[go: up one dir, main page]

WO2018120123A1 - Navigation vers un objectif - Google Patents

Navigation vers un objectif Download PDF

Info

Publication number
WO2018120123A1
WO2018120123A1 PCT/CN2016/113741 CN2016113741W WO2018120123A1 WO 2018120123 A1 WO2018120123 A1 WO 2018120123A1 CN 2016113741 W CN2016113741 W CN 2016113741W WO 2018120123 A1 WO2018120123 A1 WO 2018120123A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
individual
group including
goal
nodes
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.)
Ceased
Application number
PCT/CN2016/113741
Other languages
English (en)
Inventor
Dong Wang
Yunqing XIA
Xue Liu
Hong Yu
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Microsoft Technology Licensing LLC
Original Assignee
Microsoft Technology Licensing LLC
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Microsoft Technology Licensing LLC filed Critical Microsoft Technology Licensing LLC
Priority to CN201680080880.5A priority Critical patent/CN108604247A/zh
Priority to PCT/CN2016/113741 priority patent/WO2018120123A1/fr
Publication of WO2018120123A1 publication Critical patent/WO2018120123A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning

Definitions

  • a career navigation service may help users to identify a path to their desired career goals.
  • the career navigation service may associate the user with a role model which is based on a real-life person having achieved the goal job.
  • the career navigation service may provide the user with a career path of the real-life person as advice for achieving the goal job. This career path is based on experiences of the real-life person.
  • Embodiments of the present disclosure may establish a goal navigation advice model based on numerous training samples through data mining and machine learning.
  • the goal navigation advice model may comprise information on requirements for achieving various goals.
  • Goal navigation services may be provided based on the goal navigation advice model.
  • FIG. 1 illustrates an exemplary application scenario for providing goal navigation.
  • FIG. 2 is a flowchart of an exemplary method for providing career goal navigation according to an embodiment.
  • FIG. 3A illustrates an exemplary presentation of career goal navigation according to an embodiment.
  • FIG. 3B illustrates an exemplary presentation of career goal navigation according to an embodiment.
  • FIG. 4 illustrates an exemplary presentation of career goal navigation according to an embodiment.
  • FIG. 5 is a flowchart of an exemplary method for providing career goal navigation according to an embodiment.
  • FIG. 6 is a flowchart of an exemplary process for generating rules associated with job positions according to an embodiment.
  • FIG. 7 is a flowchart of an exemplary method for navigating among a plurality of nodes according to an embodiment.
  • FIG. 8 is a flowchart of an exemplary method according to an embodiment.
  • FIG. 9 illustrates an exemplary apparatus for navigating among a plurality of nodes according to an embodiment.
  • FIG. 10 illustrates an exemplary apparatus for navigating to a goal according to an embodiment.
  • FIG. 11 illustrates an exemplary system according to an embodiment.
  • FIG. 12 illustrates an exemplary system for providing goal navigation according to an embodiment.
  • Embodiments of the present disclosure may provide a goal navigation system.
  • the goal navigation system may be configured for providing goal navigation services.
  • the goal navigation system may establish a goal navigation advice model based on numerous training samples through data mining and machine learning.
  • the training samples may refer to data containing various attributes of individuals, and may be used as source data for the data mining and training data for the machine learning.
  • the goal navigation advice model may comprise information on requirements for achieving various goals.
  • the goal navigation system may provide advice on how to achieve a goal being queried. For example, if a user is querying how to achieve a goal, the goal navigation system may provide information on at least one path directed to the goal based on the goal navigation advice model.
  • FIG. 1 illustrates an exemplary application scenario 100 for providing goal navigation.
  • a network 110 is applied for interconnecting various network entities.
  • the network 110 may be any type of networks capable of interconnecting the network entities.
  • the network 110 may be a single network or a combination of various networks.
  • the network 110 may be a Local Area Network (LAN) , a Wide Area Network (WAN) , etc.
  • the network 110 may be a wired network or a wireless network.
  • the network 110 may be a packet switching network, a message switching network, etc.
  • a terminal device 120 may be any type of electronic computing devices capable of connecting to the network 110, assessing servers or websites on the network 110, processing data or signals, etc.
  • the terminal device 120 may be a desktop computer, laptop, tablet, smart phone, etc. Although only one terminal device 120 is shown in FIG. 1, it should be appreciated that any other terminal devices may connect to the network 110.
  • a goal navigation system 130 connects to the network 110.
  • the goal navigation system 130 is configured for providing goal navigation services.
  • the goal navigation system 130 may include a goal navigation search engine 132, a model establishing unit 134 and a goal navigation database 136.
  • the goal navigation search engine 132 may interact with users and provide advice on how to achieve goals being queried by the users.
  • the model establishing unit 134 may establish a goal navigation advice model based on numerous training samples through data mining and machine learning.
  • the goal navigation database 136 may store information, such as the goal navigation advice model, that is used by the goal navigation search engine 132 to provide advice.
  • the goal navigation search engine 132 may obtain information on how to achieve the goal based on the goal navigation advice model stored in the goal navigation database 136, and provide the information to the user of the terminal device 120 as advice.
  • the goal navigation search engine 132, the model establishing unit 134 and the goal navigation database 136 are included in the goal navigation system 130, it should be appreciated that anyone of the goal navigation search engine 132, the model establishing unit 134 and the goal navigation database 136 may be separated from the goal navigation system 130.
  • the model establishing unit 134 may be a single network entity, and thus the functionality of establishing the goal navigation advice model may be separated from the goal navigation system 130.
  • the goal navigation system 130 may receive and store a goal navigation advice model established by the model establishing unit 134.
  • a sample source 140 in FIG. 1 refers to various data sources that may provide training samples. These training samples can be used by the model establishing unit 134 to establish the goal navigation advice model.
  • the sample source 140 may be a single database containing training samples and connecting to the network 110, or be a part of a server which may obtain training samples based on users’access.
  • the goal navigation system may be applied in various approaches.
  • the goal navigation system may be applied for providing career goal navigation.
  • career goal navigation For the sake of facilitating to explain technical details of some exemplary embodiments, the following description will take career goal navigation as an example. However, it should be appreciated that the present disclosure is not limited to career goal navigation in any way.
  • FIG. 2 is a flowchart of an exemplary method 200 for providing career goal navigation according to an embodiment.
  • the method 200 may be used by the goal navigation system according to the embodiments of the present disclosure to provide advice on how to achieve a career goal.
  • the method 200 may be triggered by a query submitted by a user through a terminal device to the goal navigation system, wherein the user may want to know how to achieve a career goal, e.g., a target job position.
  • the goal navigation system may initiatively perform a query on how to achieve a target job position from a start job position.
  • the method 200 may be performed by a client application preloaded or downloaded to the terminal device of the user through interacting with the goal navigation system.
  • the method 200 starts at 202, and proceeds to 204.
  • a start job position and a target job position may be determined.
  • the start job position may be a current job position of a user, and the target job position may be a career goal.
  • job position may also be referred to as “node” or “milestone” .
  • Node or milestone may indicate a critical position or point along a trajectory.
  • a “node” may refer to a “job position” along a career path.
  • the method 200 may be triggered by a query on how to achieve the target job position that is submitted by a user through a terminal device to the goal navigation system.
  • the user may be currently at a job position, for example, the start job position.
  • the goal navigation system may determine the start job position from registration information of the user.
  • the user may have registered on a website providing goal navigation services, and submitted an individual profile at least including a current job position to the website.
  • the goal navigation system may retrieve at least the current job position of the user from registration information of the user as the start job position.
  • the goal navigation system may obtain the start job position inputted by the user together with the query.
  • the goal navigation system may further determine the target job position from the query.
  • the method 200 may be triggered by a query initiatively performed by the goal navigation system.
  • the start job position and the target job position would already be known to the goal navigation system.
  • a client application preloaded or downloaded to the terminal device of the user may determine the start job position from registration information of the user, or receive the start job position inputted by the user.
  • the client application may further determine the target job position from the query of the user.
  • the start job position and the target job position may be transmitted by the client application to the goal navigation system through networks.
  • information on at least one path navigated from the start job position to the target job position may be determined.
