CN110503242A - Businessman's order method, apparatus, electronic equipment and storage medium - Google Patents
Businessman's order method, apparatus, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110503242A CN110503242A CN201910671992.3A CN201910671992A CN110503242A CN 110503242 A CN110503242 A CN 110503242A CN 201910671992 A CN201910671992 A CN 201910671992A CN 110503242 A CN110503242 A CN 110503242A
- Authority
- CN
- China
- Prior art keywords
- order
- current
- businessman
- mode
- order mode
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Record Information Processing For Printing (AREA)
- Calculators And Similar Devices (AREA)
- Electronic Switches (AREA)
Abstract
The embodiment of the present application discloses a kind of businessman's order method, apparatus, electronic equipment and storage medium, the described method includes: obtaining the first order related data of current businessman and the order related data of similar businessman, as data sample, and obtain order mode label, label includes order mode, further includes order success rate or loses single rate;Data sample is encoded, sample input data is obtained;According to sample input data and label, order recommended models are trained;The the second order related data for obtaining current businessman encodes as current order related data, and to current order related data, obtains input data;According to input data, the recommendation probability of a variety of order modes is predicted by order recommended models;According to probability is recommended, at least one order mode is chosen from a variety of order modes, and determine priority;At least one order mode and corresponding priority are recommended into current businessman.Order success rate can be improved in the embodiment of the present application.
Description
Technical field
This application involves Internet technical fields, more particularly to a kind of businessman's order method, apparatus, electronic equipment and deposit
Storage media.
Background technique
It is taken out in order scene in businessman, after new order generates, order meeting time-out is taken if the non-order of businessman in more than 5 minutes
Disappear.
Currently, the optional way of businessman's order has: computer order, APP (Application, application program) order, GPRS
(General Packet Radio Service, general packet radio service) order, voice reminder and open platform order etc.,
The mode of order is more dispersed, and for businessman, businessman does not simultaneously know about all order modes, even if understand do not know as
What is selected, and still using the order mode got used to, this, which is just easily caused, loses list.
Summary of the invention
The embodiment of the present application provides a kind of businessman's order method, apparatus, electronic equipment and storage medium, helps to improve quotient
Family's order success rate.
To solve the above-mentioned problems, in a first aspect, the embodiment of the present application provides a kind of businessman's order method, comprising:
The first order related data of current businessman and the order related data of similar businessman are obtained, as data sample
This, and the corresponding order mode label of each data sample is obtained, it further includes connecing that the order mode label, which includes order mode,
Single success rate loses single rate;
The data sample is encoded, the sample input data of order recommended models is obtained;
According to the sample input data and corresponding order mode label, order recommended models are trained, so that
The output of order recommended models is the recommendation probability of a variety of order modes, obtains the order recommended models of training completion;
The the second order related data for obtaining current businessman, as current order related data, and to the current order
Related data is encoded, and the input data of order recommended models is obtained;
According to the input data, the recommendation of a variety of order modes is predicted by the order recommended models that the training is completed
Probability;
According to the recommendation probability, at least one order mode is chosen from a variety of order modes, and described in determination
The priority of at least one order mode;
Order mode is recommended to combine at least one order mode and corresponding priority composition;
Give the recommendation order mode combined recommendation to the current businessman.
Optionally, the data sample and the current order related data respectively include: user behavior day 2, history are ordered
Odd-numbered day will, businessman's location distribution, order mode vivo identification, order scene, time of received orders distribution, history order mode with
And history loses at least one of single reason distribution.
Optionally, after giving the recommendation order mode combined recommendation to the current businessman, further includes:
Obtain the current state of current the order mode and order equipment of the current businessman;
According to the current order mode, the recommendation order mode combines and the current state, loses determining to exist
When single risk, risk prompting and settling mode are pushed to the current businessman.
Optionally, after giving the recommendation order mode combined recommendation to the current businessman, further includes:
When detecting that the current businessman loses single, obtains corresponding single reason of losing and be distributed, and obtain the current businessman
Current order mode and order equipment current state;
Single reason distribution, the current order mode, the current state and the recommendation order mode are lost according to described
Single settling mode is lost in combination, determination;
It loses single settling mode by described and is sent to the current businessman.
Optionally, after giving the recommendation order mode combined recommendation to the current businessman, further includes:
Obtain the current state of current the order mode and order equipment of the current businessman;
When the current state and the current order mode mismatch, connect according to the current state and the recommendation
The combination of folk prescription formula, determines best order mode, and the current order mode is switched to the best order mode.
