CN109726847A - Position predicting method, device, electronic equipment and readable storage medium storing program for executing - Google Patents
Position predicting method, device, electronic equipment and readable storage medium storing program for executing Download PDFInfo
- Publication number
- CN109726847A CN109726847A CN201811379242.0A CN201811379242A CN109726847A CN 109726847 A CN109726847 A CN 109726847A CN 201811379242 A CN201811379242 A CN 201811379242A CN 109726847 A CN109726847 A CN 109726847A
- Authority
- CN
- China
- Prior art keywords
- sample
- target
- position score
- score
- target object
- 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.)
- Withdrawn
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000012549 training Methods 0.000 claims abstract description 29
- 238000010801 machine learning Methods 0.000 claims abstract description 13
- 230000002123 temporal effect Effects 0.000 claims description 63
- 238000004590 computer program Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 description 7
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 241000406668 Loxodonta cyclotis Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Embodiment of the disclosure provides a kind of position predicting method, device, electronic equipment and readable storage medium storing program for executing, which comprises obtains object time and target object；Object time feature vector is generated according to the object time；Object time feature vector is input to the position score prediction model for target object and obtains target position score；Position score prediction model is the machine learning model using set of data samples training；Sample data set includes the sample position score whether sample time feature vector and instruction target object are located at predeterminated position；Determine target object whether in predeterminated position according to target position score.It can be according to object time and model automatic Prediction position score, automatically determine whether target object is located at predeterminated position, in scene of the method that this programme is provided using businessman's visit, whether the target (responsible person of businessman) that can be very good prediction visit is interior at shop (predeterminated position) in the object time, helps to improve visit success rate.
Description
Technical field
Embodiment of the disclosure is related to machine prediction technical field more particularly to a kind of position predicting method, device, electronics
Equipment and readable storage medium storing program for executing.
Background technique
The commodity of businessman tend to rely on sales force in sales process, and sales force can frequent target businessman.So
And sales force often will appear situation of the businessman responsible person not in shop in visit, and sales force is caused not obtain in time
Commodity or other merchandise newss are taken, the final sale for influencing commodity.It is hereby understood that whether confirmation businessman responsible person is in businessman in advance
The sales volume of commodity is helped to improve in shops.
In the prior art, whether sales force needs empirically determined businessman responsible person in businessman shops.For example,
During visit before, businessman responsible person often eat out between 12 points to 14 points or businessman responsible person at 14 points extremely
There is meeting between 15 points, can not receive.It is visitd in addition, sales force can also be reserved by phone, short message or other communication modes
It visits.
For above scheme, whether accuracy in businessman shops is lower by empirically determined businessman responsible person, especially
It is to accumulate experience less, accuracy is lower for new sales force；Reservation visit can have the case where reservation failure, cause
It is lower to visit success rate.
Summary of the invention
Embodiment of the disclosure provides a kind of position predicting method, device, electronic equipment and readable storage medium storing program for executing.
It is according to an embodiment of the present disclosure in a first aspect, providing a kind of position predicting method, which comprises
Obtain object time and target object；
Object time feature vector is generated according to the object time；
The object time feature vector is input to the position score trained in advance for the target object and predicts mould
In type, target position score is obtained；The position score prediction model is the machine learning model using set of data samples training；
The sample data set includes sample time feature vector and corresponding sample position score；The sample position score is used for
Indicate whether the target object is located at predeterminated position；
Determine the target object whether in the predeterminated position according to the target position score.
Second aspect according to an embodiment of the present disclosure, provides a kind of predicted position device, and described device includes:
Data obtaining module, for obtaining object time and target object；
Temporal characteristics vector generation module, for generating object time feature vector according to the object time；
Position score prediction module, it is preparatory for the target object for being input to the object time feature vector
In trained position score prediction model, target position score is obtained；The position score prediction model is to use data sample
Collect the machine learning model of training；The sample data set includes that sample time feature vector and corresponding sample position obtain
Point；The sample position score is used to indicate whether the target object is located at predeterminated position；
Position determination module, for determining the target object whether in the default position according to the target position score
It sets.
The third aspect according to an embodiment of the present disclosure, provides a kind of electronic equipment, comprising:
Processor, memory and it is stored in the computer journey that can be run on the memory and on the processor
Sequence, which is characterized in that the processor realizes aforementioned location prediction technique when executing described program.
Fourth aspect according to an embodiment of the present disclosure provides a kind of readable storage medium storing program for executing, when in the storage medium
Instruction by electronic equipment processor execute when so that electronic equipment is able to carry out aforementioned location prediction technique.
Embodiment of the disclosure provides a kind of position predicting method, device, electronic equipment and readable storage medium storing program for executing, described
Method includes: to obtain object time and target object；Object time feature vector is generated according to the object time；By the mesh
Mark temporal characteristics vector is input in the position score prediction model trained in advance for the target object, obtains target position
Score；The position score prediction model is the machine learning model using set of data samples training；The sample data set packet
Include sample time feature vector and corresponding sample position score；The sample position score is used to indicate the target object
Whether predeterminated position is located at；Determine the target object whether in the predeterminated position according to the target position score.It can
According to object time and model automatic Prediction position score, automatically determine whether target object is located at predeterminated position, by this programme
In scene of the method for offer using businessman's visit, the target (responsible person of businessman) of prediction visit can be very good in target
Between whether in shop (predeterminated position), help to improve visit success rate.