  • the start job position and the target job position may be used for building one or more paths from the path navigation advice model.
  • a path directly navigated from the start job position to the target job position may be built.
  • there may be one or more intermediate job positions between the start job position and the target job position and thus a path including the start job position, the one or more intermediate job positions, and the target job position may be found.
  • Information on the path may also be obtained from the path navigation advice model.
  • the information on the path may include a probability of a job position on the path being navigated to another job position on the path.
  • the information on the path may further include individual attributes associated with at least one job position on the path and weights of the individual attributes. “Individual attributes” herein relate to personal attributes. In the scenario of career goal navigation, individual attributes may refer to skill, education, work experience, and other relevant attributes of a user.
  • path information may be presented. For example, all or a portion of the information determined at 206 may be presented to the user who is querying. Accordingly, the user may view the presented information, e.g., on a display screen of a terminal device, and thus obtain advice on how to achieve the career goal he is querying.
  • the information may be presented in various approaches. For example, the information may be presented in one or more graphs or in one or more tables, presented all at once or step by step, etc.
  • notifications may be further pushed to the user.
  • the notifications may include reminders that remind the user to try to achieve the career goal, information on how much way is left until achieving the career goal, etc.
  • the method 200 for providing career goal navigation may be performed dynamically. For example, after presenting the path information, a status of the user may be monitored based on performance of the user. If it is detected that the status of the user is updated, e.g., reaching a new job position, this new job position may be used as an updated start job position, and accordingly, through the method 200, updated path information may be presented. Moreover, the path navigation advice model of the goal navigation system may also be dynamically updated based on information on the update of the status of the user.
  • FIG. 3A illustrates an exemplary presentation 300a of career goal navigation according to an embodiment.
  • a start job position is “College Student” and a target job position is “Software Test Engineer (STE) ”
  • the presentation 300a in a graph form may be provided by the goal navigation system as advice.
  • the college student herein may refer to a college student majoring in computer science.
  • the first path is formed by four sequential job positions including College Student 310, Support Engineer Intern 312, Support Engineer 314 and STE 320.
  • the second path is formed by four sequential job positions including College Student 310, Software Develop Engineer (SDE) Intern 316, SDE 318 and STE 320.
  • SDE Software Develop Engineer
  • the presentation 300a may include information on navigation, or called “transitions” , between any two sequential job positions.
  • block 330 shows information on transition from College Student 310 to Support Engineer Intern 312.
  • the information may include a probability of College Student 310 being navigated to Support Engineer Intern 312, e.g., “40%” . That is, a college student majoring in computer science may have a probability of “40%” to become a support engineer intern.
  • the information may include individual attributes required to achieve the transition and weights of the individual attributes.
  • the individual attributes shown in block 330 may be used for indicating what types of attributes the user should have such that the user could transit from College Student 310 to Support Engineer Intern 312.
  • a weight of an individual attribute may indicate importance of the individual attribute in the transition.
  • Block 330 shows an individual attribute “Network Security” and its weight “0.462” , an individual attribute “Troubleshooting” and its weight “0.458” , an individual attribute “Windows Server” and its weight “0.315” , an individual attribute “Data Optimization” and its weight “0.132” , etc.
  • the presentation 300a further includes blocks 332, 334, 336, 338 and 340 for showing information on corresponding transitions.
  • the blocks 330 to 340 may be shown all at once. In another implementation, the blocks 330 to 340 may be hidden firstly, and when a user is clicking on or pointing to a line of a transition, a corresponding block is shown.
  • the individual attributes herein may refer to skill, education, work experience, etc.
  • the “Network Security” , “Troubleshooting” , “Windows Server” in block 330 are skills
  • the “Master Degree” in block 334 is an education level
  • the “3-year experience in software development” in block 340 is a type of work experience.
  • Each of the blocks 330 to 340 may show all individual attributes required to achieve a corresponding transition, or only show one or more of the individual attributes, such as the top 5 or 10 weighted individual attributes.
  • each of the blocks 330 to 340 shows the weights of individual attributes, the weights may also be hidden.
  • the probability can be shown or indicated in any other approaches.
  • the probability may be indicated by a color of a line between two sequential job positions. Different colors may indicate different probabilities.
  • the probability may be indicated by a thickness of a line between two sequential job positions. Thicker lines may indicate higher probabilities.
  • the probability may also be hidden from the presentation 300a.
  • the presentation 300 may firstly show the transition from 310 to 312, and in response to an instruction from the user, further show the next transition from 312 to 314, and so on.
  • relevant courses may be associated with the individual attributes. For example, a hyperlink to a relevant course may be shown beside a corresponding individual attribute.
  • FIG. 3B illustrates an exemplary presentation 300b of career goal navigation according to an embodiment, which is in a different graph form from the presentation 300a.
  • a start job position is “College Student”
  • a target job position is “Research Software Development Engineer (RSDE) ”
  • the presentation 300b may be provided by the goal navigation system as advice.
  • the first path is formed by three sequential job positions including College Student 350, Researcher Intern 352 and RSDE 360.
  • the second path is formed by three sequential job positions including College Student 350, SDE Intern 354 and RSDE 360.
  • the presentation 300b may include information on navigation, or called “transitions” , between any two sequential job positions.
  • the area 370 shows information on transition from College Student 350 to Researcher Intern 352.
  • the showing of information in the area 370 may be triggered by, such as, a user’s operation of pointing to a line of the transition between College Student 350 and Researcher Intern 352. Information on other transitions may also be shown in a similar way.
  • the information in the area 370 may include individual attributes and corresponding weights.
  • the individual attributes may be, such as, “C#” , “C++” , “Data Structures” , “Software Development” , etc.
  • the weights of the individual attributes are shown by progress bars, instead of specific numbers as shown in FIG. 3A.
  • relevant courses may also be shown in the area 370. For example, the text “ [course] ” beside “C#” may be a hyperlink to a course about C#, and thus if the user clicks on the text “ [course] ” , the user may be redirected to, such as, a relevant online course.
  • the area 370 may show all individual attributes required to achieve a corresponding transition, or only show one or more of the individual attributes, such as the top 5 or 10 weighted individual attributes. Although the area 370 shows the weights of individual attributes, the weights may also be hidden.
  • the probability can be additionally shown or indicated in any approaches.
  • the probability may be indicated by a color of a line between two sequential job positions. Different colors may indicate different probabilities.
  • the probability may be indicated by a thickness of a line between two sequential job positions. Thicker lines may indicate higher probabilities.
  • the entire two paths from College Student to RSDE are shown in FIG. 3B, it should be appreciated that these two paths may be shown separately, and each transition in a path may also be shown separately, e.g., step by step.
  • the presentation 300b may firstly show the transition from 350 to 352, and in response to an instruction from the user, further show the next transition from 352 to 360.
  • FIG. 4 illustrates an exemplary presentation 400 of career goal navigation according to an embodiment.
  • the presentation 400 is in a table form, and may be viewed as a transformation of the presentation 300a which is in the graph form.
  • the presentation 400 includes two groups of tables, each showing a path from College Student to STE.
  • the first group of tables includes a table 410 showing information on transition from College Student to Support Engineer Intern, a table 420 showing information on transition from Support Engineer Intern to Support Engineer, and a table 430 showing information on transition from Support Engineer to STE.
  • the second group of tables includes a table 440 showing information on transition from College Student to SDE Intern, a table 450 showing information on transition from SDE Intern to SDE, and a table 460 showing information on transition from SDE to STE.
  • Information in each of the tables 410 to 460 is similar as the information in blocks 330 to 340 in FIG. 3A respectively.
  • FIG. 3A, FIG. 3B and FIG. 4 are exemplary. The embodiments of the present disclosure are not limited to these exemplary presentations. According to actual application requirements, any elements in FIG. 3A, FIG. 3B and FIG. 4 may be adjusted, replaced or hidden. Moreover, besides the graph form in FIG. 3A, FIG. 3B and the table form in FIG. 4, any other forms suitable for showing information on transitions in paths may be adopted by the goal navigation system for presenting advice of career goal navigation.
  • FIG. 5 is a flowchart of an exemplary method 500 for providing career goal navigation according to an embodiment.
  • the method 500 may be performed by the goal navigation system for generating the goal navigation advice model based on numerous training samples through data mining and machine learning.
