CN113762621A - Network taxi appointment driver departure prediction method and system - Google Patents
Network taxi appointment driver departure prediction method and system Download PDFInfo
- Publication number
- CN113762621A CN113762621A CN202111056641.5A CN202111056641A CN113762621A CN 113762621 A CN113762621 A CN 113762621A CN 202111056641 A CN202111056641 A CN 202111056641A CN 113762621 A CN113762621 A CN 113762621A
- Authority
- CN
- China
- Prior art keywords
- type
- characteristic information
- driver
- information
- car booking
- 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
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000012549 training Methods 0.000 claims abstract description 38
- 230000008569 process Effects 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims description 27
- 238000004458 analytical method Methods 0.000 claims description 17
- 238000011156 evaluation Methods 0.000 claims description 9
- 230000006399 behavior Effects 0.000 claims description 7
- 238000007689 inspection Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 5
- 230000000875 corresponding effect Effects 0.000 description 36
- 238000012353 t test Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 10
- 238000004590 computer program Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000000692 Student's t-test Methods 0.000 description 4
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000001755 vocal effect Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013398 bayesian method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000003867 tiredness Effects 0.000 description 1
- 208000016255 tiredness Diseases 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- 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
- G06Q10/10—Office automation; Time management
- G06Q10/105—Human resources
- G06Q10/1053—Employment or hiring
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a system for predicting the leaving of a driver in a network car appointment, which relate to the technical field of leaving prediction and comprise the following steps: acquiring various characteristic information of a target network car booking driver, wherein the characteristic information comprises personal characteristic information and/or working characteristic information; determining the leaving probability of a driver of the car appointment on the target network by using the multiple kinds of characteristic information as input through a leaving model; the leaving model is obtained by training in the following way: acquiring various characteristic information of a plurality of network car booking drivers and the leaving situation of each network car booking driver; and removing the characteristic information of a target type from the multiple characteristic information of the plurality of network car booking drivers, wherein the target type is the type of the characteristic information of which the capability of distinguishing whether the network car booking drivers leave the office is smaller than the preset capability, and training a basic model to obtain a leaving office model. According to the embodiment of the invention, the feature information of the type with weaker distinguishing capability is removed in the training process, and the model is trained by adopting the removed feature information, so that the training speed is increased.
Description
Technical Field
The invention relates to the technical field of departure prediction, in particular to a method and a system for predicting departure of a net car booking driver.
Background
With the continuous development of the mobile internet, the network car booking industry is scaled, so that a platform for network car booking is formed. For drivers, the platform creates employment opportunities for the drivers and increases income. Similarly, the driver is a core asset of the online taxi appointment platform, and the driver's hard work brings stable single quantity and running water to the online taxi appointment platform. The high frequency of driver often can influence the efficiency of meeting an order of whole platform, to the platform that is main from camping, the influence is particularly great, for example the vehicle is idle, the transport capacity is extravagant etc.. The prediction of departure from the net appointment driver is also a key topic.
The existing job leaving prediction is used for collecting information training job leaving models of a plurality of users during model training, however, some information does not have a distinguishing function for the fact that one user leaves or does not leave the job, so that the data size of the training job leaving models is large, and the training speed is low.
Disclosure of Invention
The invention provides a method and a system for predicting the leaving of a net car booking driver, which can eliminate the characteristic information of the type of the characteristic information of which the capability of distinguishing whether the net car booking driver leaves the car or not is smaller than the preset capability in the training process, and train a leaving model by adopting the eliminated characteristic information, thereby improving the training speed.
In a first aspect, an embodiment of the present invention provides a method for predicting a departure of a vehicle booking driver, including:
acquiring various feature information of a target network car booking driver, wherein the feature information comprises personal feature information and/or working feature information;
determining the leaving probability of the target network car appointment driver by using the various kinds of characteristic information as input through a leaving model;
wherein, the leaving model is obtained by training in the following way:
acquiring various characteristic information of a plurality of network car booking drivers and the leaving situation of each network car booking driver;
removing the characteristic information of the target type from a plurality of kinds of characteristic information of a plurality of network car booking drivers; the target type is a type of characteristic information, wherein the capability of distinguishing whether a vehicle booking driver leaves a job or not is smaller than the preset capability;
and training a basic model according to the removed various feature information and the job leaving condition of each network car booking driver to obtain a job leaving model.
According to the method, when the departure model is trained, the type of the characteristic information for distinguishing whether the departure ability of the car booking driver is smaller than the preset ability is found, the characteristic information of the type is removed, the characteristic information without the distinguishing ability is adopted, and the departure model is trained, so that the training speed can be increased.
In one possible embodiment, the target type is determined by some or all of the following feature analysis methods;
determining the variance of each type of feature information according to a plurality of kinds of feature information of a plurality of networked car booking drivers, and determining the type as a target type if the variance of the type of feature information does not exceed a variance preset threshold value aiming at the variance of each type of feature information;
determining a correlation coefficient of each type of characteristic information according to a plurality of kinds of characteristic information of a plurality of network car booking drivers, and determining the type as a target type if the correlation coefficient of the type of characteristic information does not exceed a preset threshold value of the correlation coefficient aiming at the correlation coefficient of each type of characteristic information;
determining the difference degree of each type of characteristic information of a driver leaving the network car appointment and a driver not leaving the network car appointment through t inspection according to various characteristic information of a plurality of network car appointment drivers, and determining the type as a target type according to the difference degree of each type of characteristic information if the difference degree of the type of characteristic information exceeds a preset degree.
According to the method, when the quit model is trained, the feature information of the type without the distinguishing capability is determined through the variance, the correlation coefficient and a partial or whole mode of t test, and the quit model is trained by adopting the feature information after the feature information of the type without the distinguishing capability is removed, so that the training speed can be improved.