Detailed description of the invention
In order to illustrate more clearly of the technical solution of embodiment of the disclosure, below by the description to embodiment of the disclosure
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only the implementation of the disclosure
Some embodiments of example for those of ordinary skill in the art without any creative labor, can be with
It obtains other drawings based on these drawings.
Fig. 1 shows the step flow chart of position predicting method in one embodiment of the present disclosure；
Fig. 2 shows the step flow charts of the position predicting method in another embodiment of the disclosure；
Fig. 3 shows the structure chart of predicted position device in one embodiment of the present disclosure；
Fig. 4 shows the structure chart of the predicted position device in another embodiment of the disclosure；
The structure chart for the electronic equipment that one embodiment that Fig. 5 shows the disclosure provides.
Specific embodiment
Below in conjunction with the attached drawing in embodiment of the disclosure, the technical solution in embodiment of the disclosure is carried out clear
Chu is fully described by, it is clear that described embodiment is embodiment of the disclosure a part of the embodiment, rather than whole realities
Apply example.Based on the embodiment in embodiment of the disclosure, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, belong to embodiment of the disclosure protection range.
Embodiment one
Referring to Fig.1, it illustrates the step flow chart of position predicting method in one embodiment of the present disclosure, packets
It includes:
Step 101, object time and target object are obtained.
Wherein, object time and target object can be obtained by specified application.Certainly, object time and target pair are obtained
Specified application must be installed as not limiting user, in practical applications, interface can also be will acquire and be integrated in other application.
It is appreciated that specified application can be connect with background server by wireless network or wired network communication.
In embodiment of the disclosure, it is described in detail with the background server side.
Connected applications scene, object time can be the time of sales force businessman responsible person to be visited, and target object is
The responsible person of sales force businessman to be visited.
Step 102, object time feature vector is generated according to the object time.
In practical applications, the time is usually indicated by way of minute second date hour, for example, October 19 in 2018
Day 14:20:00.In machine learning and operation, need to convert the time into machine convenient for the form of identification.
Specifically, firstly, extracting one or more time elements from the time；Then, which is encoded；
Finally, the time element after coding is spliced into temporal characteristics vector.
Step 103, the object time feature vector position trained in advance for the target object is input to obtain
Divide in prediction model, obtains target position score；The position score prediction model is the machine using set of data samples training
Learning model；The sample data set includes sample time feature vector and corresponding sample position score；The sample bit
It sets score and is used to indicate whether the target object is located at predeterminated position.
Wherein, predeterminated position can be set according to practical application scene, and the embodiment of the present invention is without restriction to its.
The input of position score prediction model is object time feature vector, is exported as target position score.It is to be appreciated that
Position score prediction model is different for different objects, so that for each object, trained to obtain position score pre- in training
Model is surveyed, for example, being directed to the responsible person of different businessmans, is respectively trained to obtain position score prediction model.Position score predicts mould
Type is integrated in background server described in step 101, and the background server is facilitated to call the position score prediction model.
It is appreciated that position score can between 0 to 1 value, position score is higher, and target object is in the object time
Interior a possibility that being in predeterminated position, is higher；Position score is lower, and target object is in predeterminated position within the object time
Possibility is lower.
Step 104, determine the target object whether in the predeterminated position according to the target position score.
Specifically, position score can be compared with a standard value, determines target object whether in predeterminated position.
When position score is greater than the standard value, determine target object in predeterminated position；When position score is less than or equal to the standard value
When, determine target object not in predeterminated position.
In conjunction with practical application scene, for example, sales force can go to businessman shops to visit when responsible person businessman shops
Visit responsible person；When responsible person not businessman shops, sales force can be to avoid visit failure.
In conclusion embodiment of the disclosure provides a kind of position predicting method, which comprises when obtaining target
Between and target object；Object time feature vector is generated according to the object time；The object time feature vector is inputted
Into the position score prediction model trained in advance for the target object, target position score is obtained；The position score
Prediction model is the machine learning model using set of data samples training；The sample data set includes sample time feature vector
And corresponding sample position score；The sample position score is used to indicate whether the target object is located at predeterminated position；
Determine the target object whether in the predeterminated position according to the target position score.It can be according to object time and model
Automatic Prediction position score, automatically determines whether target object is located at predeterminated position, and the method that this programme is provided applies businessman
In the scene of visit, whether the target (responsible person of businessman) that can be very good to predict visit is in the object time in shop (default position
Set) in, help to improve visit success rate.
Embodiment two
Referring to Fig. 2, it illustrates the step flow chart of the position predicting method in another embodiment of the disclosure, tools
Body is as follows.
Step 201, the set of data samples for being directed to target object is generated.
Wherein, set of data samples is made of mass data sample, wherein each data sample include sample time feature to
Amount and sample position score.
In embodiment of the disclosure, set of data samples can be from multi-party acquisition.For example, visiting businessman for sales force
The application scenarios of responsible person, set of data samples can be generated according to the call data of sales force, can also be uploaded according to businessman
Location data generate.
Optionally, in another embodiment of the disclosure, above-mentioned steps 201 include sub-step 2011 to 2015:
Sub-step 2011, the first data sample that acquisition target object uploads, and for the second of the target object
Data sample, the data sample include sample uplink time, target object position, predeterminated position, second data sample
Including actual time, actual position score.