  • the method 500 starts at 502, and proceeds to 504.
  • a plurality of training samples may be obtained.
  • Each of the training samples may be associated with various individual attributes, and may be associated with at least one node.
  • a “training sample” may refer to an individual profile of a user, and individual attributes associated with the training sample may refer to any information identified from the individual profile, such as, skill, education, work experience, and other relevant attributes of the user.
  • the training samples may be obtained from social networking sites or online professional networks, such as, LinkedIn, Facebook, Google+, etc.
  • a training sample such as an individual profile, that includes personal information.
  • the individual profile may include skill, education, work experience, summary, and other relevant attributes of the user.
  • the “skill” may include professional skills that the user is or was having.
  • the “education” may include schools attended, majors, degrees, time periods, etc.
  • the “work experience” may include companies worked at, job positions, job description, time periods, etc.
  • the summary may be a section of text which includes information about, such as, self-introduction, awards, social activities, etc. of the user.
  • job positions and individual attributes may be extracted from the training samples.
  • job positions and individual attributes may be directly extracted from each training sample.
  • at least one job position may be extracted as a node from a training sample.
  • individual attributes such as, skills, schools attended, degrees, companies worked at, etc. may also be extracted from the training sample.
  • job positions and individual attributes may be extracted from sections of text in each training sample based on semantic analysis.
  • the training sample may include sections of text freely input by the user, such as, some literal paragraphs included in the summary, the work experience, etc. of the training sample.
  • the summary may include information on self-introduction, awards, social activities, etc.
  • the information in the summary may also be helpful for establishing the goal navigation advice model and providing advice on career goal navigation. For example, if the summary recites that the user worked at Microsoft Corporation, which is a Fortune 500 company, then an individual attribute “Worked for Fortune 500 companies” may be extracted.
  • LDA Latent Dirichlet Allocation
  • LDA is a document topic generating model adopting unsupervised learning. Each document may be viewed as a set of topics, and each topic may be viewed as a set of words.
  • the LDA may be used for identifying topics of documents based on probability distributions among documents, topics and words, wherein each topic may be a cluster of key words/phrases.
  • the LDA may be applied in combination with a preset label library.
  • the label library may be generated empirically through associating key words/phrases with corresponding topic descriptions. Thus, when the LDA identifies a topic, a corresponding topic description may be determined from the label library through key word/phrase matching.
  • each section of text may be viewed as a document.
  • a label library may be generated empirically based on knowledge about career goal navigation. For example, a list of Fortune 500 companies may be compiled into a set of key words which correspond to a topic description of “Worked for Fortune 500 companies” ; a set of phrases like “responsible for building software products at a software company” may be compiled into a set of key phrases which correspond to a topic description of “3-year experience in software development” ; a set of phrases like “managing a team of 10 people” , “managing a team of 5-10 people” and so on may be compiled into a set of key phrases which correspond to a topic description of “Team management experience” .
  • job positions and individual attributes may be extracted from the sections of text in the training samples.
  • the job positions or individual attributes may be normalized.
  • the job positions or individual attributes extracted from the training samples are not in a normalized form. For example, when users enter text to describe their job positions, these users may enter various names for the same job position. Taking “SDE” as an example, some users may use the term “Software Engineer” , some users may use the abbreviation “SDE” , and other users may use the abbreviation “SWE” . However, all these names should correspond to the same job position “SDE” . The same issue may occur in some individual attributes, such as, company names, school names, etc. Thus, it would be beneficial to normalize the job positions or individual attributes and avoid treating various names of the same job position or individual attribute as different job positions or individual attributes, and this can further improve the accuracy of the goal navigation advice model to be generated.
  • Various approaches may be adopted by the embodiments of the present disclosure to normalize the job positions or individual attributes.
  • a combination of K-means clustering algorithm and Convolutional Deep Structured Semantic Model (CDSSM) tool may be used for normalizing the job positions or individual attributes.
  • CDSSM Convolutional Deep Structured Semantic Model
  • the CDSSM tool may be used for generating deep semantic vectors of the job positions or individual attributes.
  • the K-means clustering algorithm may be further used for clustering the deep semantic vectors of the job positions or individual attributes.
  • Each cluster of deep semantic vectors may be viewed as relating to the same job positions or individual attributes, and thus can be given a normalized name.
  • deep semantic vectors of “Software Engineer” , “SDE” and “SWE” may be generated through the CDSSM tool and then clustered together through the K-means clustering algorithm. This cluster of deep semantic vectors may be uniformly named as “SDE” .
  • transitions among job positions are identified based on the individual attributes.
  • the job positions herein may be nodes in the training samples. That is, each job position in a training sample may be a node for the user. If a first job position in a training sample is identified as being navigated to a second job position in the training sample, a transition between these two job positions may be identified accordingly. Similarly, a set of transitions among job positions in the training samples may be identified.
  • a decision on whether there is one more transition may be made, thus the following processes may be iterated for all the transitions in the set of transitions.
  • the method 500 proceeds to 514. It can be assumed that the transition determined at 512 indicates a transition from a first job position to a second job position.
  • training samples associated with the first job position of the transition determined at 512 may be determined.
  • all training samples identifying the first job position may be selected from the training samples obtained at 504. For example, if the transition indicates a transition from “College Student” to “Support Engineer Intern” , then all the training samples identifying a job position of “College Student” may be selected at 514.
  • a training data set may be determined for the transition.
  • the training data set may include attribute vectors and tags for the training samples determined at 514.
  • the attribute vectors may be generated based on the individual attributes of the training samples.
  • the dimension of attribute vectors may be predetermined as covering all individual attributes of the training samples obtained at 504.
  • the dimension of attribute vectors may be predetermined as covering all individual attributes of the training samples obtained at 504, and one or more predictable individual attributes.
  • the dimension of attribute vectors may be determined based on individual attributes of the training samples determined at 514.
  • the tags may be generated based on whether the transition from the first job position to the second job position is identified in the training samples.
  • Table 1 shows an example of the generating of the training data set. It is assumed that the transition is from “College Student” to “Support Engineer Intern” , the training samples are from “Bob” , “Tom” , “Jack” , etc., and the individual attributes include “C++” , “Java” , “Network Security” , “Master Degree” , etc.
  • the first column in Table 1 shows the users of the training samples.
  • the second column in Table 1 shows tags generated for the training samples, where “1” indicates that the transition is identified from the corresponding training sample, and “0” indicates that the transition is not identified from the corresponding training sample.
  • the third and following columns in Table 1 shows the individual attributes a 1 , a 2 , ...a n , where “1” indicates that the corresponding training sample includes the individual attribute, and “0” indicates that the corresponding training sample does not include the individual attribute. Accordingly, an attribute vector of (1, 0, 1, 1...) may be generated for the training sample of Bob, and a corresponding tag is “1” .
  • An attribute vector of (1, 1, 0, 1...) may be generated for the training sample of Tom, and a corresponding tag is “0” .
  • An attribute vector of (0, 1, 1, 1%) may be generated for the training sample of Jack, and a corresponding tag is “1” .
  • a collection of an attribute vector and a tag generated for each training sample in Table 1 may form the training data set for the transition.
  • a probability of achieving the transition e.g., the first job position being navigated to the second job position, may be calculated.
  • the probability may be determined based on the number of training samples identifying both the first job position and the second job position and the number of training samples identifying the first job position. For example, for a transition from “College Student” to “Support Engineer Intern” , if the number of training samples identifying both “College Student” and “Support Engineer Intern” is 400, and the number of training samples identifying “College Student” is 1000, then the probability may be calculated as 40%.
  • machine learning may be performed so as to establish a goal navigation advice model.
  • Various machine learning models such as, a decision tree model or a linear fitting model, may be adopted for establishing the goal navigation advice model.
  • machine learning through decision tree may be performed based on the training data set.
  • the decision tree may be used for determining weights of the individual attributes in the training data set.
