CN106205288A - A kind of implementation method training robot - Google Patents
A kind of implementation method training robot Download PDFInfo
- Publication number
- CN106205288A CN106205288A CN201610844764.8A CN201610844764A CN106205288A CN 106205288 A CN106205288 A CN 106205288A CN 201610844764 A CN201610844764 A CN 201610844764A CN 106205288 A CN106205288 A CN 106205288A
- Authority
- CN
- China
- Prior art keywords
- training
- requirement
- treating
- treat
- training object
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B19/00—Teaching not covered by other main groups of this subclass
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The implementation method of the training robot that the present invention provides, the training requirement identification model for identifying training requirement is trained by machine learning, and according to training requirement identification model and treat that the attribute information of training object automatically generates training requirement, and automatically obtain, according to the training requirement generated, the training contents mated with training requirement further, and finally treat training object according to training contents and give training, solve existing enterprise or unit is carried out the link of training and is substantially dependent on being accomplished manually, thus cause training heavy, the problem that human cost is higher.Whole process, without manually determining training requirement, is formulated training scheme and arrangement is treated that training object gives training, save human cost, improve training efficiency.
Description
Technical field
The present invention relates to communication technical field, be specifically related to a kind of implementation method training robot.
Background technology
Training is a routine work content of enterprise or unit.Existing Training Management pattern, is typically all first by people
Work determines training requirement, then formulates training scheme, then arranges start-up to perform training.As can be seen here, existing enterprise or list
Position is carried out the link of training and is substantially dependent on being accomplished manually, thus causes training heavy, higher the asking of human cost
Topic.
Additionally, accurate not by manually determining the training requirement that training requirement may cause determining, training scheme is inadequate
Rationally.For this problem, the present embodiment proposes one and utilizes robot to automatically generate training requirement, training scheme and to training
The method that instruction object is trained automatically.
Summary of the invention
The invention provides a kind of implementation method training robot, carry out training solving existing enterprise or unit
Link be substantially dependent on being accomplished manually, thus cause training heavy, the problem that human cost is higher.
Massage robot cannot provide the technical problem of the massage service of personalization for client.
The implementation method of the training robot that the present invention provides, including:
Preset input entry and training requirement classification;
Gather the sample input data corresponding with input entry of training sample, and extract training sample according to sample input data
Input feature value;
Training requirement classification training grader according to input feature value and training sample, it is thus achieved that training requirement identification model;
According to treat training object with input attribute information corresponding to entry and training requirement identification model, determine wait to train right
The training requirement of elephant;
Obtain the training contents mated with the training requirement treating training object preset;
Treat training object according to training contents to give training.
Further, according to the attribute information corresponding with input entry and the training requirement identification mould treating training object
Type, determines that the training requirement treating training object includes:
Gather the attribute information corresponding with input entry treating training object;
The input feature value treating training object is extracted according to attribute information;
The input feature value treating training object is inputted training requirement identification model, it is thus achieved that treat the training requirement of training object.
Further, collection treats that the attribute information corresponding with input entry of training object includes:
Identify the identity treating training object;
According to the attribute information that the identity-acquiring treating training object is corresponding with input entry.
Further, identification treats that the identity of training object includes:
Obtain the biological characteristic treating training object;
Biological characteristic is mated with the sample biological characteristic being stored in advance in data base, thus identifies and treat training object
Identity.
Further, when biological characteristic does not matches that with the sample biological characteristic being stored in advance in data base, also wrap
Include:
Search for and treat the username information of biometric matches of training object on the internet;
The identity of training object is treated according to username information identification.
Further, include according to the attribute information that the identity-acquiring treating training object is corresponding with input entry:
According to the username information that the identities match treating training object is corresponding;
The electronic register information treating training object is obtained according to username information;
The attribute information that training object is corresponding with input entry is treated according to electronic register acquisition of information.
Further, treat training object according to training contents to give training and include:
Treating training object and send training notice, training notice at least includes training contents, training time and place for training;
According to training contents, training object is giveed training.
Further, treat after training object gives training according to training contents and also include:
Treat training object and give examination training.
Further, input entry includes:
Sex, age, education background, occupation, post, academic title, target, target completion rate, history training record bar
One or more combinations in mesh.
Further, training requirement classification includes:
Managerial skills, sales skills, communication skill, programming skill, skill in design, recruitment technical ability, planning technical ability, market development skill
One or more combinations in energy, marketing technical ability.
The method have the advantages that
The implementation method of the training robot that the present invention provides, trains the training for identifying training requirement by machine learning
Demand identification model, and according to training requirement identification model and treat that the attribute information of training object automatically generates training requirement,
And automatically obtain, according to the training requirement generated, the training contents mated with training requirement further, and final according to training contents
Treat training object to give training, solve existing enterprise or unit is carried out the link of training and is substantially dependent on the completeest
Become, thus cause training heavy, the problem that human cost is higher.Whole process, without manually determining training requirement, is formulated
Training scheme and arrangement treat that training object gives training, save human cost, improve training efficiency.