In one possible implementation, after determining the probability of leaving of the driver of the target network car appointment by the leaving model using a plurality of kinds of the characteristic information as input, the method further includes:
and determining a preset retrieval strategy corresponding to the probability range to which the leaving probability of the target network car booking driver belongs according to the corresponding relation between the probability range and the preset retrieval strategy, and displaying the preset retrieval strategy corresponding to the probability range to which the leaving probability of the target network car booking driver belongs.
According to the method, the preset retrieval strategy can be obtained and displayed according to the determined level of the leaving probability of the target network car appointment driver by the leaving model, so that platform management personnel can timely deal with the situation.
In one possible implementation, various characteristic information of the driver of the target network car appointment is acquired, including:
determining various personal characteristic information of a target network car booking driver according to the registration information of the target network car booking driver; and/or
Determining various kinds of working characteristic information of a target network car booking driver according to the information of the working process of the target network car booking driver; wherein, the information of the working process comprises the following parts or all: attendance information, income information, order information, passenger evaluation information, platform reward and punishment information, driving behavior information and driver voice information.
According to the method, the data of the network car booking driver is provided to contain the personal characteristic information of the target network car booking driver, and part or all of attendance information, income information, order information, passenger evaluation information, platform reward and punishment information, driving behavior information and driver voice information in the working process, and the working characteristic information is determined, so that the multi-dimensional characteristic information is adopted for carrying out the departure prediction, and the prediction accuracy is improved.
In one possible implementation, after obtaining the plurality of characteristic information of the target network car booking driver, the method further comprises:
if the form of recording the characteristic information is a text form and the type of the characteristic information is a first preset type, carrying out numerical processing on the characteristic information of a target network car booking driver according to a preset meaning value corresponding to each meaning; wherein the first preset type is a type of feature information expressing opposite meanings;
if the form of recording the characteristic information is a text form and the type of the characteristic information is a second preset type, carrying out numerical processing on the characteristic information of the target network car booking driver according to a preset level value corresponding to each level; wherein the second preset type is a type of feature information expressing a level meaning;
if the form of recording the characteristic information is a character form and the type of the characteristic information is a third preset type, carrying out numerical processing on the characteristic information of a target network car booking driver according to a preset violation score corresponding to a preset violation item; the third preset type is the type of characteristic information expressing the violation of the net car booking driver.
The method can carry out the numerical processing on the characteristic information in the character form, thereby balancing the problem that the character information cannot be processed with the numerical information in a unified way.
In one possible embodiment, after obtaining a plurality of characteristic information of a plurality of network car booking drivers, the method further comprises:
if the target type without distinguishing capability is determined by more than one feature analysis mode, the types without distinguishing capability determined by the feature analysis modes with the number exceeding the preset number are the target type.
According to the method, the feature information of the type without distinguishing capability can be determined to be deleted when the number of the feature analysis modes exceeds the preset number, so that the reliability of determining the feature information of the type without distinguishing capability is improved.
In a second aspect, an embodiment of the present invention provides a system for predicting a departure of a web taxi appointment driver, including:
the acquisition module is used for acquiring various characteristic information of a target network car booking driver; the characteristic information comprises personal characteristic information and/or working characteristic information;
the prediction module is used for taking various personal characteristic information and various working characteristic information as input and determining the leaving probability of the target network car appointment driver through a leaving model;
the training module is used for training the job leaving model in the following modes:
acquiring various characteristic information of a plurality of network car booking drivers and the leaving situation of each network car booking driver;
removing the characteristic information of the target type from a plurality of kinds of characteristic information of a plurality of network car booking drivers; the target type is a type of characteristic information, wherein the capability of distinguishing whether a vehicle booking driver leaves a job or not is smaller than the preset capability;
and training a basic model according to the removed various feature information and the job leaving condition of each network car booking driver to obtain a job leaving model.
In a possible implementation manner, the training module is specifically configured to:
determining the variance of each type of feature information according to a plurality of kinds of feature information of a plurality of networked car booking drivers, and determining the type as a target type if the variance of the type of feature information does not exceed a variance preset threshold value aiming at the variance of each type of feature information;
determining a correlation coefficient of each type of characteristic information according to a plurality of kinds of characteristic information of a plurality of network car booking drivers, and determining the type as a target type if the correlation coefficient of the type of characteristic information does not exceed a preset threshold value of the correlation coefficient aiming at the correlation coefficient of each type of characteristic information;
determining the difference degree of each type of characteristic information of a driver leaving the network car appointment and a driver not leaving the network car appointment through t inspection according to various characteristic information of a plurality of network car appointment drivers, and determining the type as a target type according to the difference degree of each type of characteristic information if the difference degree of the type of characteristic information exceeds a preset degree.
In one possible implementation, the system further comprises:
and the retrieval module is used for determining a preset retrieval strategy corresponding to the probability range to which the leaving probability of the target network car booking driver belongs according to the corresponding relation between the probability range and the preset retrieval strategy, and displaying the preset retrieval strategy corresponding to the probability range to which the leaving probability of the target network car booking driver belongs.
In a possible implementation manner, the obtaining module is specifically configured to:
determining various personal characteristic information of a target network car booking driver according to the registration information of the target network car booking driver; and/or
Determining various kinds of working characteristic information of a target network car booking driver according to the information of the working process of the target network car booking driver; wherein, the information of the working process comprises the following parts or all: attendance information, income information, order information, passenger evaluation information, platform reward and punishment information, driving behavior information and driver voice information.