Wherein, the first data sample and the second data sample are the data that two different identities upload, the embodiment of the present invention
The data that multiple identity upload can be received in conjunction with practical application scene.For example, visiting businessman responsible person's for sales force
Application scenarios, the first data sample can upload for businessman responsible person oneself, and the second data sample can be to visit the businessman
The sales force of responsible person uploads.
Sales force can upload call data sample with the specified application in login step 101；Businessman responsible person can also be with
The specified application is logged in, current position determination data is uploaded.Certainly, sales force and businessman responsible person are obtained with different identity logs
To different interfaces, so that the information for needing to input is different.
Wherein, sample uplink time can be obtained by system.
Target object position is the current location of businessman responsible person, can be obtained by positioning system.
Predeterminated position can be the position of businessman shops, the position that can be provided in merchant registration.
Actual time is the time that sales force visits businessman responsible person.
Actual position be scored at sales force visit target businessman when businessman responsible person whether businessman shops position.It can manage
Solution, actual position score can be with values for 0, and corresponding not in businessman shops position, value 1 is corresponding in businessman shops position.
Optionally, in another embodiment of the disclosure, above-mentioned predeterminated position passes through following table step such as and determines:
Step A1, the registered location for obtaining the target businessman obtain predeterminated position.
A kind of typical application scenarios of the disclosure are that sales force visits before businessman responsible person, determine businessman responsible person
Whether in shops.
For above-mentioned application scenarios, predeterminated position is the registered location of target businessman.In practical applications, businessman is selling
When being registered on platform, it is desirable to provide the geographical location of businessman, the position are registered location.
Sub-step 2012, respectively according to the sample uplink time, the actual time generate first time feature vector,
Second temporal characteristics vector.
Specifically, firstly, sample uplink time and extraction different time period corresponding temporal information in actual time；So
Afterwards, it is encoded the temporal information of sample uplink time and actual time to obtain temporal characteristics；Finally, sample is uploaded respectively
The temporal characteristics of time and actual time are spliced into first time feature vector, the second temporal characteristics vector.
The step be referred to position score prediction when step 205 to 207 detailed description, details are not described herein.
Sub-step 2013 determines first position score according to the target object position and predeterminated position.
Specifically, the distance between target object position and predeterminated position are calculated；When distance is less than certain distance threshold value,
Target object is indicated in predeterminated position, first position is scored at 1；When distance is greater than or equal to certain distance threshold value, mesh is indicated
Object is marked not in predeterminated position, first position is scored at 0.
Wherein, the distance before target object position and predeterminated position can be calculated using Euclidean distance.
Sub-step 2014, respectively using the first time feature vector, corresponding first position score as sample time
Feature vector and sample position score, are added to set of data samples.
In embodiment of the disclosure, when sales force visited businessman responsible person, visit time and businessman can be responsible for
Whether people is the businessman shops the case where, as data sample.
Sub-step 2015, respectively using the second temporal characteristics vector, corresponding actual position score as sample time
Feature vector and sample position score, are added to set of data samples.
In embodiment of the disclosure, the time and position data obtained time that are uploaded according to businessman responsible person and whether
The case where businessman shops, it can also be used as data sample.
Wherein, position score is used for the training process of monitor model.
Step 202, the set of data samples is trained by Logic Regression Models, is obtained for the target object
Position score prediction model.
Wherein, the formula of Logic Regression Models S (X) is as follows:
S (X)=1/ (1+e-p)(1)
Wherein, the calculation formula of p is as follows:
Wherein, N is the length of temporal characteristics vector, xiFor i-th of element of temporal characteristics vector, ciFor xiCorresponding ginseng
Number, c0For constant parameter.
It is appreciated that the training process of position score prediction model is to be trained to the parameter in above-mentioned formula, directly
To the target position score that above-mentioned formula is calculated closest to the sample position score in sample.
Optionally, in another embodiment of the disclosure, above-mentioned steps 202 include sub-step 2021 to 2025:
Sub-step 2021, the parameter of initialization logic regression model.
Wherein, parameter is the c in formula (2)iAnd c0。
Specifically, the initial value of parameter can be determined based on experience value, so as to effectively reduce the training time, improve instruction
Practice speed.
The sample time feature vector is input in the Logic Regression Models, is calculated pre- by sub-step 2022
Location sets score.
Specifically, sample time feature vector is input in formula (1), obtains the predicted value of target position score.
Sub-step 2023 calculates penalty values according to the predicted position score and sample position score.
Wherein, penalty values can be using Mean square error loss function, absolute error loss function, logarithm loss function, global loss
Function etc..The embodiment of the present invention is without restriction to the calculation of penalty values.
Sub-step 2024 adjusts the logistic regression mould in the case where the penalty values are greater than default penalty values threshold value
The parameter of type, to continue to train.
Wherein, penalty values threshold value can be set according to practical application scene, and embodiment of the disclosure is without restriction to its.
It is appreciated that penalty values threshold value is smaller, the training time is longer, and position score prediction model is more accurate；Penalty values threshold value is bigger, instruction
The white silk time is shorter, and position score prediction model accuracy is poor.
Specifically, it can be directed to the partial derivative of each parameter according to loss function, determine the adjustment direction of the parameter.Example
Such as, partial derivative is greater than 0, then turns the parameter down；Partial derivative then tunes up the parameter less than 0.