In addition to objects, features and advantages described above, the present invention also has other objects, features and advantages.Below
Will be with reference to figure, the present invention is further detailed explanation.
Accompanying drawing explanation
The accompanying drawing of the part building the application is used for providing a further understanding of the present invention, and the present invention's is schematic real
Execute example and illustrate for explaining the present invention, not building inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 is the flow chart of the implementation method of preferred embodiment of the present invention training robot;
Fig. 2 is the preferred embodiment of the present invention flow chart for the implementation method of first training robot simplifying embodiment;
Fig. 3 is the preferred embodiment of the present invention flow chart for the implementation method of second training robot simplifying embodiment.
Detailed description of the invention
Below in conjunction with accompanying drawing, embodiments of the invention are described in detail, but the present invention can be defined by the claims
Implement with the multitude of different ways covered.
With reference to Fig. 1, the preferred embodiments of the present invention provide a kind of implementation method training robot, including:
Step S101, presets input entry and training requirement classification.
Step S102, gathers the sample input data corresponding with input entry of training sample, and inputs number according to sample
According to the input feature value extracting training sample.
Step S103, trains grader according to the training requirement classification of input feature value and training sample, it is thus achieved that training
Demand identification model.
Step S104, according to the attribute information corresponding with input entry and the training requirement identification mould for the treatment of training object
Type, determines the training requirement treating training object.
Step S105, obtains the training contents mated with the training requirement treating training object preset;
Step S106, treats training object according to training contents and gives training.
The implementation method of the training robot that the present invention provides, is trained for identifying training requirement by machine learning
Training requirement identification model, and according to training requirement identification model and treat that the attribute information of training object automatically generates training need
Ask, and automatically obtain, according to the training requirement generated, the training contents mated with training requirement further, and final according to training
Content is treated training object and is giveed training, solves existing enterprise or unit is carried out the link of training and is substantially dependent on manually
Complete, thus cause training heavy, the problem that human cost is higher.Whole process, without manually determining training requirement, is made
Determine training scheme and arrangement treats that training object gives training, save human cost, improve training efficiency.
Additionally, the implementation method training robot that the present invention provides, the training requirement identification mould trained by employing
Type is treated the training requirement of trainer and is carried out Intelligent Recognition, improves the degree of accuracy that in training, training requirement obtains;And it is logical
Cross and automatically obtain the training contents mated with training requirement, it is possible to obtain scientific and reasonable training scheme, to whole training
Intellectuality realize significant.
It should be noted that the present embodiment presets input entry primarily to gather targetedly and determine training
The relevant information of instruction demand, such as in actual implementation process, can preset the sex of employee, the age, education background,
Occupation, post, academic title, target, target completion rate, history training record, the total amount of makeing mistakes of work, place team appoint
The input entries such as business completion rate, it is thus possible to gather the attribute information corresponding with these input entries targetedly, and further
According to attribute information and the training requirement identification model trained, automatically obtain the most reasonably training requirement.Additionally, this enforcement
Example indication treat training object can be individual, it is also possible to be team, with specific reference to it needs to be determined that.
Alternatively, according to treat training object with input attribute information corresponding to entry and training requirement identification model,
Determine that the training requirement treating training object includes:
Gather the attribute information corresponding with input entry treating training object;
The input feature value treating training object is extracted according to attribute information;
The input feature value treating training object is inputted training requirement identification model, it is thus achieved that treat the training requirement of training object.
The present embodiment according to treat training object with input attribute information corresponding to entry and training requirement identification model,
Determine when the training requirement of training object, first gather the attribute information corresponding with input entry treating training object, then
Extract the input feature value treating training object according to attribute information, finally the input feature value treating training object is inputted training
Instruction demand identification model, it is thus achieved that treat the training requirement of training object.Specifically, the input entry in the present embodiment and training requirement
Classification is all by user-defined, and such as input entry can be sex, age, education background, occupation, post, academic title, work
Make in target, target completion rate, history training record, the total amount of makeing mistakes of work, place team tasks completion rate entry
One or more combinations, training requirement classification can be sales skills, communication skill, skill in design, recruitment technical ability, planning skill
One or more combinations in energy, market development technical ability, marketing technical ability.Such as, training requirement classification can only include
Sales skills classification, it is also possible to include the service plan by the combination of sales skills and communication skill, or sales skills and communication
Technical ability, and the classification of skill in design combination.
