CN110032681B - Resume content-based job recommendation method - Google Patents
Resume content-based job recommendation method Download PDFInfo
- Publication number
- CN110032681B CN110032681B CN201910308204.4A CN201910308204A CN110032681B CN 110032681 B CN110032681 B CN 110032681B CN 201910308204 A CN201910308204 A CN 201910308204A CN 110032681 B CN110032681 B CN 110032681B
- Authority
- CN
- China
- Prior art keywords
- information
- resume
- salary
- component
- category
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/105—Human resources
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Operations Research (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a resume content-based job recommendation method, which comprises the following steps: acquiring the recruitment information of each recruitment position, and generating a position vector P for each recruitment position according to the recruitment informationiAll the position vectors form a position vector set P; obtaining the resume information in each user resume, and generating a resume vector R for each user resume according to the resume informationjAll resume vectors form a resume vector set R; traversing the position vectors in the position vector set according to the resume vectors of the target resume, eliminating the position vectors by adopting a vote rejection method, and generating a candidate position vector set by the rest position vectors; and calculating the similarity between the resume information corresponding to the target resume and the recruitment information corresponding to each position vector of the candidate position vector set, generating a position recommendation list, and recommending the position recommendation list to the job seeker. The method has the advantages that the pre-elimination is carried out by adopting a vote rejection method, the calculation amount of the similarity algorithm is effectively reduced, and the recommendation efficiency is improved.
Description
Technical Field
The invention relates to the technical field of network recruitment. More particularly, the present invention relates to a resume content-based job recommendation method.
Background
With the popularization of the internet in China, the network recruitment field of China shows a vigorous development. The rapidly increasing scale of network recruitment brings more selection opportunities to job seekers and increases the cost of job seekers for searching for required positions. Therefore, position recommendation systems are introduced into various famous network recruitment websites to different degrees, and help job seekers decide to deliver positions. The current focus of the research of the network recruitment recommendation technology is Collaborative exclusion recommendation technology (Collaborative Filtering), and a Collaborative exclusion recommendation algorithm is mainly based on a group of job seekers with the same interests for recommendation. The method has the advantages that the characteristic attribute of the object does not need to be analyzed, no special requirement is required for the recommended object, and better recommendation quality is shown when the data density reaches a certain degree. However, this method has the following significant disadvantages: the job seeker recommendation method based on the job seeker data has the advantages that firstly, the data sparsity problem (sparse publishing), namely, the recommendation quality is low when the job seeker scores less data, and particularly the problem is prominent when the job seeker data is very much and the job seeker data is less; and when no score data exists when a new job is added, the collaborative elimination technology cannot recommend the job to the job seeker, the problem is called a new job problem, similarly, a new job seeker does not score any data, the collaborative elimination technology cannot recommend the job to the job seeker, and the problem is called a new job seeker problem. The new job question and the new job seeker question are both referred to as cold-start questions (cold-start programs). Secondly, the scalability problem (scalability) is that the calculation time of the collaborative elimination algorithm is increased in a linear relationship with the increase of the numbers of job seekers and positions, so that the scalability is poor and the recommendation requirement of a large recruitment website cannot be met.
Disclosure of Invention
An object of the present invention is to solve at least the above problems and to provide at least the advantages described later.
The invention also aims to provide a resume content-based job recommendation method, which adopts a vote rejection method for pre-elimination, effectively reduces the calculation amount of a similarity algorithm and improves the recommendation efficiency.
To achieve these objects and other advantages in accordance with the present invention, there is provided a resume content-based job recommendation method, comprising the steps of:
acquiring the recruitment information of each recruitment position, and generating a position vector P for each recruitment position according to the recruitment informationiAll position vectors form a position vector set P, P ═ P1、P2、…、Pi、…、Pm)；
Obtaining the resume information in each user resume, and generating a resume vector R for each user resume according to the resume informationjAll resume vectors form a resume vector set R, R ═ R (R)1、R2、…、Rj、…、Rm) The job position information of each recruitment position and the resume information in each user resume respectively comprise keywords for representing the position information, the work place information, the salary information, the position type information, the academic information, the work experience information and the industry information;
selecting a user resume as a target resume, traversing a position vector centralized position vector according to the resume vector of the target resume, eliminating the position vector by adopting a ticket veto method according to position information, workplace information, salary information, position type information and academic information, and generating a candidate position vector set by using the rest position vector;
and calculating the similarity between the resume information corresponding to the target resume and the recruitment information corresponding to each position vector of the candidate position vector set according to the position information, the work place information, the salary information, the academic information, the work experience information and the industry information, screening the first N recruitment positions with the highest similarity to the target resume to generate a position recommendation list, and recommending the position recommendation list to a job seeker.
