliated partner of determining keyword can be improved
Rate avoids the model merely with one based on probability graph from carrying out data processing and unreasonable prediction result occur.Further,
The affiliated partner that can use carries out information push, so that terminal user be facilitated to obtain data processed result, improves user
Experience.
With further reference to Fig. 3, it illustrates according to another embodiment of the method for handling data of the disclosure
Process 300.This is used to handle the process 300 of the method for data, comprising the following steps:
In the step 310, to pre-generated key class, first kind keyword and the second class keyword are obtained.
It in the present embodiment, can be by the natural language in executing subject (for example, server 105 as shown in Figure 1)
Unit is managed, user is pre-processed by the text information that terminal device inputs (for example, text resolution, identification, feature extraction
Deng), to obtain pre-generated keyword.It further, can also be by natural language processing unit to the keyword of pre-generatmg
Classify, first kind keyword and the second class keyword is obtained, based on the model of probability graph with first kind keyword, the second class
Keyword is input parameter, determines incidence degree sequence.
In step 320, the prior probability of the object in object intersection is obtained.
In the present embodiment, the prior probability of object indicates probability of the object under the limitation of no other conditions, with object
For certain disease, the prior probability of object can be the disease incidence of the disease.
The prior probability of object also serves as the input parameter of the model based on probability graph, i.e., based on the model of probability graph with
At least one of a kind of keyword, second class keyword and prior probability are that input parameter is calculated, from object set
Determine the affiliated partner of keyword.
The prior probability of object can store on executing subject (for example, server 105 in Fig. 1), also can store
In on other electronic equipments communicated to connect with executing subject, communicated to connect when the prior probability of object is stored in executing subject
Other electronic equipments on when, executing subject can be corresponding to obtain by sending data requesting instructions to the electronic equipment
The prior probability of object.
In a step 330, using pre-set at least two models based on probability graph, degree of association sequence is determined respectively
Column, wherein incidence degree sequence is used to characterize the association of pre-generated keyword and the object in the object set obtained in advance
Degree.
In step 340, identified incidence degree sequence is merged, obtains fusion incidence degree sequence.
In step 350, it is based on the fusion incidence degree sequence, the keyword is determined from the object set
Affiliated partner.
330~step 350 of above-mentioned steps can be according to similar with step 210~step 230 in embodiment illustrated in fig. 2
Mode executes, and details are not described herein.
The method provided by the above embodiment for handling data of the disclosure, is divided by the keyword to pre-generatmg
Class carries out input parameter as the model based on probability graph using the keyword of classification, can be realized the model based on probability graph
Layered method, incidence degree sequence determined based on different types of keyword to realize, further increases determining keyword pass
Join the accuracy rate of object.
With further reference to Fig. 4, it illustrates according to another embodiment of the method for handling data of the disclosure
Process 400.This is used to handle the process 400 of the method for data, comprising the following steps:
In step 410, using pre-set at least two models based on probability graph, degree of association sequence is determined respectively
Column, wherein incidence degree sequence is used to characterize the association of pre-generated keyword and the object in the object set obtained in advance
Degree.
At step 420, identified incidence degree sequence is merged, obtains fusion incidence degree sequence；
In step 430, it is based on the fusion incidence degree sequence, the keyword is determined from the object set
Affiliated partner.
410~step 430 of above-mentioned steps can be according to similar with step 210~step 230 in embodiment illustrated in fig. 2
Mode executes, and details are not described herein.
In step 441, by first kind keyword and prior probability input concatenated two probability graph models (PGM,
Probabilistic graphical model) in first calculated, by the output knot of first probability graph model
Input in concatenated two probability graph models second of fruit and the second class keyword is calculated, to obtain the first degree of association sequence
Column.
In the present embodiment, run in executing subject (for example, server 105 as shown in Figure 1) based on probability graph
Model include concatenated two probability graph models, refering to what is shown in Fig. 5, appended drawing reference 510 is shown in the model based on probability graph
Probability graph model be properly termed as father PGM, the probability graph model shown in appended drawing reference 520 is properly termed as sub- PGM.
When executing subject executes calculation of relationship degree, as an embodiment of the present embodiment, probability graph model 510 with
First kind keyword and prior probability are input parameter, determine the degree of association of the object in the first keyword and object set.