In one possible implementation, the system further comprises:
the data processing module is used for carrying out numerical processing on the characteristic information of the target network car booking driver according to a preset meaning value corresponding to each meaning if the form of recording the characteristic information is a character form and the type of the characteristic information is a first preset type; wherein the first preset type is a type of feature information expressing opposite meanings;
if the form of recording the characteristic information is a text form and the type of the characteristic information is a second preset type, carrying out numerical processing on the characteristic information of the target network car booking driver according to a preset level value corresponding to each level; wherein the second preset type is a type of feature information expressing a level meaning;
if the form of recording the characteristic information is a character form and the type of the characteristic information is a third preset type, carrying out numerical processing on the characteristic information of a target network car booking driver according to a preset violation score corresponding to a preset violation item; the third preset type is the type of characteristic information expressing the violation of the net car booking driver.
In a possible implementation manner, the training module is further configured to, if the target type without the distinguishing capability is determined by more than one feature analysis manner, determine the type without the distinguishing capability as the target type, where the types without the distinguishing capability are determined by all the feature analysis manners whose number exceeds a preset number.
In a third aspect, an embodiment of the present invention provides an electronic device, including: a processor and a memory for storing processor-executable instructions;
the processor is configured to execute the net appointment driver departure prediction method of any one of the first aspect.
In a fourth aspect, an embodiment of the present invention provides a storage medium, where instructions are executed by a processor of an electronic device, so that the electronic device can execute the net appointment driver departure prediction method according to any one of the first aspect.
In addition, for technical effects brought by any one implementation manner of the second aspect to the fourth aspect, reference may be made to technical effects brought by different implementation manners of the first aspect, and details are not described here.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
Drawings
Fig. 1 is a flowchart of a method for predicting a departure of a network car booking driver according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a system for predicting the departure of a network car booking driver according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail with reference to the accompanying drawings, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
At present, a large amount of data can be obtained for training in the training process of the leaving model, and the training speed is low due to the fact that the data amount is large in the training process.
Based on this, an embodiment of the present invention provides a method for predicting a departure of a network car booking driver, as shown in fig. 1, including:
s100: acquiring various characteristic information of a plurality of network car booking drivers and the leaving situation of each network car booking driver; the characteristic information comprises personal characteristic information and/or working characteristic information;
wherein each networked car appointment driver has a plurality of personal characteristic information and/or a plurality of work characteristic information.
The departure situation of each net car booking driver, such as departure or non-departure, is used as a label of the net car booking driver for subsequent training.
S101: removing the characteristic information of the target type from a plurality of kinds of characteristic information of a plurality of network car booking drivers; the target type is a type of characteristic information, wherein the capability of distinguishing whether the taxi appointment driver leaves is smaller than the preset capability;
s102: training a basic model according to the removed various feature information and the situation of leaving of each network car booking driver to obtain a leaving model;
s103: acquiring various characteristic information of a target network car booking driver;
the target network car booking driver is a network car booking driver needing to carry out departure prediction. Acquiring various personal characteristic information and various working characteristic information of the driver of the target network car appointment in the same manner as the step 100.
S104: and determining the leaving probability of the car appointment driver of the target network by using the various characteristic information as input through a leaving model.
In the scheme, the departure model can be trained through the characteristic information after the precision subtraction, and the departure probability of the target network car booking driver is predicted by adopting the departure model, so that the speed of the training model is improved, and the contribution can be made to the driver management of the network car booking platform.
Further, whether the net car booking driver or the target net car booking driver is used as a training sample, the method for acquiring the multiple kinds of characteristic information comprises the following steps:
aiming at each network car booking driver, determining the personal characteristic information of the network car booking driver according to the registration information of the network car booking driver;
determining various kinds of working characteristic information of the network car booking driver according to the information of the working process of the network car booking driver; wherein, the information of the working process comprises the following parts or all: attendance information, income information, order information, passenger evaluation information, platform reward and punishment information, driving behavior information and driver voice information.
The registration information includes age, gender, culture level, marital status, whether to engage in the online car-booking industry, the time when the vehicle engaged in the online car-booking industry, and the like.
For example: the personal characteristic information of the taxi appointment driver comprises: age: age 40; sex: male; culture level: graduating the department; marital status: marrying; whether to engage in the network-passing car booking industry: after the practice; time spent in the network appointment industry: and 5 years.
Determining various kinds of work characteristic information of the networked car booking driver according to the attendance information, income information and order information of the networked car booking driver, wherein the types of the work characteristic information comprise the attendance days, the leave-on days, the daily average online time, the daily average offline latest time, the daily average running water, the week average income and the month average income of a preset time period.
And determining various kinds of working characteristic information of the network car booking driver according to the passenger evaluation information and the platform reward and punishment information of the network car booking driver, wherein the types of the working characteristic information comprise the poor evaluation number, the poor evaluation rate, the customer complaint number, the customer complaint rate, the platform violation number, the blackened number of the passengers and the violation deduction total number in a preset time period.
And determining various kinds of working characteristic information of the network car booking driver according to the driving behavior information of the network car booking driver, wherein the types of the working characteristic information comprise the number of times of rejecting orders and the average driving receiving time.
According to the voice information of the driver of the networked car booking driver, various kinds of working characteristic information of the networked car booking driver are determined, and the type of the working characteristic information is obtained, for example, a conversation text is obtained through a voice-to-text interface. The method comprises the steps of segmenting words of a text, counting the number and the times of negative emotion words (such as anger generation, pain, regret and the like) in a conversation within nearly 7 days, and counting the number and the times of departure tendency words (such as no dryness, no earning money, too tiredness, no worth and the like) in a conversation within nearly 15 days.