The present embodiment when training demand identification model, according to gather training sample with input entry corresponding
Sample input data extract the input feature value of training sample, thus train grader, wherein, grader can be support to
Amount machine grader, Bayes classifier, maximum entropy classifiers etc..And in actual implementation process, the quantity of training sample
Should be big as far as possible, thus ensure the accuracy of identification of training requirement identification model.
Additionally, the present embodiment is after gathering the attribute information corresponding with input entry treating training object, according to attribute
Attribute information can also be normalized before the input feature value that information retrieval treats training object.Such as when get with
When occupation attribute information corresponding to attributes entries is teacher, according to normalization set in advance rule, it can be carried out normalizing
Change.Normalization rule the most set in advance can be predefined mark rule, such as, pre-define the occupation of N kind, when adopting
Collect to a certain kind in this N kind occupation to time corresponding then to gather occupation be identified, thus realize attribute information is returned
One changes.And for when getting the attribute information corresponding with other input entries, it is also possible to take the method being similar to that attribute is believed
Breath is normalized, thus establishes good data basis for improving the discrimination identifying model.
The present embodiment by training for identifying the training requirement identification model of training requirement, can not only be right according to waiting to train
The attribute information coupling of elephant is corresponding, have training requirement pointedly, and uses the training that the present embodiment trains
The accuracy rate that the training requirement identification of training object treated by demand identification model is high, and follow-up training robot can be according to knowledge
The training requirement not gone out is treated that training object provides training contents targetedly, embodies higher intelligent level.
Alternatively, collection treats that the attribute information corresponding with input entry of training object includes:
Identify the identity treating training object;
According to the attribute information that the identity-acquiring treating training object is corresponding with input entry.
The present embodiment is when gathering the attribute information corresponding with input entry, and the identity first treating training object is known
Not, then according to the attribute information that the identity-acquiring treating training object is corresponding with input entry.And the present embodiment treated by identification
The identity of training object, and the attribute information corresponding with input entry according to treating the identity-acquiring of training object, just can basis
Treating attribute information and the training requirement identification model of training object, it is thus achieved that treat the training requirement of training object, intelligence degree is high,
Consumer's Experience is good.
Alternatively, identification treats that the identity of training object includes:
Obtain the biological characteristic treating training object;
Biological characteristic is mated with the sample biological characteristic being stored in advance in data base, thus identifies and treat training object
Identity.
Specifically, first the present embodiment obtains the biological characteristic treating training object when identifying until the identity of training object,
Then the biological characteristic treating training object is mated with the sample biological characteristic being stored in advance in data base, thus identify
Treat the identity of training object.Wherein, treat that the biological characteristic of training object includes: face, fingerprint, sound, abnormal smells from the patient, iris, hand,
One or more combinations in palmmprint, signature, gait.
The data base of the intelligent training robot in the present embodiment often stores the sample biological characteristic of some users,
Such as face, fingerprint, sound, abnormal smells from the patient, iris, hand, palmmprint, signature, gait etc., therefore wait to train when training robot collects
After the biological characteristic of instruction object, often first mate in its database and whether include that this treats that the sample biology of training object is special
Levy, if the match is successful, then treat that the sample biological characteristic that training object stores in data base carries out identification to it according to this.
Such as, when collecting until the face characteristic information of training object, then use face recognition technology by itself and intelligent training machine
In the data base of people, face sample characteristics mates, thus the identity realizing treating training object is identified.
It should be noted that the identity treating training object of indication refers to characterize or uniquely characterize treat in the present embodiment
The information of training object identity, such as, treat the name of training object, telephone number, email address, ID (identity number) card No. etc..Therefore this
Embodiment is when gathering until the attribute information of training object, and the identity first treating training object is identified, follow-up
Identity according to treating training object quickly and accurately obtains the attribute information corresponding with treating training object, and according to treating training object
Attribute information obtain accurate training requirement.
Alternatively, when biological characteristic does not matches that with the sample biological characteristic being stored in advance in data base, also include:
Search for and treat the username information of biometric matches of training object on the internet;
The identity of training object is treated according to username information identification.
Due to by when the biological characteristic of training object mates with the sample biological characteristic in data base, in fact it could happen that
The situation that biological characteristic does not matches that with the sample biological characteristic being stored in advance in data base, namely can not be from being stored in advance in
Sample biological characteristic in data base matches the biological characteristic identical with treating training object.For this situation, this enforcement
Example passes through to search on the internet and treat the username information of the biometric matches of training object, then according to username information
Identify the identity treating training object.
Such as, can not match when system from the face sample biological characteristic being stored in advance in data base and wait to train
During the identical face biological characteristic of object, then according to the face biological characteristic treating training object gathered search on the internet with
The username information of correspondence.This is mainly based upon existing user when carrying out identity registration on the internet, is often required
Corresponding biological characteristic, such as face, signature etc. are provided.The present embodiment is based on the dependency of both, first passes through and treats
The biological characteristic of training object is searched on the internet and is obtained the username information matched, and then knows according to username information
Do not treat the identity of training object.