The degree of association of object in the first keyword that probability graph model 520 determines probability graph model 510 and object set and
Second class keyword as input parameter, determine first kind keyword, in the second class keyword and object set object pass
Connection degree, to obtain the first incidence degree sequence.
The step 441 realizes with different levels PGM model, and the output result of subsequent sub- PGM using father PGM are general as priori
Rate improves and is based on this first incidence degree sequence, determines the accuracy rate of the object of keyword association.
In step 442, by one, prior probability input are in parallel in first kind keyword and the second class keyword two
One is calculated in probability graph model, simultaneously by another one, prior probability input in first kind keyword and the second class keyword
Another one is calculated in two probability graph models of connection, is summed to the output result of two probability graph models, and to asking
It normalizes with result to obtain the second incidence degree sequence.
In the present embodiment, run in executing subject (for example, server 105 as shown in Figure 1) based on probability graph
Model include two probability graph models in parallel, refering to what is shown in Fig. 6, as an embodiment of the present embodiment, attached drawing mark
Probability graph model shown in note 610 is input parameter with first kind keyword and prior probability, determine the first keyword with it is right
As the degree of association of the object in set；Probability graph model 620 shown in appended drawing reference 620 is with the second class keyword and prior probability
To input parameter, the degree of association of the object in the second class keyword and object set is determined, then, executing subject is closed first
Key word is asked with the degree of association of the object in object set and the degree of association of the second class keyword and the object in object set
With, and summed result is normalized, to obtain the second incidence degree sequence.
The step 442 also achieves with different levels probability graph model, true based on different classes of keyword by multiple PGM
The degree of association made, summation, normalized to multiple PGM degree of association result determined, can eliminate single PGM by not knowing
Result error caused by factor improves the accuracy rate for determining the object of keyword association.
It should be noted that above-mentioned steps 441 and step 442 are only two kinds of preferred embodiments of the embodiment of the present disclosure,
Any restrictions are not constituted to the technical solution of the disclosure.
In step 451, the degree of association that same target is corresponded in the first incidence degree sequence and the second incidence degree sequence is calculated
Average value perhaps weighted average obtains fusion degree of association sequence using average value or weighted average as the fusion degree of association
Column.
In the present embodiment, it runs in executing subject (for example, server 105 as shown in Figure 1) and is used to handle number
According to method further include that the incidence degree sequence that determines of the models to multiple based on probability graph merges, with eliminate it is different based on
The error of the model of probability graph reduces influence of the uncertain factor to the accuracy rate for the object for determining keyword association.
As the disclosure previous embodiment in, the degree of association is condition of object under conditions of keyword in object set
Probability, therefore, calculate the degree of association that same target is corresponded in the first incidence degree sequence and the second incidence degree sequence average value or
Person's weighted average, in fact, being exactly the average value for calculating the conditional probability for corresponding to the same object in two incidence degree sequences
Or weighted average.
Wherein, when calculating weighted average, the value of weight is depending on the quantity of the model based on probability graph, specifically
Ground, in the present embodiment, the quantity of the model based on probability graph is two, then the value of weight is just between (0,2), accuracy
For one decimal place.
In step 452, the degree of association that any object is indicated in the first incidence degree sequence and the second incidence degree sequence is counted
Quantity, using quantity as fusion the degree of association, obtain fusion incidence degree sequence.
In the present embodiment, using the quantity of the degree of association of statistics as the fusion degree of association, having in statistical significance can
Reliability and feasibility, and cost is relatively low for calculating.
In step 460, the corresponding object of maximum value for determining the fusion degree of association is the affiliated partner of keyword.
In the present embodiment, as an implementation, the maximum value of the average value or weighted average of the degree of association,
That is, the maximum value of the conditional probability of certain corresponding an object, the maximum value indicate the object in object intersection with keyword association
A possibility that it is maximum.
As another embodiment, the corresponding degree of association quantity of certain an object is maximum, shows that this is right from the statistical significance
As with it is maximum a possibility that keyword association.
In step 440, it is filtered based on affiliated partner of the preset rules to the keyword determined.
Using preset rules to the object filter determined, can further filter out based on probability graph model determine
What is occurred in object is unreasonable as a result, to improve the accuracy for determining the affiliated partner of keyword in object set, that is, mentions
The accuracy of high prediction result.