Optionally, after giving the recommendation order mode combined recommendation to the current businessman, further includes:
If the current order mode of the current businessman is that the recommendation order mode one of combines and loses single rate and reaches
Single rate is lost to default, then obtains and loses single reason distribution;
Single reason distribution is lost according to described, the order recommended models are optimized.
Second aspect, the embodiment of the present application provide a kind of businessman's order device, comprising:
Sample acquisition module, for obtaining the first order related data of current businessman and the order correlation of similar businessman
Data as data sample, and obtain the corresponding order mode label of each data sample, and the order mode label includes connecing
Folk prescription formula further includes order success rate or loses single rate;
Encoding samples module obtains the sample input number of order recommended models for encoding to the data sample
According to;
Model training module, for recommending order according to the sample input data and corresponding order mode label
Model is trained, so that the output of order recommended models is the recommendation probability of a variety of order modes, obtains connecing for training completion
Single recommended models;
Data acquisition module, for obtaining the second order related data of current businessman, as current order related data,
And the current order related data is encoded, obtain the input data of order recommended models;
Recommend probabilistic forecasting module, for passing through the order recommended models of the training completion according to the input data
Predict the recommendation probability of a variety of order modes;
Order mode chooses module, for choosing at least one from a variety of order modes according to the recommendation probability
Kind order mode, and determine the priority of at least one order mode;
Recommend order mode composite module, for recommending at least one order mode and corresponding priority composition
Order mode combines;
Order recommending module, for giving the recommendation order mode combined recommendation to the current businessman.
Optionally, the data sample and the current order related data respectively include: User action log, history are ordered
Odd-numbered day will, businessman's location distribution, order mode vivo identification, order scene, time of received orders distribution, history order mode with
And history loses at least one of single reason distribution.
Optionally, described device further include:
Current order data acquisition module, for obtaining the current order mode of the current businessman and working as order equipment
Preceding state;
Risk reminding module, for being combined according to the current order mode, the recommendation order mode and described current
State pushes risk prompting and settling mode to the current businessman when determining in the presence of single risk is lost.
Optionally, described device further include:
It loses forms data and obtains module, divide for when detecting that the current businessman loses single, obtaining corresponding single reason of losing
Cloth, and obtain the current state of current the order mode and order equipment of the current businessman;
Settling mode determining module, for losing single reason distribution, the current order mode, the current shape according to
State and the recommendation order mode combine, and single settling mode is lost in determination;
Settling mode sending module, for losing single settling mode by described and being sent to the current businessman.
Optionally, described device further include:
Current order data acquisition module, for obtaining the current order mode of the current businessman and working as order equipment
Preceding state;
Order mode switching module, for when the current state and the current order mode mismatch, according to institute
It states current state and the recommendation order mode combines, determine best order mode, and the current order mode is switched to
The best order mode.
Optionally, described device further include:
It loses single reason and obtains module, if the current order mode for the current businessman is the recommendation order mode group
It one of closes and loses single rate and reach default and lose single rate, then obtain and lose single reason distribution;
Model optimization module optimizes the order recommended models for losing single reason distribution according to.
The third aspect, the embodiment of the present application also disclose a kind of electronic equipment, including memory, processor and are stored in institute
The computer program that can be run on memory and on a processor is stated, the processor realizes this when executing the computer program
Apply for businessman's order method described in embodiment.
Fourth aspect, the embodiment of the present application provide a kind of computer readable storage medium, are stored thereon with computer journey
Sequence, when which is executed by processor disclosed in the embodiment of the present application the step of businessman's order method.
Businessman's order method, apparatus, electronic equipment and storage medium disclosed in the embodiment of the present application, by obtaining current quotient
The order related data of family and the order related data of similar businessman, as data sample, and obtain each data sample pair
The order mode label answered, encodes the data sample, the sample input data of order recommended models is obtained, according to institute
Sample input data and corresponding order mode label are stated, order recommended models are trained, so that order recommended models
Output is the recommendation probability of a variety of order modes, obtains the order recommended models of training completion, and obtain current businessman second connects
Simple correlation data are encoded as current order related data, and to the current order related data, obtain order recommendation
The input data of model predicts a variety of order sides by the order recommended models that the training is completed according to the input data
The recommendation probability of formula is chosen at least one order mode from a variety of order modes according to the recommendation probability, and is determined
At least one order mode and corresponding priority composition are recommended order by the priority of at least one order mode
Mode combines, and gives the recommendation order mode combined recommendation to the current businessman, since order recommended models are to current
What the related data of businessman and similar businessman were learnt, the recommendation of the current businessman determined by order recommended models connects
The combination of folk prescription formula includes each order mode and corresponding priority, thus in current businessman is combined using recommendation order mode
Order mode when carrying out order, order success rate can be improved.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application, below will be in embodiment or description of the prior art
Required attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is only some realities of the application
Example is applied, it for those of ordinary skill in the art, without any creative labor, can also be attached according to these
Figure obtains other attached drawings.