It should be noted that the present embodiment is the most right according to the biological characteristic search acquisition treating training object on the internet
During the username information answered, more than one username information may be obtained.Now, the present embodiment can choose wherein any one
The identity of training object is treated in username information identification, it is also possible to chooses and treats the use that the biometric matches degree of training object is the highest
Name in an account book information treats the identity of training object for identifying.
Alternatively, include according to the attribute information that the identity-acquiring treating training object is corresponding with input entry:
According to the username information that the identities match treating training object is corresponding;
The electronic register information treating training object is obtained according to username information;
The attribute information that training object is corresponding with input entry is treated according to electronic register acquisition of information.
Specifically, the present embodiment is when determining until the identity of training object, easily according to the identity treating training object
Join corresponding username information, and obtain the electronic register information treating training object according to username information, and according to
Electronic register acquisition of information treats the attribute information that training object is corresponding with input entry, it is achieved that obtains rapidly and accurately and waits to train
The attribute information that object is corresponding with input entry, further increases and uses training requirement identification Model Identification employee's to be trained
The recognition efficiency of training requirement.
Alternatively, treat training object according to training contents to give training and include:
Treating training object and send training notice, training notice at least includes training contents, training time and place for training;
According to training contents, training object is giveed training.
Specifically, in the present embodiment, robot treats training object and sends after training notice, according in training notice time
Between and location information treat training object.For example, it is assumed that the when and where that training object gives training is treated in robot arrangement
Be respectively A and B, then robot can treat the arrival of training object in corresponding A time and B place, and receives it
And training.It should be noted that the present embodiment carries out aspectant training side except using training robot to treat training object
Outside method, it is also possible to take online training method, such as by methods such as long-distance video, voice, word such as mobile phone, PC, TVs.
By robot notice, the present embodiment treats that training object participates in training, and train the object participating in training
Instruction, without manpower notice, whole process is treated that training object and the trainer without specialty treat training object and give training, is flowed
Journey specification is simple, saves substantial amounts of human cost.
Alternatively, treat after training object gives training according to training contents and also include:
Treat training object and give examination training.
The training of existing enterprise or unit, often just for treating that training object gives training, does not arrange corresponding
Training and examination, so that training organizer cannot be known whether reaches training purpose, also cannot understand result of training, treat
Training object cannot understand the grasp situation to training contents.For this problem, the present embodiment treats training according to training contents
Object also includes after giveing training that treating training object gives examination training, beneficially training organization person and treat training object pair
Training carries out oneself's detection.
Alternatively, input entry includes:
Sex, age, education background, occupation, post, academic title, target, target completion rate, history training record bar
One or more combinations in mesh.
Specifically, the input entry in the present embodiment include but not limited to sex, the age, education background, occupation, post,
One or more combinations in academic title, target, target completion rate, history training record entry, such as, can also wrap
Include the total amount of makeing mistakes of work, place team tasks completion rate etc., specifically by User Defined.
Alternatively, training requirement classification includes:
Managerial skills, sales skills, communication skill, programming skill, skill in design, recruitment technical ability, planning technical ability, market development skill
One or more combinations in energy, marketing technical ability.
Specifically, the training requirement classification in the present embodiment includes but not limited to managerial skills, sales skills, communication skill
One or more combinations in energy, skill in design, recruitment technical ability, planning technical ability, market development technical ability, marketing technical ability, tool
Body is by User Defined.
Simplify embodiment below for two the present invention trains the implementation method of robot to illustrate further.
Simplify embodiment one
With reference to Fig. 2, the present invention simplifies the implementation method training robot that embodiment one provides, including:
Step S201, presets input entry and training requirement classification.
Specifically, the input entry in the present embodiment include but not limited to sex, the age, education background, occupation, post,
One or more combinations in academic title, target, target completion rate, history training record entry.Assume the present embodiment
In treat that training object is individual, and the input entry pre-set includes employee work post, miswork number of times, working year
Limit three input entries.
Moreover, it is assumed that the training requirement classification that the present embodiment pre-sets be sales skills, communication skill, programming skill,
One or more combinations in skill in design, recruitment technical ability, planning technical ability, market development technical ability, marketing technical ability.
Step S202, gathers the sample input data corresponding with input entry of training sample, and inputs number according to sample
According to the input feature value extracting training sample.