Sub-step 2025 terminates training, and will in the case where the penalty values are less than or equal to default penalty values threshold value
Current logic regression model is as the position score prediction model for being directed to the target object.
It is appreciated that the parameter c of Logic Regression Models at the end of trainingiAnd c0For the parameter of position score prediction model.
Step 203, object time and the target businessman of user's input are received.
In conjunction with the application scenarios that the embodiment of the present invention visits businessman responsible person in sales force, boundary can be provided a user
Face, user input object time and target businessman in the corresponding region at the interface, so that system prediction obtains businessman responsible person is
It is no in businessman shop.
Step 204, it obtains the corresponding responsible person of the target businessman and obtains target object.
In practical applications, the relationship between target businessman and responsible person is specified in merchant registration, so that system records
The corresponding relationship.As user selection target businessman, responsible person can be determined.
The step is referred to the detailed description of step 101, and details are not described herein.
Step 205, different time period corresponding temporal information is extracted from the object time.
Wherein, the time cycle can be the moon, day, hour, minute, second etc..
For example, the object time is 14:20:00 on October 19th, 2018, can the moon corresponding temporal information be 10, day is corresponding
Temporal information be 19 and 5 (Fridays are the 5th day of one week), hour corresponding temporal information is 14, the minute corresponding time
Information is 20, and second corresponding temporal information is 0.
Step 206, the temporal information is encoded to obtain temporal characteristics.
In embodiment of the disclosure, each time cycle corresponds at least one temporal information, so that the object time can
To obtain 6 temporal informations, wherein day corresponds to two temporal informations.
In practical applications, each temporal information is encoded, needs to consider each temporal information in coding
Value number, so that for information at the same time, the corresponding unique coding of every kind of value.For example, the value number of the moon is
12, the value number of day is 31 and 7, and the value number of hour is 24, and the value number of minute and second are 60.
It is appreciated that the length of temporal characteristics is related to the value number and encryption algorithm of each temporal information.It is encoding
One timing of algorithm, value number is more, and temporal characteristics are longer；Value number is fewer, and temporal characteristics are shorter.
Optionally, in another embodiment of the disclosure, above-mentioned steps 206 include sub-step 2061:
Sub-step 2061 carries out one-hot coding to the temporal information and obtains temporal characteristics.
Wherein, one-hot coding is one-hot coding, it may be assumed that feature has n value, then feature is indicated with n bit.
For example, there is 7 values day, be then indicated with 7 bits, that is, exist 0000001,0000010,0000100,0001000,
0010000,0100000,1,000,000 7 kind of value.
Step 207, the temporal characteristics are spliced into object time feature vector.
Specifically, using each temporal characteristics as each element of object time feature vector.For example, for the above-mentioned time
Feature splices the moon, day, hour, minute, second corresponding temporal characteristics in sequence, and it is special to obtain the object time that length is 194
Levy vector, wherein account for 12 the moon, account for 38 day, hour accounts for 24, and minute and second account for 60 respectively.
In practical applications, since the precision of minute and second are too low, the moon, day, hour can be taken to generate the mesh that length is 74
Mark temporal characteristics vector.
It is appreciated that temporal characteristics can preferentially be spliced from big to small according to the time cycle in splicing.Certainly, may be used
Not limit sequence.
Step 208, the object time feature vector position trained in advance for the target object is input to obtain
Divide in prediction model, obtains target position score；The position score prediction model is the machine using set of data samples training
Learning model；The sample data set includes sample time feature vector and corresponding sample position score；The sample bit
It sets score and is used to indicate whether the target object is located at predeterminated position.
The step is referred to the detailed description of step 103, and details are not described herein.
Step 209, in the case where the target position score is greater than or equal to predeterminated position score threshold, described in determination
Target object is in the predeterminated position.
Wherein, position score threshold can be set according to practical application scene, and the embodiment of the present invention is without restriction to its.
It is appreciated that target position score very great talent will can determine that target pair when in the very high situation of position score threshold
As if it is no in predeterminated position, so that the accuracy in predeterminated position is higher, it is lower in the accuracy of predeterminated position, it is more likely that
Reality is judged as not in predeterminated position in predeterminated position；In the case that position score threshold is lower, target position score is very
It is small to be assured that in predeterminated position, so that the accuracy in predeterminated position is lower, it is not higher in the accuracy of predeterminated position, very
It is possible that not being judged as practical in predeterminated position in predeterminated position.
Step 210, in the case where the target position score is less than predeterminated position score threshold, the target pair is determined
As not in the predeterminated position.
It is appreciated that when target object is not in predeterminated position, it can also be by target position score and position score threshold
User (such as sales force) is showed, so that user be helped to further determine whether to go out to visit with the target object.For example, if mesh
Cursor position score and position score threshold relatively, so that user thinks that target object is possible to use at this time in predeterminated position
Family can determine to go out to visit with；If target position score and position score threshold difference are larger, so that user thinks that target object is agreed
Determine not in predeterminated position, user can determine not go out to visit at this time.
Judgment basis can also be showed user by embodiment of the disclosure, so that user voluntarily judges, and combine this public affairs
The machine judging result opened is realized and is more accurately predicted in shop.