  • Each internal node of the decision tree represents a test on an attribute, each branch represents the outcome of the test, and each leaf node represents a class label.
  • the paths from root node to leaf node represent classification rules.
  • the decision tree may determine a first individual attribute with the best classifying capability as the root node. Then, for a branch having the first individual attribute, the decision tree may determine a second individual attribute with the best classifying capability as an internal node of this branch. Similarly, for a branch not having the first individual attribute, the decision tree may determine a third individual attribute with the best classifying capability as an internal node of this branch. In the same way, the decision tree may traverse all individual attributes, and determine all internal nodes.
  • the decision tree may calculate a statistic value, such as, information gain, for each individual attribute during determining the root node and the internal nodes.
  • a statistic value such as, information gain
  • a higher information gain may indicate a better classifying capability.
  • the information gain calculated for an individual attribute may be used as the weight of the individual attribute.
  • a training data set for this transition may be provided to the decision tree algorithm.
  • the decision tree algorithm may calculate information gains for the individual attributes in the training data set. For example, information gain for an individual attribute “Network Security” may be calculated as “0.462” and used as its weight, information gain for an individual attribute “Troubleshooting” may be calculated as “0.458” and used as its weight, information gain for an individual attribute “Windows Server” may be calculated as “0.315” and used as its weight, and so on.
  • machine learning through linear fitting may be performed based on the training data set.
  • the linear fitting may be used for determining weights of the individual attributes in the training data set.
  • f i denotes a tag value of the i-th training sample
  • (a i1 , a i2 , ...a in ) denotes an attribute vector of the i-th training sample
  • (w 1 , w 2 , ...w n ) denotes weights of individual attributes
  • n is the dimension of the attribute vector.
  • Equation (1) Through inputting, into Equation (1) , an attribute vector and a tag value of each training sample in the training data set, a set of linear equations with variables w 1 , w 2 , ...w n may be obtained. Then, the variables w 1 , w 2 , ...w n may be solved, thus obtaining the weights of the individual attributes.
  • either of the machine learning through decision tree at 520 and the machine learning through linear fitting at 522 may be applied for determining the weights of the individual attributes. Any other machine learning models, such as, regression tree, may also be adopted for determining the weights of the individual attributes.
  • a rule associated with the transition may be formed.
  • the rule may be formed by one or more of the probability of the transition, individual attributes associated with the first job position of the transition, and weights of the individual attributes.
  • the method 500 may return to 512 to determine whether there is a next transition. If it is determined at 512 that there is no next transition, the method 500 proceeds to 526.
  • a goal navigation advice model may be formed. For example, rules associated with all the transitions identified at 510 may be gathered to form the goal navigation advice model. Then, the method 500 ends at 528.
  • the goal navigation advice model generated by the method 500 includes information on transitions between job positions, e.g., rules associated with the transitions.
  • the information on transitions may be presented to users based on the goal navigation advice model. For example, if a user is querying how to achieve a target job position, the embodiments of the present disclosure may utilize the goal navigation advice model to provide one or more paths from a start job position to the target job position, and information on each transition in the one or more paths may also be presented.
  • the goal navigation advice model generated by the method 500 may be dynamically updated during the goal navigation system provides career goal navigation. For example, if a status of a user, who queries how to achieve a career goal, is changed, e.g., the user has achieved a transition in the path suggested by the goal navigation advice, a training sample based on this transition may be provided to the process of the method 500, and thus the goal navigation advice model may be updated in a consideration of this training sample.
  • the goal navigation advice model may include rules associated with transitions, it should be appreciated that the goal navigation advice model may also include any other information useful for providing advice to users. In some implementations, the goal navigation advice model may include rules associated with job positions.
  • FIG. 6 is a flowchart of an exemplary process 600 for generating rules associated with job positions according to an embodiment.
  • the method 600 may be performed by the goal navigation system for generating rules associated with job positions based on numerous training samples through data mining and machine learning.
  • a plurality of job positions may be identified from training samples.
  • the training samples herein may be the same as the training samples obtained at 504 in FIG. 5.
  • a training data set may be determined for a job position of the plurality of job positions.
  • the training data set may include attribute vectors and tags for the training samples.
  • the attribute vectors may be generated based on the individual attributes of the training samples.
  • the tags may be generated based on whether the job position is identified in the training samples. For example, for the i-th training sample, an attribute vector (a i1 , a i2 , ...a in ) may be generated based on the individual attributes of the i-th training sample, where the attribute vector is a vector with n dimension.
  • a collection of an attribute vector and a tag generated for each training sample may form the training data set for the job position.
  • machine learning may be performed so as to generate a rule associated with the job position.
  • Various machine learning models such as, a decision tree model or a linear fitting model, may be adopted for generating the rule associated with the job position.
  • machine learning through decision tree may be performed based on the training data set.
  • the decision tree may be used for determining weights of the individual attributes in the training data set.
  • the operation at 606 may be performed in a similar way as the operation 520 in FIG. 5.
  • machine learning through linear fitting may be performed based on the training data set.
  • the linear fitting may be used for determining weights of the individual attributes in the training data set.
  • the operation at 608 may be performed in a similar way as the operation 522 in FIG. 5.
  • either of the machine learning through decision tree at 606 and the machine learning through linear fitting at 608 may be applied for determining the weights of the individual attributes.
  • Any other machine learning models, such as, regression tree, may also be adopted for determining the weights of the individual attributes.
  • a rule associated with the job position may be formed.
  • the rule may be formed by individual attributes associated with the job position and weights of the individual attributes.
  • the process 600 may repeat the operations of 604 to 610 until rules associated with all job positions are formed.
  • the process 600 for generating rules associated with job positions may also be viewed as identifying at least one individual attribute to distinguish a job position from other job positions.
  • the rules associated with job positions generated by the process 600 may be further added into the goal navigation advice model.
  • the information on job positions may be presented to users. For example, if a user is querying how to achieve a target job position, the embodiments of the present disclosure may utilize the goal navigation advice model to provide one or more paths from a start job position to the target job position, and information on each job position in the one or more paths may also be presented.
  • job positions are extracted from the training sample, it should be appreciated that the job positions may also be predetermined empirically.
  • the goal navigation advice model may also include relevant courses associated with the individual attributes. For example, a hyperlink to a course may be associated with a corresponding individual attribute. Through learning this course, a user can possess the individual attribute, such as, master a skill.
  • the goal navigation system may be applied for providing career goal navigation.
  • the goal navigation system may also be applied in any other approaches.
  • the goal navigation system may be applied for providing personal finance goal navigation.
  • a personal finance goal navigation service may help a user to make a plan that enables the user to achieve a financial goal, such as, a financial position, to better manage his money, or to better prepare for retirements.
  • a financial goal such as, a financial position
  • the personal finance goal navigation service may provide the user with a path directed to the financial position, wherein the path may include one or more nodes, and information associated with the nodes may be shown.
  • personal finance information of numerous users may be collected, which may include various items, such as, salary, saving, investment, income to spending ratio, budget, long-term goal, etc.
  • the personal finance information of the users may be viewed as training samples, the items in the personal finance information may be viewed as individual attributes, and the financial positions may be viewed as nodes.
  • these training samples and individual attributes may be used for generating a goal navigation advice model in a similar way as mentioned above, and accordingly, the goal navigation system may provide personal finance goal navigation.
  • the goal navigation system may be applied for providing game role goal navigation.
  • a game role goal navigation service may provide players of role-playing games with advice on how to grow from a current role to a target role, such as, a game role level. For example, if a player is querying how to achieve a game role level from the current game role level, the game role goal navigation service may provide the player with a path directed to the game role level, wherein the path may include one or more nodes, such as, intermediate game role levels, and information associated with the nodes may be shown.