Specification is needed, preset rules are that related fields expert presets according to the knowledge of its fields, should
Preset rules programmable, to run in executing subject (for example, server 105 as shown in Figure 1).
As an optional implementation manner, the method for handling data of the disclosure can be applied to medical assistance and examine
Disconnected, preset rules can be medical expert and be preset according to medical knowledge, exclude class disease for filtering strong symptom, for example,
Male can not suffer from gynecological disease, and cough is unlikely to be acute bronchitis etc. in 20 years.
The method for handling data that embodiment of the disclosure provides, the model based on probability graph can realize stratified probability
Graph model calculates, and the incidence degree sequence that different probability graph model is determined is merged, and can eliminate individual probability figure
Model existing for error, improve determine keyword affiliated partner accuracy rate, additionally include the association pair to determining
It is unreasonable as a result, to further increase the accuracy rate of prediction in object to filter out as carrying out the filtering based on preset rules.
With continued reference to the process that Fig. 7, Fig. 7 are according to the application scenarios of the method for handling data of the present embodiment
700.The process 700 of the application scenarios, comprising the following steps:
Step 701 receives case history text.
Executing subject (for example, server 105 as shown in Figure 1) receives the case history text that terminal is sent.User (for example,
Doctor) by terminal (such as terminal shown in FIG. 1 101,102,103) input case history text, which can be electronic health record.
Step 702, natural language processing.
Executing subject includes natural language processing unit, integrated intelligent algorithm, executing subject in natural language processing unit
The case history text received is parsed using the natural language processing unit, and extracts illness based on analysis result.
Step 703~step 704.
The natural language processing unit of executing subject executes step 703 and step 704, classifies to illness, obtains positive disease
Disease and negative sense illness, that is, realize that positive illness extracts the extraction with negative sense illness.Wherein, positive illness has already appeared for patient
Illness, negative sense illness is the illness that does not occur of patient.
For example, case history input is as follows: intermittent more than ten years of right hypochondrial region, recurrence aggravate two months；With Nausea and vomiting；
Without fever, shiver with cold, no spitting blood, melena.Then wherein positive illness includes: right hypochondrial region, Nausea and vomiting；Negative sense illness includes: hair
Heat, shiver with cold, spitting blood, melena.
Step 705 and step 706.
The natural language processing unit of executing subject is based on positive illness and negative sense illness is classified, acquisition symptom (the
A kind of keyword) and sign (the second class keyword), wherein symptom includes positive symptom and negative sense symptom, and positive symptom is to suffer from
The symptom that person occurs, negative sense symptom are the symptom that patient does not occur；Sign includes positive sign and negative sense sign, positive sign are
The sign that patient occurs, negative sense sign are the sign that patient does not occur.
Step 707 obtains disease incidence.
In the present embodiment, executing subject is based on symptom (first kind keyword) and sign (the second class keyword), to number
Request of data is sent according to storage unit (for example, server 106 as shown in Figure 1), symptom and body occurs to obtain can result in
The disease incidence (prior probability) of the disease of sign；Meanwhile executing subject also obtains positive symptom under disease occurrence condition respectively
First induces probability, and first inhibition probability of the negative sense symptom under disease occurrence condition, positive sign is under disease occurrence condition
Second induce probability, negative sense sign under disease occurrence condition second inhibit probability.
Step 708~step 710.
In the present embodiment, operation executes Unified Policy, simultaneously there are three the model based on probability graph respectively in executing subject
Connection strategy and series connection strategy.
Wherein, Unified Policy includes a PGM, and Unified Policy is to the parameter of input without classificating requirement, that is, directly by first
It induces probability, the first inhibition probability, the second induction probability, the second inhibition probability and prior probability and is input to execution Unified Policy
PGM in calculated, to obtain the incidence degree sequence of corresponding Unified Policy；
Strategy in parallel includes and associated two PGM, strategy in parallel have classificating requirement to input parameter, that is, lures first
Hair probability, the first inhibition probability and be input in two PGM one of prior probability are calculated, by second induce probability,
Second inhibition probability and prior probability be input in two PGM another calculated, the pass that two PGM are calculated
Join degree series summation, and summed result is normalized, to obtain the incidence degree sequence of corresponding strategy in parallel.