However, some of the above-mentioned various types of feature information are recorded in a numerical form, for example, the number of times of words including a tendency to leave, and the age, some of the feature information are recorded in a text form, for example, the culture level, sex, marital status, whether or not to work in the internet appointment industry, and the like, and are recorded in a text form and a numerical form, which cannot be processed in a unified manner, and the feature information in a text form cannot be removed from the feature information in a feature analysis manner. Based on this, the present invention proposes a way of digitizing, specifically:
if the form of the recorded feature information is a text form and the type of the feature information is a first preset type, carrying out numerical processing on the feature information of the target network car booking driver according to a preset meaning value corresponding to each meaning; wherein, the first preset type is the type of the characteristic information expressing the opposite meaning;
if the form of the recorded characteristic information is a text form and the type of the characteristic information is a second preset type, carrying out numerical processing on the characteristic information of the target network car booking driver according to a preset level value corresponding to each level; the second preset type is the type of the characteristic information expressing the level meaning;
if the form of the recorded characteristic information is a text form and the type of the characteristic information is a third preset type, carrying out numerical processing on the characteristic information of the target network car booking driver according to a preset violation score corresponding to a preset violation item; the third preset type is the type of characteristic information expressing the violation of the net taxi booking driver.
The first preset type is a type of characteristic information expressing opposite meanings, such as gender, marital status, whether to work in the internet-of-vehicles industry, and the like. Whether the sex is male or female, the marital status is married or not, and whether the online car-booking industry is engaged in the online car-booking industry or not is determined.
The preset meaning value corresponding to the sex of the male is 1, and the preset meaning value corresponding to the sex of the female is 0; when the sex of the taxi appointment driver is male, the characteristic information is numerically processed to be 1; when the sex of the online taxi appointment driver is female, the characteristic information is numerically processed to be 0.
The marriage state is that the preset meaning value corresponding to married is 1, and the marriage state is that the preset meaning value corresponding to unmarried is 0; when the marriage state of the online car appointment driver is married, the characteristic information is processed into 1 in a numerical mode; when the marriage status of the online car reservation driver is not married, the characteristic information is digitized to 0.
The preset meaning value corresponding to whether the networked car booking industry is engaged is 1, and the preset meaning value corresponding to whether the networked car booking industry is engaged is not engaged is 0; when the online car booking driver is engaged in the online car booking industry, the characteristic information is numerically processed to be 1; when the online car booking driver does not work in the online car booking industry, the characteristic information is processed into 0 in a numerical mode.
The second preset type is the type of the characteristic information expressing the level meaning; for example: the culture level specifically comprises: this family, specialty, Master, doctor, specialty; the preset level value corresponding to the special department is 1, the preset level value corresponding to the special department is 2, the preset level value corresponding to the family is 3, the preset level value corresponding to the master is 4, and the preset level value corresponding to the doctor is 5.
For example: when the culture level of the online taxi booking driver is the subject, the characteristic information is numerically processed to be 3. When the culture level of the online car booking driver is master, the characteristic information is numerically processed to be 4.
The third preset type is the type of the characteristic information expressing the violation of the network car booking driver; for example, the predetermined violations include verbal attacks, harassment of passengers, advance billing, detour, mid-way rejection, pricing, bargaining, verbal inducement cancellation, claim for good comments, and multiple tolls.
The default violation points corresponding to various default violations are, for example, the speech attack is-8 points, the harassed passenger is-8 points, the advance charge is-8 points, the detour is-8 points, the midway passenger throwing is-8 points, the pricing negotiation price is-8 points, the speech induction cancellation is-4 points, the asking for help is rated as-4 points, and the multi-rate charge is-4 points.
The characteristic information of the net car booking driver of the type comprises: if the violation of a preset time period is specifically that the speech attacks for 3 times, carrying out numerical processing on the violation of the preset time period to obtain a value of-8 times 3 to obtain-24; the illegal items in the preset time period of the taxi appointment driver comprise 1 time of speech attack, 2 times of midway passenger throwing and 1 time of price adding and bargaining, and the numerical processing is carried out to be-8 times and 4 times to be-32 minutes.
Wherein the target type is determined by the following characteristic analysis mode;
mode 1: determining the variance of each type of feature information according to a plurality of kinds of feature information of a plurality of networked car booking drivers, and determining the type as a target type if the variance of the type of feature information does not exceed a variance preset threshold value aiming at the variance of each type of feature information;
mode 2: determining a correlation coefficient of each type of characteristic information according to a plurality of kinds of characteristic information of a plurality of network car booking drivers, and determining the type as a target type if the correlation coefficient of the type of characteristic information does not exceed a preset threshold value of the correlation coefficient aiming at the correlation coefficient of each type of characteristic information;
mode 3: determining the difference degree of each type of characteristic information of a driver leaving the network car appointment and a driver not leaving the network car appointment through t inspection according to various characteristic information of a plurality of network car appointment drivers, and determining the type as a target type according to the difference degree of each type of characteristic information if the difference degree of the type of characteristic information exceeds a preset degree.
The above mentioned variance, correlation coefficient, t-test are explained in relation to:
wherein, the variance formula is as follows:
Var(X)＝E(X-E(X))^2
wherein X is a variable value, and E (X) is an expected value of X. Through the variance formula, the variance of each type of characteristic information can be determined, for example, the age, the marital status, the academic history and the like are required to determine one variance for each type, and when the variance of the type is smaller than a preset threshold value of the variance, the type of all network taxi appointment drivers is determined to be the target type.
For example, the type of the characteristic information is age, an average value of ages of all network car booking drivers is calculated, and then a variance of the ages is determined according to the average value and the ages of a plurality of network car booking drivers. If the variance of the age does not exceed the variance preset threshold, the age is considered as a target type for a person to leave or not leave, if the variance of the age exceeds the variance preset threshold, the age is considered as a type with distinguishing capability for the person to leave or not leave, and the characteristic information of the type of the age is reserved.