Fig. 1 is the flow chart of businessman's order method of the embodiment of the present application one;
Fig. 2 is the structural schematic diagram of businessman's order device of the embodiment of the present application two.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiment is some embodiments of the present application, instead of all the embodiments.Based on this Shen
Please in embodiment, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall in the protection scope of this application.
Embodiment one
A kind of businessman's order method disclosed in the present embodiment, as shown in Figure 1, this method comprises: step 110 is to step 130.
Step 110, the first order related data of current businessman and the order related data of similar businessman are obtained, as
Data sample, and the corresponding order mode label of each data sample is obtained, the order mode label includes order mode, also
Including order success rate or lose single rate.
Wherein, similar businessman is other businessmans identical with current businessman's classification, for example, the businessman is outside food and drink
Businessman is sold, then similar businessman is that businessman is taken out in other food and drink.The order related data is number relevant to businessman's order mode
According to.The data sample includes but is not limited to: the User action log of the businessman, History Order log, businessman geographical location
Distribution, the distribution of order mode vivo identification, order scene, time of received orders, history order mode and history lose single reason distribution
At least one of.The data sample can also include user's representation data of corresponding merchant, such as the order that businessman likes
Mode etc..Order mode includes: computer order, APP order, GPRS order, voice reminder or open platform order etc..It is described
APP order includes the order of Android platform mobile phone or ios platform mobile phone, and mobile phone order includes manual order or automatic order.Institute
Stating GPRS order is the side that GPRS printer passes through USB (Universal Serial Bus, universal serial bus) interface tandem
Formula or the mode of remote service access carry out order.Voice reminder is businessman when being more than that setting time does not have order, is passed through
The mode of call voice carries out reminder, assists businessman's order, and setting time for example can be 2 point half or 3 point half etc..It is described to open
Being laid flat platform order needs businessman to buy third party's service to complete order.
Wherein, it is artificial order or the automatic order of machine that order mode vivo identification, which is to discriminate between order mode,.Order scene
Including network state (such as 3G, 4G) and/or whether long-range order (such as order mode be mobile phone order when, mobile phone is not in businessman
In shops) etc..When order mode is mobile phone order, it may include: that APP is killed or abnormal jingle bell that history, which loses single reason distribution,
Deng.
User action log and History Order log first to current businessman sample, and obtain for the first user behavior day
Will and the first History Order log, analyze first User action log and the first History Order log, and determination connects
Folk prescription formula vivo identification, order scene, time of received orders distribution, history order mode and history lose single reason distribution, and obtain
Businessman's location distribution by first User action log, the first History Order log, businessman's location distribution, connects
Folk prescription formula vivo identification, order scene, time of received orders distribution, history order mode and history lose single reason distribution as first
Order related data.Since the history order mode that current businessman uses may not be complete, some order modes not configured have
It may be showed in the use of other businessmans well, it therefore, can be according to the order related data of similar businessman, expanding data sample
This.By the first order related data of current businessman and the order related data of similar businessman, collectively as data sample.Identification
The order mode of each data sample, and calculate the corresponding order success rate of the order mode or lose single rate, obtain each data
The corresponding order mode label of sample.
Step 120, the data sample is encoded, obtains the sample input data of order recommended models.
Wherein, order recommended models are machine learning model, such as convolutional neural networks model.
The data sample carries out coded treatment, and data sample is encoded to the sample input data of order recommended models.
Step 130, according to the sample input data and corresponding order mode label, order recommended models are instructed
Practice, so that the output of order recommended models is the recommendation probability of a variety of order modes, obtains the order recommended models of training completion.
Sample input data is inputted into order recommended models, obtains the order mode and corresponding of order recommended models output
Recommend probability, and the order mode and corresponding recommendation probability are compared with order mode label, mould is recommended in adjustment order
The parameter of type, until the training of order recommended models is completed.Wherein, recommend the order success rate pair in probability and order mode label
It answers, the order success rate in order mode label is 1 with the sum of single rate is lost.
Step 140, the second order related data for obtaining current businessman, as current order related data, and to described
Current order related data is encoded, and the input data of order recommended models is obtained.
Wherein, the second order related data is order dependency number different from the first order related data in current businessman
According to.The current order related data includes but is not limited to: the User action log of the businessman, History Order log, businessman
Location distribution, order mode vivo identification, order scene, time of received orders distribution, history order mode and history lose list
At least one of reason distribution.The data sample can also include user's representation data of the current businessman, such as quotient
The order mode etc. that family is liked.