Series connection strategy includes concatenated two PGM, and series connection strategy has classificating requirement to input parameter, that is, induces first
Probability, the first inhibition probability and prior probability are input to father PGM and calculate, the priori by the output of father PGM as sub- PGM
Probability input, while the second induction probability, the second inhibition probability being input to sub- PGM and calculated, it is tactful to obtain corresponding series connection
Incidence degree sequence.
Step 711 executes fusion.
Executing subject incidence degree sequence and correspondence that the incidence degree sequence of above-mentioned corresponding Unified Policy, corresponding parallel connection is tactful
The incidence degree sequence of series connection strategy is merged, and fusion incidence degree sequence is obtained, wherein fusion incidence degree sequence for example can be
Joint probability point of the every kind of disease under a variety of positive symptoms, negative sense symptom, positive sign, negative sense sign simultaneously occurrence condition
The method and step of cloth, fusion is identical with previous embodiment, and details are not described herein.
Based on fusion probability sequence, determine the corresponding disease of maximum value in fusion probability sequence as prediction result.
Step 712~step 714.
Further, executing subject is filtered prediction result based on preset rules, is pushed away with the disease for eliminating unreasonable
Reason, feeds back to terminal for filtered prediction result, that is, informs diagnostic result of the user based on current case history.
Preset rules can be preset for medical expert according to medical knowledge, be stored in data storage cell, execute
Main body can obtain preset rules by sending request of data.
As an implementation, preset rules can be set based on gender, symptom duration parameter.
It should be appreciated that it is above-mentioned it is shown in fig. 7 be only exemplary application scene for handling the method for data, not generation
Restriction of the table to the disclosure.
With further reference to Fig. 8, as the realization to method shown in above-mentioned each figure, present disclose provides one kind for handling number
According to method one embodiment, the Installation practice is corresponding with embodiment of the method shown in Fig. 2, which can specifically answer
For in various electronic equipments.
As shown in figure 8, the device 800 for being used to handle data may include: degree of association determination unit 810, it is configured to
Using pre-set at least two models based on probability graph, incidence degree sequence is determined respectively, wherein incidence degree sequence is used for
The degree of association of the pre-generated keyword of characterization and the object in the object set obtained in advance；Integrated unit 820, is configured to
Identified incidence degree sequence is merged, fusion incidence degree sequence is obtained；Affiliated partner determination unit 830, is configured to
Based on fusion incidence degree sequence, the affiliated partner of keyword is determined from object set.
In some optional embodiments of the present embodiment, for handling the device 800 of data further include: key class
Unit 840 is configured to obtain first kind keyword and the second class keyword to pre-generated key class, so that
Incidence degree sequence is determined to input parameter with first kind keyword and the second class keyword based on the model of probability graph.
In some optional embodiments of the present embodiment, for handling the device 800 of data further include: prior probability obtains
Unit 850 is taken, is configured to obtain the prior probability of the object in object intersection, so that based on the model of probability graph with object
Prior probability be input parameter, determine incidence degree sequence.
In some optional embodiments of the present embodiment, the pre-set at least two model packets based on probability graph
Include: concatenated two probability graph models or two probability graph models of parallel connection, degree of association determination unit 810 are further configured
At: first that first kind keyword and prior probability are inputted in concatenated two probability graph models calculates, by this
The output result of one probability graph model and the second class keyword input second progress in concatenated two probability graph models
It calculates, to obtain the first incidence degree sequence；One, prior probability input in first kind keyword and the second class keyword is in parallel
Two probability graph models in one calculated, by another one, prior probability in first kind keyword and the second class keyword
It inputs another one in two probability graph models in parallel to be calculated, sum to the output result of two probability graph models,
And summed result is normalized to obtain the second incidence degree sequence.
In some optional embodiments of the present embodiment, integrated unit 820 is further configured to: calculating the first association
The average value or weighted average that the degree of association of same target is corresponded in degree series and the second incidence degree sequence, by average value or
Person's weighted average obtains fusion incidence degree sequence as the fusion degree of association.
In some optional embodiments of the present embodiment, integrated unit 820 is further configured to: the first association of statistics
The quantity that the degree of association of any object is indicated in degree series and the second incidence degree sequence is obtained using quantity as the fusion degree of association
Merge incidence degree sequence.
In some optional embodiments of the present embodiment, affiliated partner determination unit 830 is further configured to: being determined
The corresponding object of maximum value for merging the degree of association is the affiliated partner of keyword.