Wherein, the correlation coefficient is Pearson correlation coefficient, and the calculation formula is as follows:
wherein, X is independent variable, Y dependent variable is not off duty, and the range of the correlation coefficient is [ -1, +1 ]. The closer the absolute value of the correlation coefficient is to 1, the more correlated the two variables are. And the correlation coefficient eliminates the influence of dimension and can reflect the relation between variables more directly.
That is, X is the feature information, Y is whether the user of the feature information leaves the job,
is the average value of the characteristic information of this type,
the average of the departure situations of the car drivers is scheduled for a plurality of nets.
For example, the type of the feature information is age, and similarly, the average value of the ages of all the net car booking drivers is calculated, and then, the average value of the leaving situations of all the net car booking drivers, that is, the net car booking drivers serving as training samples is calculated, where the average value of the leaving situations is calculated according to the leaving situation of 1 and the non-leaving situation of 0.
And then determining the Pearson correlation coefficient of the type of age according to the average value of the age, the average value of the leaving situation and the age and the leaving situation of each net appointment driver. And if the Pearson correlation coefficient exceeds the preset threshold value of the correlation coefficient, the age is considered as a target type for a person to leave or not leave, and if the Pearson correlation coefficient exceeds the preset threshold value of the correlation coefficient, the age is considered as a type with the distinguishing capability for the person to leave or not leave.
Wherein, t test, deducing the probability of difference by using t distribution theory, thereby comparing whether the difference of the two averages is significant. Examples are: record the age of the driver as X1Age of driver not out of work is recorded as X2Then, the following statistics can be statistically constructed to determine whether there is a significant difference between the ages of the two populations:
where n1, n2 represent the number of two population samples, and S1 and S2 represent the variance of the two population samples. the t-test may return a P-value probability, range [0,1 ]. Similarly, the features below the threshold are retained based on a preset P value threshold.
That is to say the t-test is determined for whether there is a difference between the two populations,
is the average of the ages of the out-of-position drivers,
is the average of the age of the driver who is out of position. And (3) detecting the difference degree of the car booking driver of the off-duty network and the car booking driver of the non-off-duty network, deleting all the characteristic information of the type of the age when the difference degree exceeds the preset degree, and keeping all the characteristic information of all the car booking drivers of the type of the age when the difference degree does not exceed the preset degree.
Wherein the variance, or correlation coefficient, or t-test can be used alone to determine the target type; the target type can also be determined by adopting the variance and the correlation coefficient; or determining the type of the target by adopting variance and t test; or determining the type of the target by adopting a correlation coefficient and t test; or determining the target type by using variance, correlation coefficient and t test.
The invention provides that when the target type is determined in multiple modes, as long as one mode is determined as the target type, the characteristic information of the type can be removed, for example, the target type is determined by using variance and t test, and when one target type is determined by using variance and the other target type is determined by using t test, both the two target types can be removed.
Of course, in order to improve the reliability of the determined target type, when the target type is determined by a plurality of feature analysis methods, the embodiment of the present invention further provides a method:
if the target type without distinguishing capability is determined by more than one feature analysis mode, the types without distinguishing capability determined by the feature analysis modes with the number exceeding the preset number are the target type.
For example, whether each piece of feature information should be deleted is determined by the above three methods, and the deletion is 1, and the non-deletion is 0. The specific statistics are shown in table 1:
TABLE 1
If the preset number is 2, the types to be removed are as shown in table 2:
| age (age) |
| Cultural level |
| Marital status |
| Time in network appointment trade |
| Nearly 7 balance receiving driving time |
| Nearly 30 balance receiving driving time |
If the preset number is 1, the type to be removed is as shown in table 3:
| age (age) |
| Cultural level |
| Marital status |
| Whether to engage in network car booking industry |
| Time in network appointment trade |
| Rest days of nearly 14 days |
| Rest days of nearly 30 days |
| Nearly 30 days, the late peak finishes the single flow |
| Order number evaluated in approximately 7 days |
| Number of complained work orders in nearly 7 days |
| Order number evaluated in last 30 days |
| Number of complained work orders in nearly 30 days |
The basic model can be Catboost, parameters need to be adjusted in the model, and the super-parameter adjustment of the model is carried out by adopting a Bayesian method. The specific process is as follows:
determining the quality evaluation standard of the model, and selecting an index AUC;
setting the range of each hyper-parameter, such as depth of tree: {2,10 };
setting iteration times and cross validation times;
and obtaining the hyper-parametric combination of the optimal AUC.
After the training is finished, the leaving probability of the driver of the target network car appointment can be determined, and a platform manager is informed of the leaving probability, specifically:
and determining a preset retrieval strategy corresponding to the probability range of the leaving probability of the target network car booking driver according to the corresponding relation between the probability range and the preset retrieval strategy, and displaying the preset retrieval strategy corresponding to the probability range of the leaving probability of the target network car booking driver.
For example, the preset retrieval policy may be to continue to observe for a period of time, ask face-to-face whether there is a departure idea; continuously observing for a period of time with the probability range of 0-0.5; and inquiring whether the face-to-face inquiry has the departure idea or not with the probability range of 0.51-1. When the leaving probability of the target network car booking driver is 0.6, the preset retrieval strategy can be displayed for inquiring whether a leaving idea exists or not in a face-to-face mode, and therefore when the manager of the network car booking platform knows that the target network car booking driver is likely to leave, the manager can carry out retrieval processing according to the preset retrieval strategy.