  • game role profiles of numerous players may be collected, which may include various items, such as, skill, equipment, experience, time spent, etc.
  • the game role profiles of the players may be viewed as training samples, the items in the game role profiles may be viewed as individual attributes, and the game role levels may be viewed as nodes.
  • these training samples and individual attributes may be used for generating a goal navigation advice model in a similar way as mentioned above, and accordingly, the goal navigation system may provide game role goal navigation.
  • the goal navigation system may be applied for providing romantic partnership goal navigation.
  • a romantic partnership goal navigation service may provide users with advice on how to pursue a target romantic partner. For example, if a user is querying how to possess a romantic partnership with a target romantic partner who has requirements on a desired personal condition of a romantic partner, the romantic partnership goal navigation service may provide the user with a path directed to the desired personal condition required by the target romantic partner.
  • individual profiles of numerous users may be collected, which may include various items, such as, personality, social activity, hobby, education, appearance and individual wealth, etc. of the users.
  • the individual profiles of the users may be viewed as training samples, the items in the individual profiles may be viewed as individual attributes, and personal conditions for romantic partnership may be viewed as nodes.
  • these training samples and individual attributes may be used for generating a goal navigation advice model in a similar way as mentioned above, and accordingly, the goal navigation system may provide romantic partnership goal navigation.
  • the goal navigation system may be applied for providing fitness goal navigation.
  • a fitness goal navigation service may provide users with advice on how to achieve a fitness goal. For example, if a user is querying how to achieve a target fitness level from the current fitness level, the fitness goal navigation service may provide the user with a path directed to the target fitness level, wherein the path may include one or more nodes, such as, intermediate fitness levels, and information associated with the nodes may be shown.
  • personal fitness information of numerous users may be collected, which may include various items, such as, Basal Metabolic Rate (BMR) , Body Mass Index (BMI) , body fat ratio, muscle mass, body moisture rate, etc.
  • BMR Basal Metabolic Rate
  • BMI Body Mass Index
  • the personal fitness information of the users may be viewed as training samples, the items in the personal fitness information may be viewed as individual attributes, and the fitness levels may be viewed as nodes.
  • these training samples and individual attributes may be used for generating a goal navigation advice model in a similar way as mentioned above, and accordingly, the goal navigation system may provide fitness goal navigation.
  • the goal navigation system may be applied for providing personal health goal navigation.
  • a personal health goal navigation service may provide users with advice on how to achieve a personal health goal. For example, if a user is querying how to achieve a target personal health level from the current personal health level, the personal health goal navigation service may provide the user with a path directed to the target personal health level, wherein the path may include one or more nodes, such as, intermediate personal health levels, and information associated with the nodes may be shown.
  • personal health information of numerous users may be collected, which may include various items, such as, blood pressure, blood fat, blood sugar, etc.
  • the personal health information of the users may be viewed as training samples, the items in the personal health information may be viewed as individual attributes, and the personal health levels may be viewed as nodes.
  • these training samples and individual attributes may be used for generating a goal navigation advice model in a similar way as mentioned above, and accordingly, the goal navigation system may provide personal health goal navigation.
  • FIG. 7 is a flowchart of an exemplary method 700 for navigating among a plurality of nodes according to an embodiment.
  • the method 700 may be used for a goal navigation system.
  • a plurality of training samples may be obtained.
  • Each of the training samples may include at least one individual attribute.
  • At 704 at least weights of individual attributes associated with a first node may be calculated based on the training samples associated with the first node with a machine learning model, the weights being used to navigate the first node to a second node.
  • the individual attributes may comprise at least one of the following groups: a first group including at least one of skill, education and work experience; a second group including at least one of salary, saving, investment, income to spending ratio, budget and long-term goal; a third group including at least one of skill, equipment, experience and time spent; a fourth group including at least one of personality, social activity, hobby, education, appearance and individual wealth; a fifth group including at least one of BMR, BMI, body fat ratio, muscle mass and body moisture rate; and a six group including at least one of blood pressure, blood fat and blood sugar.
  • a first group including at least one of skill, education and work experience
  • a second group including at least one of salary, saving, investment, income to spending ratio, budget and long-term goal
  • a third group including at least one of skill, equipment, experience and time spent
  • a fourth group including at least one of personality, social activity, hobby, education, appearance and individual wealth
  • a fifth group including at least one of BMR, BMI, body fat ratio, muscle mass and
  • the machine learning model may be any one of a decision tree model and a linear fitting model.
  • each of the training samples may include an individual profile
  • the method 700 may further comprise extracting the at least one individual attribute from the individual profile by LDA.
  • any one of the plurality of nodes may be one of: job position, finance position, game role level, personal condition for romantic partnership, fitness level and personal health level.
  • the method may further comprise normalizing description of the individual attributes or description of the nodes.
  • the normalization may comprise generating semantic vectors for the description with CDSSM tool; and clustering the semantic vectors with K-means clustering algorithm.
  • the method 700 may further comprise identifying at least one individual attribute to distinguish one node from other nodes based on a machine learning model.
  • the method 700 may further comprise calculating a probability of the first node being navigated to the second node based on the training samples navigated from the first node to the second node and the training samples associated with the first node.
  • FIG. 8 is a flowchart of an exemplary method 800 according to an embodiment.
  • the method 800 may be used for a goal navigation system.
  • a start node and a target node among a plurality of nodes may be determined.
  • Each node may be associated with at least one training sample.
  • Each training sample may be associated with at least one individual attribute.
  • One or more weights in connection with individual attributes associated with a first node may be calculated based on training samples associated with the first node with a machine learning model and configured to navigate the first node to a second node.
  • information on at least one path navigated from the start node to the target node may be provided based on the weights in connection with the individual attributes associated with at least one node placed on the path.
  • the method 800 may further comprise presenting the information based on the node placed on the path.
  • the information may include a predetermined number of the individual attributes and the weights thereof.
  • the individual attributes may comprise at least one of the following groups: a first group including at least one of skill, education and work experience; a second group including at least one of salary, saving, investment, income to spending ratio, budget and long-term goal; a third group including at least one of skill, equipment, experience and time spent; a fourth group including at least one of personality, social activity, hobby, education, appearance and individual wealth; a fifth group including at least one of BMR, BMI, body fat ratio, muscle mass and body moisture rate; and a six group including at least one of blood pressure, blood fat and blood sugar.
  • a first group including at least one of skill, education and work experience
  • a second group including at least one of salary, saving, investment, income to spending ratio, budget and long-term goal
  • a third group including at least one of skill, equipment, experience and time spent
  • a fourth group including at least one of personality, social activity, hobby, education, appearance and individual wealth
  • a fifth group including at least one of BMR, BMI, body fat ratio, muscle mass and
  • any one of the plurality of nodes may be one of: job position, finance position, game role level, personal condition for romantic partnership, fitness level and personal health level.
  • description of the individual attributes or description of the nodes may be normalized by using K-means clustering algorithm with CDSSM tool.
  • each of the training samples may include an individual profile, and the at least one individual attribute may be extracted from the individual profile by LDA.
  • FIG. 9 illustrates an exemplary apparatus 900 for navigating among a plurality of nodes according to an embodiment.
  • the apparatus 900 may comprise an obtaining module 902.
  • the obtaining module 902 is for obtaining a plurality of training samples, each of the training samples including at least one individual attribute.
  • the apparatus 900 may comprise a calculating module 904.
  • the calculating module 904 is for calculating at least weights of individual attributes associated with a first node based on the training samples associated with the first node with a machine learning model, the weights being used to navigate the first node to a second node.