In some optional embodiments of the present embodiment, for handling the device 800 of data further include:
Filter element 860 is configured to be filtered based on affiliated partner of the preset rules to the keyword determined.
Below with reference to Fig. 9, it illustrates the electronic equipment that is suitable for being used to realize embodiment of the disclosure, (example is as shown in figure 1
Server) 900 structural schematic diagram.Server shown in Fig. 9 is only an example, should not be to the function of embodiment of the disclosure
Any restrictions can be brought with use scope.
As shown in figure 9, electronic equipment 900 may include processing unit (such as central processing unit, graphics processor etc.)
901, random access can be loaded into according to the program being stored in read-only memory (ROM) 902 or from storage device 908
Program in memory (RAM) 903 and execute various movements appropriate and processing.In RAM 903, it is also stored with electronic equipment
Various programs and data needed for 900 operations.Processing unit 901, ROM 902 and RAM 903 pass through the phase each other of bus 904
Even.Input/output (I/O) interface 905 is also connected to bus 904.
In general, following device can connect to I/O interface 905: including such as touch screen, touch tablet, keyboard, mouse, taking the photograph
As the input unit 906 of head, microphone, accelerometer, gyroscope etc.；Including such as liquid crystal display (LCD), loudspeaker, vibration
The output device 907 of dynamic device etc.；And communication device 909.Communication device 909 can permit electronic equipment 900 and other equipment into
Row is wirelessly or non-wirelessly communicated to exchange data.Although Fig. 9 shows the electronic equipment 900 with various devices, it should be understood that
Be, it is not required that implement or have all devices shown.It can alternatively implement or have more or fewer devices.Figure
Each box shown in 9 can represent a device, also can according to need and represent multiple devices.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communication device 909, or from storage device 908
It is mounted, or is mounted from ROM 902.When the computer program is executed by processing unit 901, the implementation of the disclosure is executed
The above-mentioned function of being limited in the method for example.It should be noted that computer-readable medium described in embodiment of the disclosure can be with
It is computer-readable signal media or computer readable storage medium either the two any combination.It is computer-readable
Storage medium for example may be-but not limited to-the system of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, device or
Device, or any above combination.The more specific example of computer readable storage medium can include but is not limited to: have
The electrical connection of one or more conducting wires, portable computer diskette, hard disk, random access storage device (RAM), read-only memory
(ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-
ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.In embodiment of the disclosure, computer
Readable storage medium storing program for executing can be any tangible medium for including or store program, which can be commanded execution system, device
Either device use or in connection.And in embodiment of the disclosure, computer-readable signal media may include
In a base band or as the data-signal that carrier wave a part is propagated, wherein carrying computer-readable program code.It is this
The data-signal of propagation can take various forms, including but not limited to electromagnetic signal, optical signal or above-mentioned any appropriate
Combination.Computer-readable signal media can also be any computer-readable medium other than computer readable storage medium, should
Computer-readable signal media can send, propagate or transmit for by instruction execution system, device or device use or
Person's program in connection.The program code for including on computer-readable medium can transmit with any suitable medium,
Including but not limited to: electric wire, optical cable, RF (radio frequency) etc. or above-mentioned any appropriate combination.
Above-mentioned computer-readable medium can be above-mentioned server；It is also possible to individualism, and it is unassembled
Enter in the server.Above-mentioned computer-readable medium carries one or more program, when said one or multiple programs
When being executed by the server, so that the server: utilizing pre-set at least two models based on probability graph, determine respectively
Incidence degree sequence, wherein the incidence degree sequence is for characterizing in pre-generated keyword and the object set obtained in advance
The degree of association of object；Identified incidence degree sequence is merged, fusion incidence degree sequence is obtained；Based on fusion degree of association sequence
Column, determine the affiliated partner of keyword from object set.
The behaviour for executing embodiment of the disclosure can be write with one or more programming languages or combinations thereof
The computer program code of work, described program design language include object oriented program language-such as Java,
Smalltalk, C++ further include conventional procedural programming language-such as " C " language or similar program design language
Speech.Program code can be executed fully on the user computer, partly be executed on the user computer, as an independence
Software package execute, part on the user computer part execute on the remote computer or completely in remote computer or
It is executed on server.In situations involving remote computers, remote computer can pass through the network of any kind --- packet
Include local area network (LAN) or wide area network (WAN) --- it is connected to subscriber computer, or, it may be connected to outer computer (such as
It is connected using ISP by internet).