As shown in fig. 2, based on the same inventive concept of the net car booking driver departure prediction method, the present invention also provides a net car booking driver departure prediction system, which includes:
the acquisition module 200 is used for acquiring various feature information of a target network car booking driver, wherein the feature information comprises personal feature information and/or working feature information;
the prediction module 201 is used for determining the leaving probability of the target network car booking driver by using a leaving model by taking various kinds of characteristic information as input;
a training module 202, configured to train the job leaving model to obtain the job leaving model by:
acquiring various characteristic information of a plurality of network car booking drivers and the leaving situation of each network car booking driver;
removing the characteristic information of the target type from a plurality of kinds of characteristic information of a plurality of network car booking drivers; the target type is a type of characteristic information, wherein the capability of distinguishing whether a vehicle booking driver leaves a job or not is smaller than the preset capability;
training a basic model according to the removed various feature information and the situation of leaving of each network car booking driver to obtain a leaving model;
optionally, the target type is determined by the following partial or complete characteristic analysis mode;
determining the variance of each type of feature information according to a plurality of kinds of feature information of a plurality of networked car booking drivers, and determining the type as a target type if the variance of the type of feature information does not exceed a variance preset threshold value aiming at the variance of each type of feature information;
determining a correlation coefficient of each type of characteristic information according to a plurality of kinds of characteristic information of a plurality of network car booking drivers, and determining the type as a target type if the correlation coefficient of the type of characteristic information does not exceed a preset threshold value of the correlation coefficient aiming at the correlation coefficient of each type of characteristic information;
determining the difference degree of each type of characteristic information of a driver leaving the network car appointment and a driver not leaving the network car appointment through t inspection according to various characteristic information of a plurality of network car appointment drivers, and determining the type as a target type according to the difference degree of each type of characteristic information if the difference degree of the type of characteristic information exceeds a preset degree.
Optionally, the system further includes:
and the retrieval module is used for determining a preset retrieval strategy corresponding to the probability range to which the leaving probability of the target network car booking driver belongs according to the corresponding relation between the probability range and the preset retrieval strategy, and displaying the preset retrieval strategy corresponding to the probability range to which the leaving probability of the target network car booking driver belongs.
Optionally, the obtaining module 200 is specifically configured to:
determining various personal characteristic information of a target network car booking driver according to the registration information of the target network car booking driver; and/or
Determining various kinds of working characteristic information of a target network car booking driver according to the information of the working process of the target network car booking driver; wherein, the information of the working process comprises the following parts or all: attendance information, income information, order information, passenger evaluation information, platform reward and punishment information, driving behavior information and driver voice information.
Optionally, the system further includes:
the data processing module is used for carrying out numerical processing on the characteristic information of the target network car booking driver according to a preset meaning value corresponding to each meaning if the form of recording the characteristic information is a character form and the type of the characteristic information is a first preset type; wherein the first preset type is a type of feature information expressing opposite meanings;
if the form of recording the characteristic information is a text form and the type of the characteristic information is a second preset type, carrying out numerical processing on the characteristic information of the target network car booking driver according to a preset level value corresponding to each level; wherein the second preset type is a type of feature information expressing a level meaning;
if the form of recording the characteristic information is a character form and the type of the characteristic information is a third preset type, carrying out numerical processing on the characteristic information of a target network car booking driver according to a preset violation score corresponding to a preset violation item; the third preset type is the type of characteristic information expressing the violation of the net car booking driver.
Optionally, the training module is further configured to, if the target type without distinguishing capability is determined through more than one feature analysis mode, determine the type without distinguishing capability that is determined by all the feature analysis modes of which the number exceeds the preset number as the target type.
In addition, the net appointment prediction method and apparatus described in conjunction with fig. 1-2 according to the embodiment of the present invention may be implemented by an electronic device.
The electronic device includes: a processor and a memory for storing processor-executable instructions; the processor is configured to execute any one of the net appointment driver departure prediction methods.
Based on the above description, the electronic device structure of fig. 3 is exemplarily presented.
The electronic device may include a processor 310 and a memory 320 storing computer program instructions.
Specifically, the processor 310 may include a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or may be configured as one or more Integrated circuits implementing an embodiment of the present invention.
The processor 310 may implement any of the above-described methods of performing tasks by reading and executing computer program instructions stored in the memory 320.
In one example, the electronic device can also include a communication interface 330 and a bus 340. As shown in fig. 3, the processor 310, the memory 320, and the communication interface 330 are connected via a bus 340 to complete communication therebetween.
The communication interface 330 is mainly used for implementing communication between modules, apparatuses, units and/or devices in the embodiments of the present invention.
In addition, in combination with the electronic device in the above embodiments, an embodiment of the present invention may provide a storage medium, where instructions of the storage medium, when executed by a processor of the electronic device, enable the electronic device to execute the web appointment driver departure prediction method as described in any one of the above.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A method for predicting the departure of a net appointment vehicle driver is characterized by comprising the following steps:
acquiring various feature information of a target network car booking driver, wherein the feature information comprises personal feature information and/or working feature information;
determining the leaving probability of the target network car appointment driver by using the various kinds of characteristic information as input through a leaving model;
wherein, the leaving model is obtained by training in the following way:
acquiring various characteristic information of a plurality of network car booking drivers and the leaving situation of each network car booking driver;
removing the characteristic information of the target type from a plurality of kinds of characteristic information of a plurality of network car booking drivers; the target type is a type of characteristic information, wherein the capability of distinguishing whether a vehicle booking driver leaves a job or not is smaller than the preset capability;
and training a basic model according to the removed various feature information and the job leaving condition of each network car booking driver to obtain a job leaving model.