Specifically, the training requirement identification model higher in order to obtain discrimination, the training sample that the present embodiment is chosen
Quantity should be tried one's best greatly.The quantity assuming the training sample that the present embodiment chooses is N, then for this N number of training sample, adopt respectively
Collect the sample input data corresponding with input entry, and it is corresponding therewith to input data extraction according to the sample of each training sample
Input feature value.The input entry pre-set due to the present embodiment is three, namely for each training sample,
Corresponding 3 data can be collected.In actual implementation process, the present embodiment gather training sample with input bar
After the sample input data that mesh is corresponding, extract the input feature value of training sample according to sample input data before, also wrap
Include and sample input data are normalized.
Step S203, trains grader according to the training requirement classification of input feature value and training sample, it is thus achieved that training
Demand identification model.
Specifically, due to input feature value and the training requirement classification of the known training sample of the present embodiment, therefore will instruction
Practice the input feature value input vector as grader of sample, using the training requirement classification of training sample as grader
Output vector, thus train training requirement identification model.Wherein, the grader in the present embodiment can be Bayes's classification
Device, support vector machine classifier, maximum entropy classifiers etc..
It should be noted that the training requirement identification model in the present embodiment has the function automatically updated, such as when adopting
Collect to the training requirement classification corresponding with the attribute information of training sample be not belonging to training requirement classification set in advance time, this reality
Execute example this newly-increased training requirement classification to be added in training requirement classification set in advance.
Step S204, identifies the identity treating training object.
The present embodiment is training for identifying after the training requirement identification model of training object, starts to gather and waits to train
The attribute information that object is corresponding with input entry.Specifically, the identity treating training object is first identified, in actual implementation process
In, the present embodiment can treat the biological characteristic of training object, its identity of match cognization, it is also possible to searched by the Internet by collection
Rope obtains the username information with biometric matches, and treats the identity of training object according to username information identification.
Step S205, according to the attribute information that the described identity-acquiring treating training object is corresponding with described input entry.
Specifically, the input entry preset due to the present embodiment includes employee work post, miswork number of times, working year
Limit three input entries, then the attribute information the most corresponding with these three attributes entries of training object is treated in the present embodiment collection,
Assume the present embodiment collect with treat training object with above three input attribute information corresponding to entry be respectively programmer,
4 times, half a year.
Step S206, treats the input feature value of training object according to the extraction of described attribute information.
Specifically, after the attribute information treating training object is normalized, can obtain and treat that the input of training object is special
Levy vector.
Step S207, inputs described training requirement identification model by the described input feature value treating training object, it is thus achieved that
The described training requirement treating training object.
Specifically, the input feature value treating training object is inputted training requirement identification model, is then readily available and waits to train
The training requirement of instruction object, for example, programming skill.
As can be seen here, the problem automatically obtaining training requirement is converted to choosing of training requirement by the present embodiment dexterously,
And by design training demand identification model, can quickly and accurately obtain the training requirement treating training object,
Achieve the training requirement that training object acquisition is different for the treatment of for different attribute information characteristics, personalized and intelligence degree
High.
Can be real-time it should be noted that the present embodiment trains robot, it is also possible to be acquired periodically and treat training object
Training requirement.Additionally, the trigger condition that the present embodiment can also obtain by setting training requirement starts to obtain training requirement, example
As arranged in the present embodiment when when training object miswork number of times is more than 3 times, start to gather and treat training object
Attribute information, and based on the current training requirement of the training requirement identification Model Identification trained.
Step S208, obtains the training contents mated with the described training requirement treating training object preset.
Specifically, the present embodiment determines training requirement when being programming skill, can be according to set in advance with programming skill
Concrete training contents, such as C language Course Training, Jave Course Training etc. can be obtained by corresponding training contents.
To described, step S209, treats that training object gives training according to described training contents.
The implementation method of the training robot of the embodiment of the present invention, is trained for identifying training requirement by machine learning
Training requirement identification model, and according to training requirement identification model and treat that the attribute information of training object automatically generates training
Demand, and automatically obtain, according to the training requirement generated, the training contents mated with training requirement further, and final according to training
Instruction content is treated training object and is giveed training, solves existing enterprise or unit is carried out the link of training and is substantially dependent on people
Work completes, thus causes training heavy, the problem that human cost is higher.Whole process without manually determining training requirement,
Formulate training scheme and arrangement treats that training object gives training, save human cost, improve training efficiency.
Simplify embodiment two
With reference to Fig. 3, the present invention simplifies the implementation method training robot that embodiment one provides, including:
Step S301, presets input entry and training requirement classification.
Specifically, the input entry in the present embodiment include but not limited to sex, the age, education background, occupation, post,
One or more combinations in academic title, target, target completion rate, history training record entry.Assume the present embodiment
Treat that training object is team, and the input entry pre-set include team sales volume target, actual sales revenue, to customer family
Amount, client's probability of transaction four input entry.
Moreover, it is assumed that the training requirement classification that the present embodiment pre-sets be sales skills, communication skill, programming skill,
One or more combinations in skill in design, recruitment technical ability, planning technical ability, market development technical ability, marketing technical ability.