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the disclosure, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in embodiment of the disclosure can be realized by way of software, can also be passed through
The mode of hardware is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor
Including degree of association determination unit, integrated unit and affiliated partner determination unit.The title of these units is under certain conditions not
The restriction to the unit itself is constituted, for example, degree of association determination unit is also described as " utilizing pre-set at least two
A model based on probability graph determines the unit of incidence degree sequence respectively ".
Above description is only the preferred embodiment of the disclosure and the explanation to institute's application technology principle.Those skilled in the art
Member it should be appreciated that embodiment of the disclosure involved in invention scope, however it is not limited to the specific combination of above-mentioned technical characteristic and
At technical solution, while should also cover do not depart from foregoing invention design in the case where, by above-mentioned technical characteristic or its be equal
Feature carries out any combination and other technical solutions for being formed.Such as disclosed in features described above and embodiment of the disclosure (but
It is not limited to) technical characteristic with similar functions is replaced mutually and the technical solution that is formed.
Claims (18)
1. a kind of method for handling data, comprising:
Using pre-set at least two models based on probability graph, incidence degree sequence is determined respectively, wherein the degree of association
Sequence is used to characterize the degree of association of pre-generated keyword and the object in the object set obtained in advance；
Identified incidence degree sequence is merged, fusion incidence degree sequence is obtained；
Based on the fusion incidence degree sequence, the affiliated partner of the keyword is determined from the object set.
2. according to the method described in claim 1, wherein, utilizing pre-set at least two moulds based on probability graph described
Type, before determining incidence degree sequence respectively, the method also includes:
To pre-generated key class, first kind keyword and the second class keyword are obtained, so that based on probability graph
Model is input parameter with the first kind keyword and the second class keyword, determines the incidence degree sequence.
3. according to the method described in claim 2, wherein, utilizing pre-set at least two moulds based on probability graph described
Type, before determining incidence degree sequence respectively, the method also includes:
The prior probability of the object in the object intersection is obtained, so that based on the model of probability graph with the priori of the object
Probability is input parameter, determines the incidence degree sequence.
4. according to the method described in claim 3, wherein, the described pre-set at least two model packets based on probability graph
It includes: concatenated two probability graph models or two probability graph models of parallel connection；And
It is described using pre-set at least two models based on probability graph, respectively determine incidence degree sequence, comprising:
First that the first kind keyword and the prior probability are inputted in concatenated two probability graph models counts
It calculates, the output result of first probability graph model and the second class keyword is inputted in concatenated two probability graph models
Second calculated, to obtain the first incidence degree sequence；
By the two probability artworks in parallel of one, prior probability input in the first kind keyword and the second class keyword
One is calculated in type, simultaneously by another one, prior probability input in the first kind keyword and the second class keyword
Another one is calculated in two probability graph models of connection, is summed to the output result of two probability graph models, and to institute
Summed result normalization is stated to obtain the second incidence degree sequence.
5. it is described that identified incidence degree sequence is merged according to the method described in claim 4, wherein, it is merged
Incidence degree sequence, comprising:
Determine the average value that the degree of association of same target is corresponded in first incidence degree sequence and second incidence degree sequence
Perhaps weighted average obtains fusion degree of association sequence using identified average value or weighted average as the fusion degree of association
Column.
6. it is described that identified incidence degree sequence is merged according to the method described in claim 4, wherein, it is merged
Incidence degree sequence, comprising:
Count the number that the degree of association of any object is indicated in first incidence degree sequence and second incidence degree sequence
Amount obtains fusion incidence degree sequence using the quantity as the fusion degree of association.
7. method according to claim 5 or 6, wherein it is described to be based on the fusion incidence degree sequence, from the object set
The affiliated partner of the keyword is determined in conjunction, comprising:
The corresponding object of maximum value for determining the fusion degree of association is the affiliated partner of the keyword.
8. method described in any one of -6 according to claim 1, wherein determine institute from the object set described
After the affiliated partner for stating keyword, the method also includes:
It is filtered based on affiliated partner of the preset rules to the keyword determined.