2. The net car booking driver departure prediction method according to claim 1, wherein the target type is determined by some or all of the following feature analysis methods;
determining the variance of each type of feature information according to a plurality of kinds of feature information of a plurality of networked car booking drivers, and determining the type as a target type if the variance of the type of feature information does not exceed a variance preset threshold value aiming at the variance of each type of feature information;
determining a correlation coefficient of each type of characteristic information according to a plurality of kinds of characteristic information of a plurality of network car booking drivers, and determining the type as a target type if the correlation coefficient of the type of characteristic information does not exceed a preset threshold value of the correlation coefficient aiming at the correlation coefficient of each type of characteristic information;
determining the difference degree of each type of characteristic information of a driver leaving the network car appointment and a driver not leaving the network car appointment through t inspection according to various characteristic information of a plurality of network car appointment drivers, and determining the type as a target type according to the difference degree of each type of characteristic information if the difference degree of the type of characteristic information exceeds a preset degree.
3. The net car booking driver departure prediction method according to claim 1, wherein after determining the probability of departure of the target net car booking driver through a departure model using a plurality of feature information as input, the method further comprises:
and determining a preset retrieval strategy corresponding to the probability range to which the leaving probability of the target network car booking driver belongs according to the corresponding relation between the probability range and the preset retrieval strategy, and displaying the preset retrieval strategy corresponding to the probability range to which the leaving probability of the target network car booking driver belongs.
4. The method for predicting the departure of a driver of a network car booking according to claim 1, wherein the obtaining of a plurality of characteristic information of the target driver of the network car booking comprises:
determining various personal characteristic information of a target network car booking driver according to the registration information of the target network car booking driver; and/or
Determining various kinds of working characteristic information of a target network car booking driver according to the information of the working process of the target network car booking driver; wherein, the information of the working process comprises the following parts or all: attendance information, income information, order information, passenger evaluation information, platform reward and punishment information, driving behavior information and driver voice information.
5. The web car booking driver departure prediction method according to claim 1, wherein after obtaining a plurality of characteristic information of a target web car booking driver, the method further comprises:
if the form of recording the characteristic information is a text form and the type of the characteristic information is a first preset type, carrying out numerical processing on the characteristic information of a target network car booking driver according to a preset meaning value corresponding to each meaning; wherein the first preset type is a type of feature information expressing opposite meanings;
if the form of recording the characteristic information is a text form and the type of the characteristic information is a second preset type, carrying out numerical processing on the characteristic information of the target network car booking driver according to a preset level value corresponding to each level; wherein the second preset type is a type of feature information expressing a level meaning;
if the form of recording the characteristic information is a character form and the type of the characteristic information is a third preset type, carrying out numerical processing on the characteristic information of a target network car booking driver according to a preset violation score corresponding to a preset violation item; the third preset type is the type of characteristic information expressing the violation of the net car booking driver.
6. The method for predicting the departure of a net car booking driver as claimed in any one of claims 1 to 5, wherein after obtaining a plurality of characteristic information of a plurality of net car booking drivers, the method further comprises:
if the target type without distinguishing capability is determined by more than one feature analysis mode, the types without distinguishing capability determined by the feature analysis modes with the number exceeding the preset number are the target type.
7. A net appointment driver departure prediction system, comprising:
the acquisition module is used for acquiring various characteristic information of a target network car booking driver; the characteristic information comprises personal characteristic information and/or working characteristic information;
the prediction module is used for taking various kinds of characteristic information as input and determining the leaving probability of the target network car booking driver through a leaving model;
the training module is used for training the job leaving model in the following modes:
acquiring various characteristic information of a plurality of network car booking drivers and the leaving situation of each network car booking driver;
removing the characteristic information of the target type from a plurality of kinds of characteristic information of a plurality of network car booking drivers; the target type is a type of characteristic information, wherein the capability of distinguishing whether a vehicle booking driver leaves a job or not is smaller than the preset capability;
and training a basic model according to the removed various feature information and the job leaving condition of each network car booking driver to obtain a job leaving model.
8. The net appointment driver departure prediction system of claim 7, wherein the training module is specifically configured to: determining the variance of each type of feature information according to a plurality of kinds of feature information of a plurality of networked car booking drivers, and determining the type as a target type if the variance of the type of feature information does not exceed a variance preset threshold value aiming at the variance of each type of feature information;
determining a correlation coefficient of each type of characteristic information according to a plurality of kinds of characteristic information of a plurality of network car booking drivers, and determining the type as a target type if the correlation coefficient of the type of characteristic information does not exceed a preset threshold value of the correlation coefficient aiming at the correlation coefficient of each type of characteristic information;
determining the difference degree of each type of characteristic information of a driver leaving the network car appointment and a driver not leaving the network car appointment through t inspection according to various characteristic information of a plurality of network car appointment drivers, and determining the type as a target type according to the difference degree of each type of characteristic information if the difference degree of the type of characteristic information exceeds a preset degree.
9. An electronic device, comprising: a processor and a memory for storing processor-executable instructions;
the processor is used for executing the network appointment driver departure prediction method of any one of claims 1 to 6.