Step S302, gathers the sample input data corresponding with input entry of training sample, and inputs number according to sample
According to the input feature value extracting training sample.
Specifically, the training requirement identification model higher in order to obtain discrimination, the training sample that the present embodiment is chosen
Quantity should be tried one's best greatly.In actual implementation process, the present embodiment gathers the sample corresponding with input entry of training sample
After input data, extract the input feature value of training sample according to sample input data before, also include sample is inputted
Data are normalized.
Step S303, trains grader according to the training requirement classification of input feature value and training sample, it is thus achieved that training
Demand identification model.
Specifically, due to input feature value and the training requirement classification of the known training sample of the present embodiment, therefore will instruction
Practice the input feature value input vector as grader of sample, using the training requirement classification of training sample as grader
Output vector, thus train training requirement identification model.Wherein, the grader in the present embodiment can be Bayes's classification
Device, support vector machine classifier, maximum entropy classifiers etc..
It should be noted that the training requirement identification model in the present embodiment has the function automatically updated, such as when adopting
Collect to the training requirement classification corresponding with the attribute information of training sample be not belonging to training requirement classification set in advance time, this reality
Execute example this newly-increased training requirement classification to be added in training requirement classification set in advance.
Step S304, according to the attribute information corresponding with input entry and the training requirement identification mould for the treatment of training object
Type, determines the training requirement treating training object.
Specifically, present embodiment assumes that collect with team sales volume target, actual sales revenue, to customer family amount, client
The attribute information that probability of transaction four inputs entry corresponding is 1,000,000,400,000,100,20%, and according to the training requirement trained
Identify that Model Identification goes out to treat in the present embodiment that the training requirement of training object (team) is sales skills+communication skill.
Step S305, obtains the training contents mated with the training requirement treating training object preset.
Specifically, the present embodiment determines training requirement when being sales skills+communication skill, can be according to set in advance
The training contents the most corresponding with sales skills and communication skill obtains concrete training contents.Additionally, the present embodiment is except basis
The training contents mated with training requirement set in advance obtains outside training contents, it is also possible to search on the internet obtains and institute
State the training contents that training requirement is corresponding, or according to company or unit history training record, choose the training corresponding with training requirement
Instruction content.And the present embodiment is according to company or unit history training record, when choosing the training contents corresponding with training requirement, can
To choose the bigger history training contents of the accounting corresponding with training requirement as training contents.Such as in company or unit history
Training record obtains 10 parts of history training records giveed training for sales skills, and wherein has 6 employees for sale
The history training contents that technical ability gives training is course A and course B, has the history that 2 employees give training for sales skills
Training contents is course C, and having 2 employees is course D, then the present embodiment for the history training contents that sales skills give training
Choose the bigger history training contents of the accounting corresponding with training requirement as the training mated with the training requirement treating training object
Instruction content, the most now chooses course A and course B as the training contents mated with the training requirement treating training object.
Step S306, treat training object send training notice, training notice at least include training contents, the training time and
Place for training.
Specifically, in the present embodiment, robot determines after the training contents of training object, treats training object and sends training
Instruction notice, wherein training notice at least includes training contents, training time and place for training.And after sending training notice, root
Training object is treated according to the when and where information in training notice.For example, it is assumed that robot arrangement treats that training object is carried out
The when and where of training is respectively A and B, then robot can treat arriving of training object in corresponding A time and B place
Come, and it is received and trains.By robot notice, the present embodiment treats that training object participates in training, and to participating in training
The object of instruction gives training, and without manpower notice, whole process treats that training object and the trainer without specialty treat training right
As giveing training, standard process is simple, saves substantial amounts of human cost.
Step S307, gives training training object according to training contents.
To described, step S308, treats that training object gives examination training.
Specifically, the present embodiment is treated according to training contents and is also included after training object gives training treating training object
Give examination training, beneficially training organization person and treat training object training is carried out oneself detection.
As can be seen here, the implementation method of the training robot that the present invention provides, it is not necessary to manually determine training requirement, formulate training
Instruction scheme and arrangement treat that training object gives training, save human cost, improve training efficiency.And trained by employing
Good training requirement identification model is treated the training requirement of trainer and is carried out Intelligent Recognition, improves training requirement in training
The degree of accuracy obtained;And by automatically obtaining the training contents mated with training requirement, it is possible to obtain scientific and reasonable training side
Case, the intellectuality to whole training realizes significant.
These are only the preferred embodiments of the present invention, be not limited to the present invention, for those skilled in the art
For Yuan, the present invention can have various modifications and variations.All within the spirit and principles in the present invention, any amendment of being made,
Equivalent, improvement etc., should be included within the scope of the present invention.