  • each of the training samples may include an individual profile
  • the apparatus 900 may further comprise an extracting module which is for extracting the at least one individual attribute from the individual profile by LDA.
  • the apparatus 900 may further comprise a normalizing module which is for normalizing description of the individual attributes or description of the nodes.
  • the apparatus 900 may also comprise any other modules configured for performing any operations of the methods for providing goal navigation according to the embodiments of the present disclosure as mentioned above.
  • FIG. 10 illustrates an exemplary apparatus 1000 according to an embodiment.
  • the apparatus 1000 may be used for a goal navigation system.
  • the apparatus 1000 may comprise a determining module 1002.
  • the determining module 1002 is for determining a start node and a target node among a plurality of nodes.
  • Each node may be associated with at least one training sample.
  • Each training sample may be associated with at least one individual attribute.
  • One or more weights in connection with individual attributes associated with a first node may be calculated based on training samples associated with the first node with a machine learning model and configured to navigate the first node to a second node.
  • the apparatus 1000 may comprise a providing module 1004.
  • the providing module 1004 is for providing information on at least one path navigated from the start node to the target node based on the weights in connection with the individual attributes associated with at least one node placed on the path.
  • apparatus 1000 may also comprise any other modules configured for performing any operations of the methods for providing goal navigation according to the embodiments of the present disclosure as mentioned above.
  • FIG. 11 illustrates an exemplary system 1100 according to an embodiment.
  • the system 1100 may comprise a computer readable medium 1102.
  • the computer readable medium 1102 may include a plurality of nodes. Each node may be associated with at least one training sample. Each training sample may be associated with at least one individual attribute.
  • the computer readable medium 1102 may include one or more weights in connection with individual attributes associated with a first node, the weights calculated based on training samples associated with the first node with a machine learning model, and configured to navigate the first node to a second node.
  • the system 1100 may comprise a processor 1104.
  • the processor 1104 may be configured to determine a start node and a target node, and provide information on at least one path navigated from the start node to the target node based on the weights in connection with the individual attributes associated with at least one node placed on the path.
  • the processor 1104 may be configured to present the information based on the node placed on the path, wherein the information includes a predetermined number of the individual attributes and the weights thereof.
  • the individual attributes may comprise at least one of the following groups: a first group including at least one of skill, education and work experience; a second group including at least one of salary, saving, investment, income to spending ratio, budget and long-term goal; a third group including at least one of skill, equipment, experience and time spent; a fourth group including at least one of personality, social activity, hobby, education, appearance and individual wealth; a fifth group including at least one of BMR, BMI, body fat ratio, muscle mass and body moisture rate; and a six group including at least one of blood pressure, blood fat and blood sugar.
  • a first group including at least one of skill, education and work experience
  • a second group including at least one of salary, saving, investment, income to spending ratio, budget and long-term goal
  • a third group including at least one of skill, equipment, experience and time spent
  • a fourth group including at least one of personality, social activity, hobby, education, appearance and individual wealth
  • a fifth group including at least one of BMR, BMI, body fat ratio, muscle mass and
  • any one of the plurality of nodes may be one of job position, finance position, game role level, personal condition for romantic partnership, fitness level and personal health level.
  • each of the training samples may include an individual profile, and the at least one individual attribute is extracted from the individual profile by LDA.
  • the processor 1104 may be configured to normalize description of the individual attributes or description of the nodes.
  • the description of the individual attributes or the description of the nodes may be normalized by: generating semantic vectors for the description with CDSSM tool; and clustering the semantic vectors with K-means clustering algorithm.
  • the processor 1104 may be configured to identify at least one individual attribute to distinguish one node from other nodes based on a machine learning model.
  • an apparatus may comprise a plurality of nodes. Each node may be associated with at least one training sample. Each training sample may be associated with at least one individual attribute. The apparatus may further comprise one or more weights in connection with individual attributes associated with a first node, the weights calculated based on training samples associated with the first node with a machine learning model and configured to navigate the first node to a second node.
  • the nodes and the weights may be stored in a computer readable medium. In some implementations, the nodes and the weights may be stored in cloud storage, and may be retrieved when providing goal navigation by the apparatus.
  • the individual attributes may comprise at least one of the following groups: a first group including at least one of skill, education and work experience; a second group including at least one of salary, saving, investment, income to spending ratio, budget and long-term goal; a third group including at least one of skill, equipment, experience and time spent; a fourth group including at least one of personality, social activity, hobby, education, appearance and individual wealth; a fifth group including at least one of BMR, BMI, body fat ratio, muscle mass and body moisture rate; and a six group including at least one of blood pressure, blood fat and blood sugar.
  • a first group including at least one of skill, education and work experience
  • a second group including at least one of salary, saving, investment, income to spending ratio, budget and long-term goal
  • a third group including at least one of skill, equipment, experience and time spent
  • a fourth group including at least one of personality, social activity, hobby, education, appearance and individual wealth
  • a fifth group including at least one of BMR, BMI, body fat ratio, muscle mass and
  • the machine learning model may be any one of a decision tree model and a linear fitting model.
  • each of the training samples may include an individual profile, wherein the at least one individual attribute is extracted from the individual profile by LDA.
  • any one of the plurality of nodes may be one of job position, finance position, game role level, personal condition for romantic partnership, fitness level and personal health level.
  • description of the individual attributes or description of the nodes may be normalized by using K-means clustering algorithm with CDSSM tool.
  • the apparatus may include a probability of the first node being navigated to the second node, wherein the probability may be calculated based on the training samples navigated from the first node to the second node and the training samples associated with the first node.
  • FIG. 12 illustrates an exemplary system 1200 for providing goal navigation according to an embodiment.
  • the system 1200 may comprise one or more processors 1202.
  • the system 1200 may further comprise a memory 1204 that is connected with the one or more processors 1202.
  • the memory 1204 may store computer-executable instructions that, when executed, cause the one or more processors 1202 to perform any operations of the methods for providing goal navigation according to the embodiments of the present disclosure as mentioned above.
  • the embodiments of the present disclosure may be embodied in a non-transitory computer-readable medium.
  • the non-transitory computer-readable medium may comprise instructions that, when executed, cause one or more processors to perform any operations of the methods for providing goal navigation according to the embodiments of the present disclosure as mentioned above.
  • modules in the apparatuses described above may be implemented in various approaches. These modules may be implemented as hardware, software, or a combination thereof. Moreover, any of these modules may be further functionally divided into sub-modules or combined together.
  • processors have been described in connection with various apparatuses and methods. These processors may be implemented using electronic hardware, computer software, or any combination thereof. Whether such processors are implemented as hardware or software will depend upon the particular application and overall design constraints imposed on the system.
  • a processor, any portion of a processor, or any combination of processors presented in the present disclosure may be implemented with a microprocessor, microcontroller, digital signal processor (DSP) , a field-programmable gate array (FPGA) , a programmable logic device (PLD) , a state machine, gated logic, discrete hardware circuits, and other suitable processing components configured to perform the various functions described throughout the present disclosure.
  • DSP digital signal processor
  • FPGA field-programmable gate array
  • PLD programmable logic device
  • a state machine gated logic, discrete hardware circuits, and other suitable processing components configured to perform the various functions described throughout the present disclosure.
  • the functionality of a processor, any portion of a processor, or any combination of processors presented in the present disclosure may be
  • a computer-readable medium may include, by way of example, memory such as a magnetic storage device (e.g., hard disk, floppy disk, magnetic strip) , an optical disk, a smart card, a flash memory device, random access memory (RAM) , read only memory (ROM) , programmable ROM (PROM) , erasable PROM (EPROM) , electrically erasable PROM (EEPROM) , a register, or a removable disk.
  • RAM random access memory
  • ROM read only memory
  • PROM programmable ROM
  • EPROM erasable PROM
  • EEPROM electrically erasable PROM