It, can be using the sample rate different from the first order related data to working as when recommending order mode to current businessman
The User action log of preceding businessman and History Order log sample, and obtain second user user behaviors log and the second History Order
The second user user behaviors log and the second History Order log are analyzed in log, are determined order mode vivo identification, are connect
Single game scape, time of received orders distribution, history order mode and history lose single reason distribution, and obtain businessman's location distribution,
By the second user user behaviors log, the second History Order log, businessman's location distribution, order mode vivo identification, connect
Single game scape, time of received orders distribution, history order mode and history lose single reason distribution as the second order related data, that is, work as
Preceding order related data.Current order related data is encoded to the input data of order recommended models.
Current businessman lose single rate it is higher when, can star businessman's order method, for current businessman recommend order
Mode at the appointed time starts businessman's order method to recommend order mode for current businessman alternatively, can also be.
Step 150, according to the input data, a variety of order sides are predicted by the order recommended models that the training is completed
The recommendation probability of formula.
The input data is inputted the order recommended models that training is completed to be obtained according to the output of order recommended models
The recommendation probability of a variety of order modes.
Step 160, according to the recommendation probability, at least one order mode is chosen from a variety of order modes, and
Determine the priority of at least one order mode.
One or more order modes are chosen according to the corresponding recommendation probability of each order mode, and true according to recommendation probability
Surely the corresponding priority of each order mode chosen.When choosing order mode, can choose recommend probability be greater than or
Equal to the order mode of predetermined probabilities, alternatively, the order mode for recommending the biggish preset quantity of probability can also be chosen, it can
According to the sequence of probability from big to small is recommended, a variety of order modes are ranked up, the forward preset quantity that sorts is chosen
Order mode, as at least one order mode.At least one order mode is determined according to recommendation probability size
Priority, i.e. the corresponding highest priority of order mode of maximum probability, according to the reduction of probability, priority is successively reduced.
Wherein, at least one order mode includes the order mode that the businessman is not configured, and this order mode is in similar businessman
Middle order success rate is relatively high.
Step 170, order mode is recommended to combine at least one order mode and corresponding priority composition.
According to the corresponding priority of at least one order mode, by least one order mode and corresponding excellent
First grade composition recommends order mode to combine.
Step 180, the recommendation order mode combined recommendation is given to the current businessman.
Order mode combined recommendation will be recommended to the current businessman, current businessman, which can choose, recommends order mode to combine
In the order mode of highest priority carry out order, can also according to their needs from recommend order mode combine in choose it is suitable
The order mode of conjunction carries out order.
Businessman's order method disclosed in the embodiment of the present application, by obtain current businessman the first order related data and
The order related data of similar businessman as data sample, and obtains the corresponding order mode label of each data sample, to institute
It states data sample to be encoded, obtains the sample input data of order recommended models, according to the sample input data and correspondence
Order mode label, order recommended models are trained, so that the output of order recommended models is a variety of order modes
Recommend probability, obtains the order recommended models of training completion, obtain the second order related data of current businessman, as currently connecing
Simple correlation data, and the current order related data is encoded, the input data of order recommended models is obtained, according to institute
Input data is stated, the recommendation probability of a variety of order modes is predicted by the order recommended models that the training is completed, according to described
Recommend probability to choose at least one order mode from a variety of order modes, and determines at least one order mode
At least one order mode and corresponding priority composition are recommended order mode to combine by priority, and by the recommendation
Order mode combined recommendation gives the current businessman, since order recommended models are the dependency numbers to current businessman and similar businessman
According to what is learnt, the recommendation order mode of the current businessman determined by order recommended models is combined including each order
Mode and corresponding priority, thus when the order mode in current businessman is combined using recommendation order mode carries out order,
Order success rate can be improved.
Based on the above technical solution, by the recommendation order mode combined recommendation give the current businessman it
Afterwards, can also include:
If the current order mode of the current businessman is that the recommendation order mode one of combines and loses single rate and reaches
Single rate is lost to default, then obtains and loses single reason distribution;
Single reason distribution is lost according to described, the order recommended models are optimized.
If the current order mode of current businessman be that recommendation order mode one of combines that (such as highest priority connects
Folk prescription formula), and it is still higher to lose single rate, reaches default and loses single rate, then obtain lose every time it is single lose single reason distribution, and according to
Lose single reason distribution and and lose other order related datas of simple correlation, adjustment is optimized to order recommended models so that
The recommendation results of order recommended models are more reasonable, to further increase order success rate.