Claims (10)
1. the implementation method training robot, it is characterised in that including:
Preset input entry and training requirement classification;
Gather the sample input data corresponding with described input entry of training sample, and extract according to described sample input data
The input feature value of described training sample;
Training requirement classification training grader according to described input feature value and described training sample, it is thus achieved that training requirement is known
Other model;
According to treating the attribute information corresponding with described input entry of training object and described training requirement identification model, determine
The described training requirement treating training object;
Obtain the training contents mated with the described training requirement treating training object preset;
Treat that training object gives training according to described training contents to described.
The implementation method of training robot the most according to claim 1, it is characterised in that according to treat training object with institute
State input attribute information corresponding to entry and described training requirement identification model, determine described in treat the training requirement of training object
Including:
Gather the attribute information corresponding with described input entry treating training object;
The input feature value of training object is treated according to the extraction of described attribute information;
The described input feature value treating training object is inputted described training requirement identification model, it is thus achieved that described in treat training object
Training requirement.
The implementation method of training robot the most according to claim 2, it is characterised in that gather treat training object with institute
The attribute information stating input entry corresponding includes:
Identify the identity treating training object;
According to the attribute information that the described identity-acquiring treating training object is corresponding with described input entry.
The implementation method of training robot the most according to claim 3, it is characterised in that identify the identity treating training object
Including:
Obtain the biological characteristic treating training object;
Described biological characteristic is mated with the sample biological characteristic being stored in advance in data base, thus identify wait to train right
The identity of elephant.
The implementation method of training robot the most according to claim 4, it is characterised in that when described biological characteristic is with in advance
When the sample biological characteristic being stored in data base does not matches that, also include:
Search on the internet and the username information of the described biometric matches treating training object;
The identity of training object is treated according to described username information identification.
6. according to the implementation method of the arbitrary described training robot of claim 3 to 5, it is characterised in that train according to described waiting
The attribute information that the identity-acquiring of instruction object is corresponding with described input entry includes:
According to the username information that the described identities match treating training object is corresponding;
The electronic register information of training object is treated according to the acquisition of described username information;
Treat, according to described electronic register acquisition of information, the attribute information that training object is corresponding with described input entry.
The implementation method of training robot the most according to claim 6, it is characterised in that according to described training contents to institute
State and treat that training object gives training and include:
Treating that training object sends training notice to described, described training notice at least includes training contents, training time and training
Place;
According to described training contents, described training object is giveed training.
The implementation method of training robot the most according to claim 7, it is characterised in that according to described training contents to institute
State and treat that training object also includes after giveing training:
Treat that training object gives examination training to described.
The implementation method of training robot the most according to claim 8, it is characterised in that described input entry includes:
Sex, age, education background, occupation, post, academic title, target, target completion rate, history training record bar
One or more combinations in mesh.
The implementation method of training robot the most according to claim 9, it is characterised in that described training requirement classification bag
Include:
Managerial skills, sales skills, communication skill, programming skill, skill in design, recruitment technical ability, planning technical ability, market development skill
One or more combinations in energy, marketing technical ability.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610844764.8A CN106205288A (en) | 2016-09-23 | 2016-09-23 | A kind of implementation method training robot |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610844764.8A CN106205288A (en) | 2016-09-23 | 2016-09-23 | A kind of implementation method training robot |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106205288A true CN106205288A (en) | 2016-12-07 |
Family
ID=58067168
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610844764.8A Pending CN106205288A (en) | 2016-09-23 | 2016-09-23 | A kind of implementation method training robot |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106205288A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107493352A (en) * | 2017-10-09 | 2017-12-19 | 陈包容 | A kind of mthods, systems and devices of tangible machine people intelligent training |