9. a kind of for handling the device of data, comprising:
Degree of association determination unit is configured to determine close respectively using pre-set at least two models based on probability graph
Join degree series, wherein the incidence degree sequence is for characterizing in pre-generated keyword and the object set obtained in advance
The degree of association of object；
Integrated unit is configured to merge identified incidence degree sequence, obtains fusion incidence degree sequence；
Affiliated partner determination unit is configured to determine institute from the object set based on the fusion incidence degree sequence
State the affiliated partner of keyword.
10. device according to claim 9, wherein described device further include:
Key class unit is configured to obtain first kind keyword to pre-generated key class and the second class is closed
Key word, so that based on the model of probability graph with the first kind keyword and the second class keyword to input parameter, really
The fixed incidence degree sequence.
11. device according to claim 10, wherein described device further include:
Prior probability acquiring unit is configured to obtain the prior probability of the object in the object intersection, so that based on general
The model of rate figure is input parameter with the prior probability of the object, determines the incidence degree sequence.
12. device according to claim 11, wherein the described pre-set at least two model packets based on probability graph
It includes: concatenated two probability graph models or two probability graph models of parallel connection；And
The degree of association determination unit is further configured to:
First that the first kind keyword and the prior probability are inputted in concatenated two probability graph models counts
It calculates, the output result of first probability graph model and the second class keyword is inputted in concatenated two probability graph models
Second calculated, to obtain the first incidence degree sequence；
By the two probability artworks in parallel of one, prior probability input in the first kind keyword and the second class keyword
One is calculated in type, simultaneously by another one, prior probability input in the first kind keyword and the second class keyword
Another one is calculated in two probability graph models of connection, is summed to the output result of two probability graph models, and to institute
Summed result normalization is stated to obtain the second incidence degree sequence.
13. device according to claim 12, wherein the integrated unit is further configured to:
Calculate the average value that the degree of association of same target is corresponded in first incidence degree sequence and second incidence degree sequence
Perhaps weighted average obtains fusion incidence degree sequence using the average value or weighted average as the fusion degree of association.
14. device according to claim 12, wherein the integrated unit is further configured to:
Count the number that the degree of association of any object is indicated in first incidence degree sequence and second incidence degree sequence
Amount obtains fusion incidence degree sequence using the quantity as the fusion degree of association.
15. device described in 3 or 14 according to claim 1, wherein the affiliated partner determination unit is further configured to:
The corresponding object of maximum value for determining the fusion degree of association is the affiliated partner of the keyword.
16. the device according to any one of claim 9-14, wherein described device further include:
Filter element is configured to be filtered based on affiliated partner of the preset rules to the keyword determined.
17. a kind of server, comprising:
One or more processors；
Storage device is stored thereon with one or more programs；
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
Now such as method described in any one of claims 1-8.
18. a kind of computer-readable medium, is stored thereon with computer program, wherein real when described program is executed by processor
Now such as method described in any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910401497.0A CN110110235B (en) | 2019-05-14 | 2019-05-14 | Method and device for pushing information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910401497.0A CN110110235B (en) | 2019-05-14 | 2019-05-14 | Method and device for pushing information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110110235A true CN110110235A (en) | 2019-08-09 |
| CN110110235B CN110110235B (en) | 2021-08-31 |
Family
ID=67490119
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910401497.0A Active CN110110235B (en) | 2019-05-14 | 2019-05-14 | Method and device for pushing information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110110235B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120207359A1 (en) * | 2011-02-11 | 2012-08-16 | Microsoft Corporation | Image Registration |
| US20130315470A1 (en) * | 2012-05-25 | 2013-11-28 | Poikos Limited | Body measurement |
| CN105956130A (en) * | 2016-05-09 | 2016-09-21 | 浙江农林大学 | Multi-information fusion scientific research literature theme discovering and tracking method and system thereof |