In conclusion embodiment of the disclosure provides a kind of position predicting method, which comprises when obtaining target
Between and target object；Object time feature vector is generated according to the object time；The object time feature vector is inputted
Into the position score prediction model trained in advance for the target object, target position score is obtained；The position score
Prediction model is the machine learning model using set of data samples training；The sample data set includes sample time feature vector
And corresponding sample position score；The sample position score is used to indicate whether the target object is located at predeterminated position；
Determine the target object whether in the predeterminated position according to the target position score.It can be according to object time and model
Automatic Prediction position score, automatically determines whether target object is located at predeterminated position, and the method that this programme is provided applies businessman
In the scene of visit, whether the target (responsible person of businessman) that can be very good to predict visit is in the object time in shop (default position
Set) in, help to improve visit success rate.
Embodiment three
Referring to Fig. 3, it illustrates the structure charts of predicted position device in one embodiment of the present disclosure, specifically such as
Under.
Data obtaining module 301, for obtaining object time and target object.
Temporal characteristics vector generation module 302, for generating object time feature vector according to the object time.
Position score prediction module 303, for being input to the object time feature vector for the target object
In advance in trained position score prediction model, target position score is obtained；The position score prediction model is to use data
The machine learning model of sample set training；The sample data set includes sample time feature vector and corresponding sample position
Score；The sample position score is used to indicate whether the target object is located at predeterminated position.
Position determination module 304, for determining the target object whether described pre- according to the target position score
If position.
In conclusion embodiment of the disclosure provides a kind of predicted position device, described device includes: acquisition of information mould
Block, for obtaining object time and target object；Temporal characteristics vector generation module, for generating mesh according to the object time
Mark temporal characteristics vector；Position score prediction module, for being input to the object time feature vector for the target
In the position score prediction model that object is trained in advance, target position score is obtained；The position score prediction model is to use
The machine learning model of set of data samples training；The sample data set includes sample time feature vector and corresponding sample
Position score；The sample position score is used to indicate whether the target object is located at predeterminated position；Position determination module is used
In determining the target object whether in the predeterminated position according to the target position score.It can be according to object time and mould
Type automatic Prediction position score, automatically determines whether target object is located at predeterminated position, and the method that this programme is provided applies quotient
In the scene of family's visit, whether the target (responsible person of businessman) that can be very good prediction visit is (default in shop in the object time
Position) in, help to improve visit success rate.
Three corresponding method embodiment one of Installation practice, detailed description are referred to embodiment one, and details are not described herein.
Example IV
Referring to Fig. 4, it illustrates the structure charts of the predicted position device in another embodiment of the disclosure, specifically such as
Under.
Set of data samples generation module 401, for generating the set of data samples for being directed to target object.
Position score prediction model training module 402, for being carried out by Logic Regression Models to the set of data samples
Training obtains the position score prediction model for the target object.
Data obtaining module 403, for obtaining object time and target object；Optionally, real in the another kind of the disclosure
It applies in example, above- mentioned information obtain module 403 and include:
Information receiving submodule 4031, for receiving object time and the target businessman of user's input.
Semantic object extraction submodule 4032 obtains target object for obtaining the corresponding responsible person of the target businessman.
Temporal characteristics vector generation module 404, for generating object time feature vector according to the object time；It is optional
Ground, in the embodiments of the present disclosure, above-mentioned temporal characteristics vector generation module 404, comprising:
Temporal information extracting sub-module 4041, for extracting the corresponding time in different time period from the object time
Information.
Time encoding submodule 4042, for being encoded to obtain temporal characteristics to the temporal information.
Temporal characteristics vector splices submodule 4043, for the temporal characteristics to be spliced into object time feature vector.
Position score prediction module 405, for being input to the object time feature vector for the target object
In advance in trained position score prediction model, target position score is obtained；The position score prediction model is to use data
The machine learning model of sample set training；The sample data set includes sample time feature vector and corresponding sample position
Score；The sample position score is used to indicate whether the target object is located at predeterminated position.
Position determination module 406, for determining the target object whether described pre- according to the target position score
If position；Optionally, in embodiment of the disclosure, above-mentioned position determination module 406 includes:
First position determines submodule 4061, for being greater than or equal to predeterminated position score threshold in the target position score
In the case where value, determine the target object in the predeterminated position.
The second position determines submodule 4062, for being less than the feelings of predeterminated position score threshold in the target position score
Under condition, determine the target object not in the predeterminated position.
Optionally, in another embodiment of the disclosure, the training data sample includes temporal characteristics vector sum position
Score is set, above-mentioned set of data samples generation module 401 includes:
Data-acquisition submodule, the first data sample uploaded for acquiring target object, and it is directed to the target pair
The second data sample of elephant, the data sample include sample uplink time, target object position, predeterminated position, and described second
Data sample includes actual time, actual position score.
Sample vector generates submodule, for generating first according to the sample uplink time, the actual time respectively
Temporal characteristics vector, the second temporal characteristics vector.
First position score determines submodule, for determining first position according to the target object position and predeterminated position
Score.
First sample set generates submodule, for respectively obtaining the first time feature vector, corresponding first position
It is allocated as being added to set of data samples for sample time feature vector and sample position score.
Second sample set generates submodule, for respectively obtaining the second temporal characteristics vector, corresponding actual position
It is allocated as being added to set of data samples for sample time feature vector and sample position score.
Optionally, in another embodiment of the disclosure, above-mentioned position score prediction model training module 402 includes:
Initialization submodule, the parameter for initialization logic regression model.