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Evolutionary Computation (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

L'invention concerne un procédé, un appareil et un système permettant de naviguer vers un objectif. Un nœud de départ et un nœud cible peuvent être déterminés parmi une pluralité de noeuds. Chaque nœud est associé à au moins un échantillon d'apprentissage. Chaque échantillon d'apprentissage est associé à au moins un attribut individuel. Un ou plusieurs poids en rapport avec des attributs individuels associés à un premier nœud peuvent être calculés d'après des échantillons d'apprentissage associés au premier nœud à l'aide d'un modèle d'apprentissage automatique. Les poids peuvent être configurés pour naviguer depuis le premier nœud vers le second nœud. Des informations concernant au moins un trajet parcouru entre le nœud de départ et le nœud cible peuvent être fournies d'après les poids en rapport avec les attributs individuels associés à au moins un nœud placé sur le trajet.
PCT/CN2016/113741 2016-12-30 2016-12-30 Navigation vers un objectif Ceased WO2018120123A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201680080880.5A CN108604247A (zh) 2016-12-30 2016-12-30 目标导航
PCT/CN2016/113741 WO2018120123A1 (fr) 2016-12-30 2016-12-30 Navigation vers un objectif

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/113741 WO2018120123A1 (fr) 2016-12-30 2016-12-30 Navigation vers un objectif