Based on the above technical solution, by the recommendation order mode combined recommendation give the current businessman it
Afterwards, can also include:
Obtain the current state of current the order mode and order equipment of the current businessman;
According to the current order mode, the recommendation order mode combines and the current state, loses determining to exist
When single risk, risk prompting and settling mode are pushed to the current businessman.
Wherein, the current state of the order equipment may include the current shape of the corresponding order equipment of current order mode
State can also include the current state of other order equipment.
According to current order mode, recommendation order mode combines and the current state of order equipment, it is determined whether exists and loses
Single risk determines settling mode, and risk is reminded and is pushed to current businessman with settling mode when in the presence of single risk is lost, real
Intelligent evaluation businessman is showed and has lost single risk, businessman can have been prompted in time when existing and losing single risk, so as to further increase
Order success rate.
For example, determining to exist and losing single wind when current order mode is not the order mode during recommendation order mode combines
Danger, determine settling mode be by current order mode be switched to recommend order mode combine in order mode, prompt current quotient
Family exist loses single risk, and can prompt current businessman by current order mode be switched to recommend order mode combine in order
Mode;If current order mode is the automatic order of mobile phone, the current state of mobile phone is then to lose list not by bluetooth connection printer
Probability is higher, it is determined that exists and loses single risk, current businessman can be prompted to exist and lose single risk, settling mode can be prompt and work as
Preceding businessman connect printer be either switched to recommend order mode combine in other order modes.
Based on the above technical solution, by the recommendation order mode combined recommendation give the current businessman it
Afterwards, can also include:
When detecting that the current businessman loses single, obtains corresponding single reason of losing and be distributed, and obtain the current businessman
Current order mode and order equipment current state;Single reason distribution, the current order mode, described is lost according to described
Current state and the recommendation order mode combine, and single settling mode is lost in determination;By it is described lose single settling mode be sent to it is described
Current businessman.
Wherein, losing single reason distribution includes order time-out is cancelled, user cancels etc..The current state of the order equipment can
Current state to include corresponding order equipment in a manner of current order can also include the current state of other order equipment.
In the prior art, after losing single problem, businessman's active feedback is needed, causes not can solve for a long time.This Shen
Please embodiment the order status of businessman can be detected, when detecting that businessman loses single, can be directly obtained corresponding
Single reason distribution is lost, and obtains the current state of current the order mode and order equipment of businessman, loses single solution in order to provide
Mode.
In the embodiment of the present application, if current order mode is not the order mode during recommendation order mode combines, Er Qietong
The corresponding recommendation probability of current order mode for crossing the output of order recommended models is lower, it is determined that current order mode loses single probability
It is higher, hence, it can be determined that lose single settling mode be by currently lose folk prescription formula be switched to recommend order mode combine in order
Mode.
If losing single reason is distributed as the cancellation of order time-out, current order mode is voice order, and order equipment is current
State is to be not switched on the tinkle of bells prompting, it is determined that the tinkle of bells can be opened for prompt businessman by losing single settling mode.
It loses single settling mode by described and is timely transmitted to the current businessman, in order to which the current businessman loses according to described
Single settling mode solve the problems, such as to lose in time it is single, reduction lose single risk, further increase order success rate.
Based on the above technical solution, by the recommendation order mode combined recommendation give the current businessman it
Afterwards, can also include:
Obtain the current state of current the order mode and order equipment of the current businessman;In the current state and institute
When stating current order mode and mismatching, is combined according to the current state and the recommendation order mode, determine best order side
Formula, and the current order mode is switched to the best order mode.
Wherein, the current state of the order equipment may include the current shape of the corresponding order equipment of current order mode
State can also include the current state of other order equipment.
It, can be with when the current state of the corresponding current order equipment of current order mode and current order mode mismatch
According to the current state of current order equipment, the current state of other order equipment and order mode is recommended to combine, determined best
Order mode.For example, determine recommend order mode combine in other order modes and the current shape of corresponding order equipment
When state matches, it can determine that best order mode is and the matched order mode of current state;Alternatively, current order equipment is worked as
Preceding state is mismatched with current order mode, and is matched with other order modes of current order equipment, and other order sides
When formula is also the order mode during recommendation order mode combines, it can determine that other order modes are best order mode, it is real
Show the most suitable order mode of state auto-switching according to order equipment, can be further improved order success rate.
For example, current order mode is the automatic order of mobile phone, if the current state of mobile phone is to beat not over bluetooth connection
Print machine, then the current state of mobile phone is mismatched with current order mode, at this moment can be connect according to the current state and recommendation of mobile phone
The combination of folk prescription formula determines that best order mode is voice order, then automatic order is switched to voice order.