10. A storage medium having instructions stored thereon that, when executed by a processor of an electronic device, enable the electronic device to perform the network appointment driver departure prediction method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111056641.5A CN113762621A (en) | 2021-09-09 | 2021-09-09 | Network taxi appointment driver departure prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111056641.5A CN113762621A (en) | 2021-09-09 | 2021-09-09 | Network taxi appointment driver departure prediction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113762621A true CN113762621A (en) | 2021-12-07 |
Family
ID=78794342
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111056641.5A Pending CN113762621A (en) | 2021-09-09 | 2021-09-09 | Network taxi appointment driver departure prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113762621A (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809188A (en) * | 2015-04-20 | 2015-07-29 | 广东工业大学 | Enterprise talent drainage data mining analysis method and device |
| CN105160498A (en) * | 2015-10-21 | 2015-12-16 | 北京普猎创新网络科技有限公司 | Personal value calculation method based on big data |
| CN106022708A (en) * | 2016-05-09 | 2016-10-12 | 陈包容 | Method for predicting employee resignation |
| CN108960528A (en) * | 2018-07-25 | 2018-12-07 | 平安科技（深圳）有限公司 | The prediction technique and relevant apparatus of labor turnover reason |
| CN109543918A (en) * | 2018-11-30 | 2019-03-29 | 平安科技（深圳）有限公司 | Data predication method, device, computer installation and computer readable storage medium |
| CN109657855A (en) * | 2018-12-14 | 2019-04-19 | 深圳壹账通智能科技有限公司 | Prediction technique, device, computer equipment and the storage medium of leaving office probability |
| WO2019218751A1 (en) * | 2018-05-16 | 2019-11-21 | 阿里巴巴集团控股有限公司 | Processing method, apparatus and device for risk prediction of insurance service |
| CN110782072A (en) * | 2019-09-29 | 2020-02-11 | 广州荔支网络技术有限公司 | Employee leave risk prediction method, device, equipment and readable storage medium |
| CN112686415A (en) * | 2020-12-29 | 2021-04-20 | 南京领行科技股份有限公司 | Method and device for monitoring network taxi appointment behaviors |
| KR20210052410A (en) * | 2020-09-02 | 2021-05-10 | 베이징 바이두 넷컴 사이언스 앤 테크놀로지 코., 엘티디. | Training method, device, equipment and storage medium of online prediction model |
-
2021
- 2021-09-09 CN CN202111056641.5A patent/CN113762621A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809188A (en) * | 2015-04-20 | 2015-07-29 | 广东工业大学 | Enterprise talent drainage data mining analysis method and device |
| CN105160498A (en) * | 2015-10-21 | 2015-12-16 | 北京普猎创新网络科技有限公司 | Personal value calculation method based on big data |
| CN106022708A (en) * | 2016-05-09 | 2016-10-12 | 陈包容 | Method for predicting employee resignation |
| WO2019218751A1 (en) * | 2018-05-16 | 2019-11-21 | 阿里巴巴集团控股有限公司 | Processing method, apparatus and device for risk prediction of insurance service |
| CN108960528A (en) * | 2018-07-25 | 2018-12-07 | 平安科技（深圳）有限公司 | The prediction technique and relevant apparatus of labor turnover reason |
| CN109543918A (en) * | 2018-11-30 | 2019-03-29 | 平安科技（深圳）有限公司 | Data predication method, device, computer installation and computer readable storage medium |
| CN109657855A (en) * | 2018-12-14 | 2019-04-19 | 深圳壹账通智能科技有限公司 | Prediction technique, device, computer equipment and the storage medium of leaving office probability |
| CN110782072A (en) * | 2019-09-29 | 2020-02-11 | 广州荔支网络技术有限公司 | Employee leave risk prediction method, device, equipment and readable storage medium |
| KR20210052410A (en) * | 2020-09-02 | 2021-05-10 | 베이징 바이두 넷컴 사이언스 앤 테크놀로지 코., 엘티디. | Training method, device, equipment and storage medium of online prediction model |
| CN112686415A (en) * | 2020-12-29 | 2021-04-20 | 南京领行科技股份有限公司 | Method and device for monitoring network taxi appointment behaviors |
Non-Patent Citations (2)
| Title |
|---|
| 杨帆: ""医院员工心理契约与离职倾向关系的实证研究"", 现代医院管理, vol. 11, no. 1, pages 66 - 68 * |
| 翁清雄 等: ""员工职业成长的结构及其对离职倾向的影响"", 工业工程与管理, no. 2009, pages 97 - 104 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116342259A (en) | Automatic user credit rating method and device, electronic equipment and medium | |
| CN106529711B (en) | User behavior prediction method and device | |
| CN111260102A (en) | User satisfaction prediction method and device, electronic equipment and storage medium | |
| WO2010045216A1 (en) | Method and system of determining and applying insurance profit scores | |
| CN110942326A (en) | User satisfaction prediction method and device and server | |
| CN111353901A (en) | Risk identification monitoring method and device and electronic equipment | |
| CN112101721B (en) | Risk assessment method and device | |
| CN113011899A (en) | Enterprise credit evaluation method, device, equipment and computer storage medium | |
| Berlemann et al. | Disposition time and the utilization of prior judicial decisions: Evidence from a civil law country | |
| CN111091401A (en) | Network appointment pricing method, system, storage medium and equipment | |
| CN112634016A (en) | Community property data-based credit granting system and method | |
| CN113762621A (en) | Network taxi appointment driver departure prediction method and system | |
| CN109191157B (en) | Vehicle use management method, server and computer readable storage medium | |
| CN111327661A (en) | Pushing method, pushing device, server and computer readable storage medium | |
| CN116596659A (en) | Enterprise intelligent credit approval method, system and medium based on big data wind control | |
| CN113421154B (en) | Credit risk assessment method and system based on control chart | |
| CN114677139A (en) | Method and device for determining loan amount, equipment, product and readable storage medium | |
| CN110580634A (en) | service recommendation method, device and storage medium based on Internet | |
| CN113807943A (en) | Multi-factor valuation method, system, medium and equipment for bad assets | |
| CN112001658A (en) | Method and device for generating vehicle insurance quotation | |
| CN112348537B (en) | Information processing method, device, electronic equipment and storage medium | |
| CN109886596B (en) | Method for improving cooperative rate of crowd sensing system based on psychological account theory | |
| CN116523627A (en) | Credit line prediction method, device, equipment, medium and product | |
| CN117689141A (en) | Case data automatic processing method, system, device and storage medium based on artificial intelligence assistance | |
| CN116228438A (en) | Vehicle mileage-based vehicle risk pricing model construction method, device and system |
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 |