Publications (1)

Publication Number Publication Date
WO2018120123A1 true WO2018120123A1 (fr) 2018-07-05

Family

ID=62707638

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/113741 Ceased WO2018120123A1 (fr) 2016-12-30 2016-12-30 Navigation vers un objectif

Country Status (2)

Country Link
CN (1) CN108604247A (fr)
WO (1) WO2018120123A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111163425A (zh) * 2020-01-02 2020-05-15 中国平安财产保险股份有限公司 基于lbs轨迹的身份识别方法、电子装置及可读存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090070316A1 (en) * 2007-09-07 2009-03-12 Christian Beauchesne Web-based succession planning
US20090164311A1 (en) * 2007-12-19 2009-06-25 Microsoft Corporation Human resource management system
CN102968698A (zh) * 2012-12-07 2013-03-13 深圳市智维通达科技有限公司 一种企业员工学习数据模型的建立方法及系统
CN104834668A (zh) * 2015-03-13 2015-08-12 浙江奇道网络科技有限公司 基于知识库的职位推荐系统
CN105426435A (zh) * 2015-11-04 2016-03-23 深圳市前海七号网络科技有限公司 专业人才数据处理方法及服务器

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1808778A1 (fr) * 2005-12-07 2007-07-18 Sap Ag Procédé de navigation dans une application de portail, système de navigation dans un système de portail, terminal d'utilisateur et support de stockage lisible par ordinateur
CN102930370A (zh) * 2012-09-27 2013-02-13 杜继俊 招聘信息处理方法及系统
CN106095931A (zh) * 2016-06-12 2016-11-09 杭州简才网络信息科技有限公司 一种求职者和招聘单位之间的匹配方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090070316A1 (en) * 2007-09-07 2009-03-12 Christian Beauchesne Web-based succession planning
US20090164311A1 (en) * 2007-12-19 2009-06-25 Microsoft Corporation Human resource management system
CN102968698A (zh) * 2012-12-07 2013-03-13 深圳市智维通达科技有限公司 一种企业员工学习数据模型的建立方法及系统
CN104834668A (zh) * 2015-03-13 2015-08-12 浙江奇道网络科技有限公司 基于知识库的职位推荐系统
CN105426435A (zh) * 2015-11-04 2016-03-23 深圳市前海七号网络科技有限公司 专业人才数据处理方法及服务器

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111163425A (zh) * 2020-01-02 2020-05-15 中国平安财产保险股份有限公司 基于lbs轨迹的身份识别方法、电子装置及可读存储介质

Also Published As

Publication number Publication date
CN108604247A (zh) 2018-09-28

Similar Documents

Publication Publication Date Title
US11537820B2 (en) Method and system for generating and correcting classification models
CA3129745C (fr) Systeme de reseau neuronal de classification de texte
US10460128B2 (en) Collection flow for confidential data
KR102265573B1 (ko) 인공지능 기반 입시 수학 학습 커리큘럼 재구성 방법 및 시스템
US9183285B1 (en) Data clustering system and methods
CN108255921A (zh) 确定并显示推荐内容的相关解释的系统
US20170236095A1 (en) System and method for generating a career path
US11205155B2 (en) Data selection based on career transition embeddings
US12008047B2 (en) Providing an object-based response to a natural language query
US20150287051A1 (en) System and method for identifying growing companies and monitoring growth using non-obvious parameters
US11275994B2 (en) Unstructured key definitions for optimal performance
CN113590792B (zh) 用户问题的处理方法、装置和服务器
US20200065422A1 (en) Document Entity Linking on Online Social Networks
KR20170027576A (ko) 연구 이력 매칭 기반의 연구자 추천 장치 및 방법
Nassiri et al. Iterative multiple imputation: A framework to determine the number of imputed datasets
CN110929169A (zh) 基于改进Canopy聚类协同过滤算法的职位推荐方法
EP2613275A1 (fr) Dispositif de recherche, procédé de recherche, programme de recherche et support de mémoire lisible par un ordinateur destiné à contenir un programme de recherche
Eskandari et al. Predicting best answer using sentiment analysis in community question answering systems
CN113392309B (zh) 一种数据处理方法、装置、设备及存储介质
WO2018120123A1 (fr) Navigation vers un objectif
JP2021056857A (ja) 情報処理システム、情報処理方法、プログラム
Lei An analytical model of college students’ mental health education based on the clustering algorithm
Zhu et al. Adaptation of cultural norms after merger and acquisition based on the heterogeneous agent-based relative-agreement model
US20200118082A1 (en) Presentation of Credible and Relevant Job History
KR102828656B1 (ko) 경쟁사 분석 정보를 제공하는 방법 및 전자 장치

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16925087

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16925087

Country of ref document: EP

Kind code of ref document: A1