Embodiment two
A kind of businessman's order device disclosed in the present embodiment, as shown in Fig. 2, businessman's order device 200 includes:
Sample acquisition module 210, for obtaining the first order related data and the order of similar businessman of current businessman
Related data as data sample, and obtains the corresponding order mode label of each data sample, the order mode label packet
Order mode is included, further includes order success rate or loses single rate;
Encoding samples module 220 obtains the sample input of order recommended models for encoding to the data sample
Data;
Model training module 230, for being pushed away to order according to the sample input data and corresponding order mode label
It recommends model to be trained, so that the output of order recommended models is the recommendation probability of a variety of order modes, obtains training completion
Order recommended models;
Data acquisition module 240, for obtaining the second order related data of current businessman, as current order dependency number
According to, and the current order related data is encoded, obtain the input data of order recommended models;
Recommend probabilistic forecasting module 250, for recommending mould by the order that the training is completed according to the input data
Type predicts the recommendation probability of a variety of order modes;
Order mode chooses module 260, for being chosen at least from a variety of order modes according to the recommendation probability
A kind of order mode, and determine the priority of at least one order mode;
Recommend order mode composite module 270, for forming at least one order mode and corresponding priority
Order mode is recommended to combine;
Order recommending module 280, for giving the recommendation order mode combined recommendation to the current businessman.
Optionally, the data sample and the current order related data respectively include: User action log, history are ordered
Odd-numbered day will, businessman's location distribution, order mode vivo identification, order scene, time of received orders distribution, history order mode with
And history loses at least one of single reason distribution.
Optionally, described device further include:
Current order data acquisition module, for obtaining the current order mode of the current businessman and working as order equipment
Preceding state;
Risk reminding module, for being combined according to the current order mode, the recommendation order mode and described current
State pushes risk prompting and settling mode to the current businessman when determining in the presence of single risk is lost.
Optionally, described device further include:
It loses forms data and obtains module, divide for when detecting that the current businessman loses single, obtaining corresponding single reason of losing
Cloth, and obtain the current state of current the order mode and order equipment of the current businessman;
Settling mode determining module, for losing single reason distribution, the current order mode, the current shape according to
State and the recommendation order mode combine, and single settling mode is lost in determination;
Settling mode sending module, for losing single settling mode by described and being sent to the current businessman.
Optionally, described device further include:
Current order data acquisition module, for obtaining the current order mode of the current businessman and working as order equipment
Preceding state;
Order mode switching module, for when the current state and the current order mode mismatch, according to institute
It states current state and the recommendation order mode combines, determine best order mode, and the current order mode is switched to
The best order mode.
Optionally, described device further include:
It loses single reason and obtains module, if the current order mode for the current businessman is the recommendation order mode group
It one of closes and loses single rate and reach default and lose single rate, then obtain and lose single reason distribution;
Model optimization module optimizes the order recommended models for losing single reason distribution according to.
Businessman's order device provided by the embodiments of the present application, for realizing businessman order side described in the embodiment of the present application
Each step of method, the specific embodiment of each module of device is referring to corresponding steps, and details are not described herein again.
Businessman's order device disclosed in the embodiment of the present application, the order for obtaining current businessman by sample acquisition module are related
Data and the order related data of similar businessman as data sample, and obtain the corresponding order mode of each data sample
Label, encoding samples module encode the data sample, obtain the sample input data of order recommended models, model instruction
Practice module according to the sample input data and corresponding order mode label, order recommended models is trained, so that connecing
The output of single recommended models is the recommendation probability of a variety of order modes, obtains the order recommended models of training completion, data acquisition
Module obtains the second order related data of current businessman, as current order related data, and it is related to the current order
Data are encoded, and the input data of order recommended models is obtained, and are recommended probabilistic forecasting module according to the input data, are passed through
The order recommended models that the training is completed predict the recommendation probability of a variety of order modes, and order mode chooses module according to
Recommend probability to choose at least one order mode from a variety of order modes, and determines at least one order mode
Priority recommends order mode composite module that at least one order mode and corresponding priority composition are recommended order side
Formula combination, order recommending module gives the recommendation order mode combined recommendation to the current businessman, due to order recommended models
It is to be learnt to the related data of current businessman and similar businessman, the current businessman determined by order recommended models
Recommendation order mode combine including each order mode and corresponding priority, thus current businessman using recommend order side
When order mode in formula combination carries out order, order success rate can be improved.