Predictor calculation submodule, for the sample time feature vector to be input in the Logic Regression Models,
Predicted position score is calculated.
Penalty values computational submodule, for calculating penalty values according to the predicted position score and sample position score.
Continue to train submodule, for being patrolled described in adjustment in the case where the penalty values are greater than default penalty values threshold value
The parameter of regression model is collected, to continue to train.
Terminate training submodule, for terminating in the case where the penalty values are less than or equal to default penalty values threshold value
Training, and using current logic regression model as the position score prediction model for being directed to the target object.
Optionally, in another embodiment of the disclosure, above-mentioned time encoding submodule 4042 includes:
Time encoding unit obtains temporal characteristics for carrying out one-hot coding to the temporal information.
In conclusion embodiment of the disclosure provides a kind of predicted position device, described device includes: acquisition of information mould
Block, for obtaining object time and target object；Temporal characteristics vector generation module, for generating mesh according to the object time
Mark temporal characteristics vector；Position score prediction module, for being input to the object time feature vector for the target
In the position score prediction model that object is trained in advance, target position score is obtained；The position score prediction model is to use
The machine learning model of set of data samples training；The sample data set includes sample time feature vector and corresponding sample
Position score；The sample position score is used to indicate whether the target object is located at predeterminated position；Position determination module is used
In determining the target object whether in the predeterminated position according to the target position score.It can be according to object time and mould
Type automatic Prediction position score, automatically determines whether target object is located at predeterminated position, and the method that this programme is provided applies quotient
In the scene of family's visit, whether the target (responsible person of businessman) that can be very good prediction visit is (default in shop in the object time
Position) in, help to improve visit success rate.
Four corresponding method embodiment two of Installation practice, detailed description are referred to embodiment two, and details are not described herein.
Embodiment of the disclosure additionally provides a kind of electronic equipment, referring to Fig. 5, comprising: processor 501, memory 502 with
And it is stored in the computer program 5021 that can be run on the memory 502 and on the processor 501, the processor is held
The position predicting method of previous embodiment is realized when row described program.
Embodiment of the disclosure additionally provides a kind of readable storage medium storing program for executing, when the instruction in the storage medium is set by electronics
When standby processor executes, so that electronic equipment is able to carry out the position predicting method of previous embodiment.
For device embodiment, since it is basically similar to the method embodiment, related so being described relatively simple
Place illustrates referring to the part of embodiment of the method.
Algorithm and display are not inherently related to any particular computer, virtual system, or other device provided herein.
Various general-purpose systems can also be used together with teachings based herein.As described above, it constructs required by this kind of system
Structure be obvious.In addition, embodiment of the disclosure is also not for any particular programming language.It should be understood that can be with
The content of embodiment of the disclosure described herein is realized using various programming languages, and is retouched above to what language-specific was done
Stating is preferred forms in order to disclose embodiment of the disclosure.
In the instructions provided here, numerous specific details are set forth.It is to be appreciated, however, that the implementation of the disclosure
The embodiment of example can be practiced without these specific details.In some instances, it is not been shown in detail well known
Methods, structures and technologies, so as not to obscure the understanding of this specification.
Similarly, it should be understood that in order to simplify the disclosure and help to understand one or more of the various inventive aspects,
Above in the description of the exemplary embodiment of embodiment of the disclosure, each feature of embodiment of the disclosure is sometimes by together
It is grouped into single embodiment, figure or descriptions thereof.However, it is as follows that the method for the disclosure should not be construed to reflection
Be intended to: embodiment of the disclosure i.e. claimed requires more more than feature expressly recited in each claim
Feature.More precisely, as reflected in the following claims, inventive aspect is single less than disclosed above
All features of embodiment.Therefore, it then follows thus claims of specific embodiment are expressly incorporated in the specific embodiment party
Formula, wherein separate embodiments of each claim as embodiment of the disclosure itself.
Those skilled in the art will understand that can be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more devices different from this embodiment.It can be the module or list in embodiment
Member or component are combined into a module or unit or component, and furthermore they can be divided into multiple submodule or subelement or
Sub-component.Other than such feature and/or at least some of process or unit exclude each other, it can use any
Combination is to all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so disclosed
All process or units of what method or apparatus are combined.Unless expressly stated otherwise, this specification is (including adjoint power
Benefit require, abstract and attached drawing) disclosed in each feature can carry out generation with an alternative feature that provides the same, equivalent, or similar purpose
It replaces.
The various component embodiments of embodiment of the disclosure can be implemented in hardware, or in one or more processing
The software module run on device is realized, or is implemented in a combination thereof.It will be understood by those of skill in the art that can be in reality
Trample it is middle realized using microprocessor or digital signal processor (DSP) it is according to an embodiment of the present disclosure in the pre- measurement equipment in shop
In some or all components some or all functions.Embodiment of the disclosure is also implemented as executing here
Some or all device or device programs of described method.Such program for realizing embodiment of the disclosure
It can store on a computer-readable medium, or may be in the form of one or more signals.Such signal can be with
It downloads from internet website, is perhaps provided on the carrier signal or is provided in any other form.