Preferably, the position information of the recruitment information of each recruitment position comprises a position category major category, a position category minor category and a position name; the industry information comprises major industry and minor industry; the work place information comprises a work place city and a work place area;
the position information of the resume information of each user resume comprises a major category expected to be engaged in the job, a minor category expected to be engaged in the job, a major category of the position, a minor category of the position and a name of the position; the workplace information comprises cities and areas of expected workplaces, living places and areas, and the salary information comprises expected monthly salaries and position monthly salaries; the industry information includes expected pursuit of an industry, an industry category; wherein, the job category major category, the job category minor category, the job name, the job monthly salary and the industry category are all corresponding attributes of the latest job of the job seeker.
Preferably, an input page frame comprising the defined keyword options for inputting the recruitment information and the resume information is respectively established;
the method for acquiring the recruitment information of each recruitment position comprises the following steps: guiding the recruiter to select and input the defined keywords, acquiring the defined keywords and the defined keyword attributes input by the recruiter, and combining to generate the recruitment information corresponding to the recruitment position;
the method for acquiring the resume information in each user resume comprises the following steps: and guiding the job seeker to select and input the defined keywords, acquiring the defined keywords and the defined keyword attributes input by the job seeker, and combining to generate the resume information in the corresponding user resume.
Preferably, when the academic information is input, the corresponding defined keyword options include a basic academic option and a skill academic option, the basic academic option includes junior high school, college, basic subject, master and doctor, and the skill academic option includes middle school, background, MBA and EMBA, wherein the basic academic option corresponding to the recruitment information further includes unlimited basic academic options, and the basic academic option corresponding to the resume information further includes others, and each defined keyword corresponding to the academic information is quantized: not limited to-1, other-0, junior-middle-1, senior-middle-2, major-3, parent-4, major-5, doctor-6, junior-2, junior-3, MBA-5, EMBA-6;
defining a position vector as a terminal B and a resume vector as a terminal C, and specifically, eliminating the position vector according to position information, workplace information, salary information, position type information and academic information by adopting a ticket veto method:
judging whether the component position name of the B terminal is contained in all position name sets covered by the component position information of the C terminal, and if not, excluding;
if yes, judging whether the B-end component work place city is contained in the component of the C-end component work place information, and if not, excluding;
if yes, judging whether the lower salary limit represented by the salary information of the B-end component is smaller than the lower salary limit represented by the expected monthly salary of the C-end component and smaller than the lower salary limit represented by the monthly salary of the position of the C-end component, and if yes, excluding;
if not, judging whether the position type represented by the B-terminal component position type information is the same as the position type represented by the C-terminal component position type information, and if not, excluding;
if yes, judging whether the quantized value corresponding to the component academic aptitude information of the B terminal is equal to-1, and if yes, storing the candidate position vector set;
if not, judging whether the quantized value corresponding to the B-end component academic information-the quantized value corresponding to the C-end component academic information | is less than or equal to 2, if not, excluding, if so, storing in a candidate position vector set.