Correspondingly, the embodiment of the present application also discloses a kind of electronic equipment, including memory, processor and it is stored in described
On memory and the computer program that can run on a processor, the processor are realized when executing the computer program such as this
Apply for businessman's order method described in embodiment.The electronic equipment can be server, PC machine, mobile terminal, individual digital
Assistant, tablet computer, server etc..
The embodiment of the present application also discloses a kind of computer readable storage medium, is stored thereon with computer program, the journey
The step of businessman's order method as described in the embodiment of the present application is realized when sequence is executed by processor.
All the embodiments in this specification are described in a progressive manner, the highlights of each of the examples are with
The difference of other embodiments, the same or similar parts between the embodiments can be referred to each other.For Installation practice
For, since it is basically similar to the method embodiment, so being described relatively simple, referring to the portion of embodiment of the method in place of correlation
It defends oneself bright.
A kind of businessman's order method, apparatus provided by the embodiments of the present application, electronic equipment and storage medium are carried out above
It is discussed in detail, specific examples are used herein to illustrate the principle and implementation manner of the present application, above embodiments
Explanation be merely used to help understand the present processes and its core concept;At the same time, for those skilled in the art,
According to the thought of the application, there will be changes in the specific implementation manner and application range, in conclusion in this specification
Hold the limitation that should not be construed as to the application.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware realization.Based on such reason
Solution, substantially the part that contributes to existing technology can embody above-mentioned technical proposal in the form of software products in other words
Come, which may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including
Some instructions are used so that a computer equipment (can be personal computer, server or the network equipment etc.) executes respectively
Method described in certain parts of a embodiment or embodiment.
Claims (10)
1. a kind of businessman's order method characterized by comprising
The first order related data of current businessman and the order related data of similar businessman are obtained, as data sample, and
Obtain the corresponding order mode label of each data sample, the order mode label includes order mode, further include order at
Power loses single rate;
The data sample is encoded, the sample input data of order recommended models is obtained;
According to the sample input data and corresponding order mode label, order recommended models are trained, so that order
The output of recommended models is the recommendation probability of a variety of order modes, obtains the order recommended models of training completion;
The the second order related data for obtaining current businessman, as current order related data, and it is related to the current order
Data are encoded, and the input data of order recommended models is obtained;
According to the input data, predict that the recommendation of a variety of order modes is general by the order recommended models that the training is completed
Rate;
According to the recommendation probability, at least one order mode is chosen from a variety of order modes, and determination is described at least
A kind of priority of order mode;
Order mode is recommended to combine at least one order mode and corresponding priority composition;
Give the recommendation order mode combined recommendation to the current businessman.
2. the method according to claim 1, wherein the data sample and the current order related data point
Do not include: User action log, History Order log, businessman's location distribution, order mode vivo identification, order scene,
Time of received orders distribution, history order mode and history lose at least one of single reason distribution.
3. the method according to claim 1, wherein the recommendation order mode combined recommendation is worked as to described
After preceding businessman, further includes:
Obtain the current state of current the order mode and order equipment of the current businessman;
According to the current order mode, the recommendation order mode combines and the current state, loses single wind determining to exist
When dangerous, risk prompting and settling mode are pushed to the current businessman.
4. the method according to claim 1, wherein the recommendation order mode combined recommendation is worked as to described
After preceding businessman, further includes:
When detecting that the current businessman loses single, obtains corresponding single reason of losing and be distributed, and obtain working as the current businessman
The current state of preceding order mode and order equipment;
It is combined according to single reason distribution, the current order mode, the current state and the recommendation order mode of losing,
Single settling mode is lost in determination;
It loses single settling mode by described and is sent to the current businessman.
5. the method according to claim 1, wherein the recommendation order mode combined recommendation is worked as to described
After preceding businessman, further includes:
Obtain the current state of current the order mode and order equipment of the current businessman;
When the current state and the current order mode mismatch, according to the current state and the recommendation order side
Formula combination, determines best order mode, and the current order mode is switched to the best order mode.
6. the method according to claim 1, wherein the recommendation order mode combined recommendation is worked as to described
After preceding businessman, further includes:
If the current order mode of the current businessman is that the recommendation order mode one of combines and loses single rate and reaches pre-
If losing single rate, then obtains and lose single reason distribution;
Single reason distribution is lost according to described, the order recommended models are optimized.