It should be noted that above-described embodiment illustrates rather than to embodiment of the disclosure embodiment of the disclosure
It is limited, and those skilled in the art can be designed replacement without departing from the scope of the appended claims and implement
Example.In the claims, any reference symbol between parentheses should not be configured to limitations on claims.Word
"comprising" does not exclude the presence of element or step not listed in the claims.Word "a" or "an" located in front of the element is not
There are multiple such elements for exclusion.Embodiment of the disclosure can be by means of including the hardware of several different elements and borrowing
Help properly programmed computer to realize.In the unit claims listing several devices, several in these devices
A can be is embodied by the same item of hardware.The use of word first, second, and third does not indicate any suitable
Sequence.These words can be construed to title.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
The foregoing is merely the preferred embodiments of embodiment of the disclosure, not to limit the implementation of the disclosure
Example, all made any modifications, equivalent replacements, and improvements etc. within the spirit and principle of embodiment of the disclosure should all include
Within the protection scope of embodiment of the disclosure.
The above, the only specific embodiment of embodiment of the disclosure, but the protection scope of embodiment of the disclosure
It is not limited thereto, anyone skilled in the art, can in the technical scope that embodiment of the disclosure discloses
Change or replacement are readily occurred in, should all be covered within the protection scope of embodiment of the disclosure.Therefore, embodiment of the disclosure
Protection scope should be subject to the protection scope in claims.
Claims (12)
1. a kind of position predicting method, which is characterized in that the described method includes:
Obtain object time and target object；
Object time feature vector is generated according to the object time；
The object time feature vector is input in the position score prediction model trained in advance for the target object,
Obtain target position score；The position score prediction model is the machine learning model using set of data samples training；It is described
Sample data set includes sample time feature vector and corresponding sample position score；The sample position score is used to indicate
Whether the target object is located at predeterminated position；
Determine the target object whether in the predeterminated position according to the target position score.
2. the method according to claim 1, wherein described generate object time feature according to the object time
The step of vector, comprising:
Different time period corresponding temporal information is extracted from the object time；
The temporal information is encoded to obtain temporal characteristics；
The temporal characteristics are spliced into object time feature vector.
3. the method according to claim 1, wherein described determine the target according to the target position score
Whether object is the predeterminated position the step of, comprising:
In the case where the target position score is greater than or equal to predeterminated position score threshold, determine the target object in institute
State predeterminated position；
In the case where the target position score is less than predeterminated position score threshold, determine the target object not described pre-
If position.
4. according to the method described in claim 2, it is characterized in that, described encode the temporal information to obtain time spy
The step of sign, comprising:
One-hot coding is carried out to the temporal information and obtains temporal characteristics.
5. the method according to claim 1, wherein the position score prediction model is instructed as follows
Practice:
Generate the set of data samples for being directed to target object；
The set of data samples is trained by Logic Regression Models, the position score for obtaining being directed to the target object is pre-
Survey model.
6. according to the method described in claim 5, it is characterized in that, the set of data samples include sample time feature vector and
Sample position score, described the step of generating the set of data samples for being directed to target businessman, comprising:
The first data sample that target object uploads is acquired, and for the second data sample of the target object, the number
It include sample uplink time, target object position, predeterminated position according to sample, second data sample includes actual time, true
Real position score；
Respectively according to the sample uplink time, the actual time generate first time feature vector, the second temporal characteristics to
Amount；
First position score is determined according to the target object position and predeterminated position；
Respectively using the first time feature vector, corresponding first position score as sample time feature vector and sample bit
Score is set, set of data samples is added to；
Respectively using the second temporal characteristics vector, corresponding actual position score as sample time feature vector and sample bit
Score is set, set of data samples is added to.
7. according to the method described in claim 6, it is characterized in that, it is described by Logic Regression Models to the set of data samples
The step of being trained, obtaining the position score prediction model for the target object, comprising:
The parameter of initialization logic regression model；
The sample time feature vector is input in the Logic Regression Models, predicted position score is calculated；
Penalty values are calculated according to the predicted position score and sample position score；
In the case where the penalty values are greater than default penalty values threshold value, the parameter of the Logic Regression Models is adjusted, to continue
Training；
In the case where the penalty values are less than or equal to default penalty values threshold value, terminate training, and current logic is returned into mould
Type is as the position score prediction model for being directed to the target object.
8. the method according to claim 1, wherein the step of acquisition object time and target object, packet
It includes:
Receive object time and the target businessman of user's input；
It obtains the corresponding responsible person of the target businessman and obtains target object.
9. according to the method described in claim 8, it is characterized in that, the predeterminated position determines as follows:
The registered location for obtaining the target businessman obtains predeterminated position.
10. a kind of predicted position device, which is characterized in that described device includes:
Data obtaining module, for obtaining object time and target object；
Temporal characteristics vector generation module, for generating object time feature vector according to the object time；
Position score prediction module is trained for being input to the object time feature vector for the target object in advance
Position score prediction model in, obtain target position score；The position score prediction model is to be assembled for training using data sample
Experienced machine learning model；The sample data set includes sample time feature vector and corresponding sample position score；Institute
It states sample position score and is used to indicate whether the target object is located at predeterminated position；
Position determination module, for determining the target object whether in the predeterminated position according to the target position score.
11. a kind of electronic equipment characterized by comprising
Processor, memory and it is stored in the computer program that can be run on the memory and on the processor,
It is characterized in that, the processor realizes the position prediction as described in one or more in claim 1-9 when executing described program
Method.