| CN106570046A (en) * | 2016-03-02 | 2017-04-19 | 合网络技术（北京）有限公司 | Method and device for recommending relevant search data based on user operation behavior |
| CN106777304A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | The method for pushing and device of theme |
| CN206741482U (en) * | 2017-03-27 | 2017-12-12 | 乐山师范学院 | A kind of network public-opinion supervising device using probability graph model |
| WO2018009643A1 (en) * | 2016-07-07 | 2018-01-11 | Aspen Technology, Inc. | Computer systems and methods for performing root cause analysis and building a predictive model for rare event occurrences in plant-wide operations |
| US20180268559A1 (en) * | 2017-03-16 | 2018-09-20 | Electronics And Telecommunications Research Institute | Method for tracking object in video in real time in consideration of both color and shape and apparatus therefor |
| CN109656541A (en) * | 2018-11-20 | 2019-04-19 | 东软集团股份有限公司 | Exploitative recommended method, device, storage medium and electronic equipment |
-
2019
- 2019-05-14 CN CN201910401497.0A patent/CN110110235B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120207359A1 (en) * | 2011-02-11 | 2012-08-16 | Microsoft Corporation | Image Registration |
| US20130315470A1 (en) * | 2012-05-25 | 2013-11-28 | Poikos Limited | Body measurement |
| CN106570046A (en) * | 2016-03-02 | 2017-04-19 | 合网络技术（北京）有限公司 | Method and device for recommending relevant search data based on user operation behavior |
| CN105956130A (en) * | 2016-05-09 | 2016-09-21 | 浙江农林大学 | Multi-information fusion scientific research literature theme discovering and tracking method and system thereof |
| WO2018009643A1 (en) * | 2016-07-07 | 2018-01-11 | Aspen Technology, Inc. | Computer systems and methods for performing root cause analysis and building a predictive model for rare event occurrences in plant-wide operations |
| CN106777304A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | The method for pushing and device of theme |
| US20180268559A1 (en) * | 2017-03-16 | 2018-09-20 | Electronics And Telecommunications Research Institute | Method for tracking object in video in real time in consideration of both color and shape and apparatus therefor |
| CN206741482U (en) * | 2017-03-27 | 2017-12-12 | 乐山师范学院 | A kind of network public-opinion supervising device using probability graph model |
| CN109656541A (en) * | 2018-11-20 | 2019-04-19 | 东软集团股份有限公司 | Exploitative recommended method, device, storage medium and electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 吴蕾等: "《异构信息网络数据上的融合概率图模型》", 《HTTP://WWW.CNKI.NET/KCMS/DOI/10.3778/J.ISSN.1673-9418.1401005.HTML》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110110235B (en) | 2021-08-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20190147430A1 (en) | Customizing payment sessions with machine learning models | |
| US20200167798A1 (en) | Customizing customer onboarding for a service with machine learning models | |
| US11412070B2 (en) | Method and apparatus for generating information | |
| Gadekallu et al. | Federated learning for big data: A survey on opportunities, applications, and future directions | |
| WO2019080662A1 (en) | Information recommendation method, device and apparatus | |
| US11929962B2 (en) | Method and system for monitoring and integration of one or more intelligent conversational agents | |
| JP2020506473A (en) | Method for adjusting risk parameters and method and device for risk identification | |
| US11334750B2 (en) | Using attributes for predicting imagery performance | |
| US10998103B2 (en) | Medical risk factors evaluation | |
| CN110837565A (en) | Model training method and device for realizing medicine recommendation and calculating equipment | |
| CN110224870A (en) | Interface monitoring method, apparatus calculates equipment and storage medium | |
| US20220350825A1 (en) | Automated response engine to implement internal communication interaction data via a secured omnichannel electronic data channel and external communication interaction data | |
| Preuveneers et al. | SAMURAI: A batch and streaming context architecture for large-scale intelligent applications and environments | |
| US20220358240A1 (en) | Adaptive data privacy platform | |
| WO2022152018A1 (en) | Method and device for identifying multiple accounts belonging to the same person | |
| CN110059172A (en) | The method and apparatus of recommendation answer based on natural language understanding | |
| US20200402153A1 (en) | Negotiation device | |
| CN109767337A (en) | The recognition methods of adverse selection user, device and computer equipment in insurance | |
| US20220311766A1 (en) | Sensor-based authentication, notification, and assistance systems | |
| CN110110235A (en) | Method and apparatus for handling data | |
| US20230418794A1 (en) | Data processing method, and non-transitory medium and electronic device | |
| US20220358366A1 (en) | Generation and implementation of dedicated feature-based techniques to optimize inference performance in neural networks | |
| CN111709784B (en) | Method, apparatus, device and medium for generating user retention time | |
| US20230237180A1 (en) | Systems and methods for linking a screen capture to a user support session | |
| CN115689202B (en) | Three-dimensional digital data processing method based on block chain |
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 |