7. a kind of businessman's order device characterized by comprising
Sample acquisition module, for obtaining the first order related data of current businessman and the order dependency number of similar businessman
According to as data sample, and obtaining the corresponding order mode label of each data sample, the order mode label includes order
Mode further includes order success rate or loses single rate;
Encoding samples module obtains the sample input data of order recommended models for encoding to the data sample;
Model training module is used for according to the sample input data and corresponding order mode label, to order recommended models
It is trained, so that the output of order recommended models is the recommendation probability of a variety of order modes, the order for obtaining training completion is pushed away
Recommend model;
Data acquisition module, for obtaining the second order related data of current businessman, as current order related data, and it is right
The current order related data is encoded, and the input data of order recommended models is obtained;
Recommend probabilistic forecasting module, for being predicted by the order recommended models that the training is completed according to the input data
The recommendation probability of a variety of order modes;
Order mode chooses module, for choosing at least one from a variety of order modes and connecing according to the recommendation probability
Folk prescription formula, and determine the priority of at least one order mode;
Recommend order mode composite module, at least one order mode and corresponding priority composition to be recommended order
Mode combines;
Order recommending module, for giving the recommendation order mode combined recommendation to the current businessman.
8. device according to claim 7, which is characterized in that the data sample and the current order related data point
Do not include: User action log, History Order log, businessman's location distribution, order mode vivo identification, order scene,
Time of received orders distribution, history order mode and history lose at least one of single reason distribution.
9. a kind of electronic equipment, including memory, processor and it is stored on the memory and can runs on a processor
Computer program, which is characterized in that the processor realizes claim 1 to 6 any one when executing the computer program
Businessman's order method.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
The step of businessman's order method described in claim 1 to 6 any one is realized when execution.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910671992.3A CN110503242A (en) | 2019-07-24 | 2019-07-24 | Businessman's order method, apparatus, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910671992.3A CN110503242A (en) | 2019-07-24 | 2019-07-24 | Businessman's order method, apparatus, electronic equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110503242A true CN110503242A (en) | 2019-11-26 |
Family
ID=68586741
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910671992.3A Pending CN110503242A (en) | 2019-07-24 | 2019-07-24 | Businessman's order method, apparatus, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110503242A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112613813A (en) * | 2020-12-22 | 2021-04-06 | 天津五八到家货运服务有限公司 | Method, equipment and medium for pushing freight work scheme |
CN113793165A (en) * | 2021-01-15 | 2021-12-14 | 北京京东拓先科技有限公司 | Order receiving response time length output method and device, electronic equipment and computer medium |
-
2019
- 2019-07-24 CN CN201910671992.3A patent/CN110503242A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112613813A (en) * | 2020-12-22 | 2021-04-06 | 天津五八到家货运服务有限公司 | Method, equipment and medium for pushing freight work scheme |
CN113793165A (en) * | 2021-01-15 | 2021-12-14 | 北京京东拓先科技有限公司 | Order receiving response time length output method and device, electronic equipment and computer medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110138982A (en) | Service based on artificial intelligence is realized | |
Bennett et al. | Why global scenarios need ecology | |
US20170357910A1 (en) | System for iteratively training an artificial intelligence using cloud-based metrics | |
CN103473831B (en) | For using the system of neural network positioned vehicle key, method and product | |
Potts et al. | International trade, eco-labelling, and sustainable fisheries–recent issues, concepts and practices | |
CN111861569A (en) | Product information recommendation method and device | |
CN106097043B (en) | The processing method and server of a kind of credit data | |
CN103971162A (en) | Method for improving BP (back propagation) neutral network and based on genetic algorithm | |
CN110503242A (en) | Businessman's order method, apparatus, electronic equipment and storage medium | |
CN112346936A (en) | Application fault root cause positioning method and system | |
CN108090218A (en) | Conversational system generation method and device based on deeply study | |
US11836590B2 (en) | User intent classification using a multi-agent reinforcement learning framework | |
CN108595493A (en) | Method for pushing and device, storage medium, the electronic device of media content | |
CN110288878B (en) | Self-adaptive learning method and device | |
CN102413378A (en) | Adaptive neural network-based lost packet recovery method in video transmission | |
CN107506877A (en) | OTA platforms are to Forecasting Methodology and system of the shop without room | |
CN108959388A (en) | information generating method and device | |
CN107977676A (en) | Text similarity computing method and device | |
US11829920B2 (en) | Intelligent prediction systems and methods for conversational outcome modeling frameworks for sales predictions | |
CN113094284A (en) | Application fault detection method and device | |
CN110727771A (en) | Information processing method and device, electronic equipment and readable storage medium | |
CN107665362B (en) | Training method, the method and device of prediction answer for realizing robot chat | |
CN111046156B (en) | Method, device and server for determining rewarding data | |
CN116611633A (en) | Vehicle production scheduling method and device and electronic equipment | |
CN109636702A (en) | Information management system, method, medium and electronic equipment |
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 |