12. a kind of readable storage medium storing program for executing, which is characterized in that when the instruction in the storage medium is held by the processor of electronic equipment
When row, so that electronic equipment is able to carry out the position predicting method as described in one or more in claim to a method 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811379242.0A CN109726847A (en) | 2018-11-19 | 2018-11-19 | Position predicting method, device, electronic equipment and readable storage medium storing program for executing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811379242.0A CN109726847A (en) | 2018-11-19 | 2018-11-19 | Position predicting method, device, electronic equipment and readable storage medium storing program for executing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109726847A true CN109726847A (en) | 2019-05-07 |
Family
ID=66294666
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811379242.0A Withdrawn CN109726847A (en) | 2018-11-19 | 2018-11-19 | Position predicting method, device, electronic equipment and readable storage medium storing program for executing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109726847A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126049A (en) * | 2019-12-14 | 2020-05-08 | 中国科学院深圳先进技术研究院 | Object relation prediction method and device, terminal equipment and readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020165755A1 (en) * | 2001-03-29 | 2002-11-07 | Kitts Brendan J. | Method of predicting behavior of a customer at a future date and a data processing system readable medium |
| CN101247392A (en) * | 2008-02-28 | 2008-08-20 | 中兴通讯股份有限公司 | Objective activity estimation device and method |
| CN104737565A (en) * | 2012-10-19 | 2015-06-24 | 脸谱公司 | Method relating to predicting the future state of a mobile device user |
| JP6035404B1 (en) * | 2015-11-17 | 2016-11-30 | 日本生命保険相互会社 | Visit preparation system |
| CN107730094A (en) * | 2017-09-27 | 2018-02-23 | 成都爱杰特电子科技有限公司 | A kind of client relation management method and system |
| CN107977980A (en) * | 2017-12-06 | 2018-05-01 | 北京飞搜科技有限公司 | A kind of method for tracking target, equipment and computer-readable recording medium |
| CN108427658A (en) * | 2018-03-12 | 2018-08-21 | 北京奇艺世纪科技有限公司 | A kind of data predication method, device and electronic equipment |
-
2018
- 2018-11-19 CN CN201811379242.0A patent/CN109726847A/en not_active Withdrawn
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020165755A1 (en) * | 2001-03-29 | 2002-11-07 | Kitts Brendan J. | Method of predicting behavior of a customer at a future date and a data processing system readable medium |
| CN101247392A (en) * | 2008-02-28 | 2008-08-20 | 中兴通讯股份有限公司 | Objective activity estimation device and method |
| CN104737565A (en) * | 2012-10-19 | 2015-06-24 | 脸谱公司 | Method relating to predicting the future state of a mobile device user |
| JP6035404B1 (en) * | 2015-11-17 | 2016-11-30 | 日本生命保険相互会社 | Visit preparation system |
| CN107730094A (en) * | 2017-09-27 | 2018-02-23 | 成都爱杰特电子科技有限公司 | A kind of client relation management method and system |
| CN107977980A (en) * | 2017-12-06 | 2018-05-01 | 北京飞搜科技有限公司 | A kind of method for tracking target, equipment and computer-readable recording medium |
| CN108427658A (en) * | 2018-03-12 | 2018-08-21 | 北京奇艺世纪科技有限公司 | A kind of data predication method, device and electronic equipment |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126049A (en) * | 2019-12-14 | 2020-05-08 | 中国科学院深圳先进技术研究院 | Object relation prediction method and device, terminal equipment and readable storage medium |
| CN111126049B (en) * | 2019-12-14 | 2023-11-24 | 中国科学院深圳先进技术研究院 | Object relation prediction method, device, terminal equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10957105B2 (en) | Augmented reality geolocation optimization | |
| CN108805091B (en) | Method and apparatus for generating a model | |
| US11026047B2 (en) | Associating multiple user devices with a single user | |
| CN111209477B (en) | Information recommendation method and device, electronic equipment and storage medium | |
| US10386190B2 (en) | Cognitive guide system for multi-objective excursion | |
| CN109064278A (en) | Target object recommended method and device, electronic equipment, storage medium | |
| KR102135477B1 (en) | Method and system for image automatic classification | |
| CN104484361B (en) | Information shows method and device | |
| CN109978642A (en) | A kind of information recommendation method, device and communication equipment | |
| CN108197203A (en) | A kind of shop front head figure selection method, device, server and storage medium | |
| CN111294620A (en) | Video recommendation method and device | |
| US20200409451A1 (en) | Personalized content for augemented reality based on past user experience | |
| CN109726847A (en) | Position predicting method, device, electronic equipment and readable storage medium storing program for executing | |
| CN111026849A (en) | Data processing method and device | |
| CN113194270B (en) | Video processing method and device, electronic equipment and storage medium | |
| CN112269943B (en) | Information recommendation system and method | |
| CN109840072A (en) | Information processing method and device | |
| CN108833971A (en) | A kind of method for processing video frequency and device | |
| US20210116251A1 (en) | Location calibration based on movement path and map objects | |
| CN112256957A (en) | Information sorting method and device, electronic equipment and storage medium | |
| US20200101375A1 (en) | Deep learning from real world and digital exemplars | |
| CN110008398A (en) | A kind of data classification management recommended method and device | |
| US20170272908A1 (en) | Location context inference based on user mobile data with uncertainty | |
| US11373407B2 (en) | Attention generation | |
| CN114500900A (en) | Method and device for searching lost object |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190507 |