CN108564508A (en) * | 2018-04-27 | 2018-09-21 | 武汉文都信息技术有限公司 | The method and apparatus of shared trainee's information |
CN108734379A (en) * | 2018-04-03 | 2018-11-02 | 四川新网银行股份有限公司 | It is a kind of that Training Methodology on the line of differentiation is realized to contact staff |
CN111340664A (en) * | 2020-03-25 | 2020-06-26 | 江苏安泰信息科技发展有限公司 | Enterprise safety production training system and operation method |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007233077A (en) * | 2006-03-01 | 2007-09-13 | Yamaha Corp | Evaluation device, control method, and program |
CN101436277A (en) * | 2008-12-04 | 2009-05-20 | 北京安博在线软件有限公司 | System for providing test comment, training, job hunting and recruitment |
CN102456344A (en) * | 2010-10-22 | 2012-05-16 | 中国电信股份有限公司 | System and method for analyzing customer behavior characteristic based on speech recognition technique |
CN103544210A (en) * | 2013-09-02 | 2014-01-29 | 烟台中科网络技术研究所 | System and method for identifying webpage types |
CN103839476A (en) * | 2013-11-04 | 2014-06-04 | 广西电网公司电力科学研究院 | Three-dimensional simulation system for production supervision of assembly of transformer |
CN104834728A (en) * | 2015-05-14 | 2015-08-12 | 无锡天脉聚源传媒科技有限公司 | Pushing method and device for subscribed video |
CN105574788A (en) * | 2016-02-05 | 2016-05-11 | 北京普猎创新网络科技有限公司 | Competency label-based training providing method |
CN105894253A (en) * | 2016-05-09 | 2016-08-24 | 陈包容 | Method and device for automatic pushing of job application demand |
-
2016
- 2016-09-23 CN CN201610844764.8A patent/CN106205288A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007233077A (en) * | 2006-03-01 | 2007-09-13 | Yamaha Corp | Evaluation device, control method, and program |
CN101436277A (en) * | 2008-12-04 | 2009-05-20 | 北京安博在线软件有限公司 | System for providing test comment, training, job hunting and recruitment |
CN102456344A (en) * | 2010-10-22 | 2012-05-16 | 中国电信股份有限公司 | System and method for analyzing customer behavior characteristic based on speech recognition technique |
CN103544210A (en) * | 2013-09-02 | 2014-01-29 | 烟台中科网络技术研究所 | System and method for identifying webpage types |
CN103839476A (en) * | 2013-11-04 | 2014-06-04 | 广西电网公司电力科学研究院 | Three-dimensional simulation system for production supervision of assembly of transformer |
CN104834728A (en) * | 2015-05-14 | 2015-08-12 | 无锡天脉聚源传媒科技有限公司 | Pushing method and device for subscribed video |
CN105574788A (en) * | 2016-02-05 | 2016-05-11 | 北京普猎创新网络科技有限公司 | Competency label-based training providing method |
CN105894253A (en) * | 2016-05-09 | 2016-08-24 | 陈包容 | Method and device for automatic pushing of job application demand |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107493352A (en) * | 2017-10-09 | 2017-12-19 | 陈包容 | A kind of mthods, systems and devices of tangible machine people intelligent training |
CN108734379A (en) * | 2018-04-03 | 2018-11-02 | 四川新网银行股份有限公司 | It is a kind of that Training Methodology on the line of differentiation is realized to contact staff |
CN108734379B (en) * | 2018-04-03 | 2022-02-11 | 四川新网银行股份有限公司 | Online training method for realizing differentiation of customer service personnel |
CN108564508A (en) * | 2018-04-27 | 2018-09-21 | 武汉文都信息技术有限公司 | The method and apparatus of shared trainee's information |
CN111340664A (en) * | 2020-03-25 | 2020-06-26 | 江苏安泰信息科技发展有限公司 | Enterprise safety production training system and operation method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2017167069A1 (en) | Resume assessment method and apparatus | |
CN110765257A (en) | Intelligent consulting system of law of knowledge map driving type | |
CN106408249A (en) | Resume and position matching method and device | |
CN109635117A (en) | A kind of knowledge based spectrum recognition user intention method and device | |
CN106205288A (en) | A kind of implementation method training robot | |
CN106649524A (en) | Improved advanced study intelligent response system based on computer cloud data | |
CN108255921A (en) | The system for determining and showing the relevant explanation of recommendation | |
JP2020521210A (en) | Information processing method and terminal, computer storage medium | |
CN101441660A (en) | Knowledge evaluating system and method in inquiry and answer community | |
CN101799849A (en) | Method for realizing non-barrier automatic psychological consult by adopting computer | |
CN101118554A (en) | Intelligent interactive request-answering system and processing method thereof | |
CN109978020B (en) | Social network account number vest identity identification method based on multi-dimensional features | |
CN103761254A (en) | Method for matching and recommending service themes in various fields | |
CN109032591A (en) | A kind of crowdsourcing software developer recommended method neural network based | |
CN117114514B (en) | Talent information analysis management method, system and device based on big data | |
CN111510368B (en) | Family group identification method, device, equipment and computer readable storage medium | |
CN112819286A (en) | Classified matching system and method for high embedding degree of elite talents and posts | |
CN111242565A (en) | Resume optimization method and device based on intelligent personnel model | |
CN109063116A (en) | Data identification method, device, electronic equipment and computer readable storage medium | |
CN110532363A (en) | A kind of automatic dialogue method of task orientation type based on decision tree | |
CN115577172A (en) | Article recommendation method, device, equipment and medium | |
CN115640074A (en) | Service data processing method and device and intelligent counter terminal | |
Ragab et al. | Design and implementation of a hybrid recommender system for predicting college admission | |
US20240127143A1 (en) | Method, device and storage medium for information processing based on data interaction | |
CN108257042A (en) | A kind of task-driven type intelligent electricity selling business support system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161207 |