Preferably, when the work experience information is input, the corresponding predefined keyword option includes: 0. defining the defined keywords corresponding to the work experience information as 1 gear, 2 gear, 3 gear, … gear and 7 gear from top to bottom in sequence for less than 1 year, 1-3 years, 3-5 years, 5-7 years, 7-9 years and more than 10 years;
the similarity between the resume information corresponding to the target resume and the recruitment information corresponding to the position vector in the candidate position vector set is Y (P)i,Rj)，Y(Pi,Rj)＝y1+y2+y3+y4+y5+y6+c，y1、y2、y3、y4、y5、y6Respectively representing the similarity of the target resume and the corresponding recruitment position on the position information, the workplace information, the salary information, the academic information, the work experience information and the industry information, wherein:
y1max＝3.5a，a>0, calculate y1When the value is taken, the calculation is carried out step by step from high to low, and the calculation comprises the following steps:
y is when the category subclass of the B-terminal component position is the same as one of the category subclass of the C-terminal component position and the expected employment subclass, and the name of the B-terminal component position is the same as that of the C-terminal component position1＝3.5a；
Y is the same as one of the category subclass of the B-end component position, the category subclass of the C-end component position and the category subclass of the expected employment, and the name of the B-end component position is the same as that of the C-end component position1＝3a；
Y is the same when the category of the B-side component position is the same as the category of the C-side component position is the same as one of the categories of the expected job jobs and the name of the B-side component position is the same as the name of the C-side component position1＝2.5a；
Y is the same when the category of the B-side component position is the same as one of the category of the C-side component position and one of the categories of the expected engaged position, and the name of the B-side component position is the same as that of the C-side component position1＝2a；
When the B-terminal component position name is the same as the C-terminal component position name, y1＝a；
When none of the above conditions is satisfied, y1＝0.5a；
y2max1.5a, calculate y2The step (2) is calculated from high to low step by step, and comprises the following steps:
when the working information of the B-end component is identical to the expected working information of the C-end component, y2＝1.5a；
When the information of the B-side component is identical to one of the information of the C-side component expected to work, y2＝1.3a；
When the information of the working place of the B-end component is the same as that of the existing residential city of the C-end component, y2＝a；
When none of the above conditions is satisfied, y2＝0；
y3maxCalculate y 2a3The step (2) is calculated from high to low step by step, and comprises the following steps:
when the salary information of the B-side component is an interview or the expected monthly salary of the C-side component is an interview and the monthly salary of the job position is confidential, y3＝a；
When the lower salary limit represented by the B-end component salary information is larger than or equal to the lower salary limit represented by the C-end component expected monthly salary, and the lower salary limit represented by the B-end component salary information is smaller than or equal to the upper salary limit represented by the C-end component expected monthly salary, y is smaller than or equal to the upper salary limit represented by the B-end component salary information3＝2a；
When the lower salary limit represented by the B-end component salary information is larger than or equal to the lower salary limit represented by the C-end component position monthly salary, and the lower salary limit represented by the B-end component salary information is smaller than or equal to the upper salary limit represented by the C-end component position monthly salary, y is smaller than or equal to the upper salary limit represented by the B-end component salary information3＝1.7a；
When the lower salary limit expressed by the | B-end component salary information-the lower salary limit expressed by the expected monthly salary of the C-end component is less than or equal to the upper salary limit expressed by the expected monthly salary of the | C-end component-the lower salary limit expressed by the expected monthly salary of the C-end component, y is3＝1.5a；
When none of the above conditions is satisfied, y3＝0.5a；
y4maxCalculating y as a4The step (2) is calculated from high to low step by step, and comprises the following steps:
when the quantized value corresponding to the B-terminal component basic academic record is equal to the quantized value corresponding to the C-terminal component basic academic record, y4＝a；
When the quantization value corresponding to the B-end component skill calendar is equal to the quantization value corresponding to the C-end component skill calendar, and the quantization value corresponding to the C-end component basic skill calendar-the quantization value corresponding to the B-end component skill calendar is less than or equal to 1, y is less than or equal to 14＝a；
When the B-terminal component skill calendar corresponding quantization value is equal to-1, y4＝0.5a；
When the quantized value corresponding to the C-end component basic academic calendar-the quantized value corresponding to the B-end component skill academic calendar is less than or equal to 1, y40.5 a; when none of the above conditions is satisfied, y4＝0；
When the gear of the C-terminal component work experience information is equal to the gear represented by the B-terminal component work experience information, y5＝2a；
When the gear of the C-side component work experience information is 1 gear and the gear represented by the C-side component work experience information-B-side component work experience information is-1, or when the gear of the C-side component work experience information is not 1 gear and the gear represented by the C-side component work experience information-B-side component work experience information is 1 gear, y is5＝1.5a；
When the gear of the C-side component work experience information is 1 gear and the gear represented by the C-side component work experience information-the gear represented by the B-side component work experience information is-2, or when the gear of the C-side component work experience information is not 1 gear and the gear represented by the C-side component work experience information-the gear represented by the B-side component work experience information is-1, y is5＝a；
When none of the above conditions is satisfied, y5＝0.5a；
y6maxCalculating y as a6The step (2) is calculated from high to low step by step, and comprises the following steps:
when the B-side component major industry is the same as the C-side component desires to engage in one of the industries and the industry category, y is the same6＝a；
When the B-side component major industry is the same as the C-side component desires to engage in one of the industries and one of the industry categories, y6＝0.8a；
When one of the B-side component minor industries is the same as one of the C-side component expected to engage in the industries and the industry category, y is the same6＝0.6a；
When one of the B-side component minor industries is the same as one of the C-side component expected to engage in one of the industries and one of the industry categories, y6＝0.4a；
When none of the above conditions is satisfied, y6＝0；
When y is2≥a、y3≥1.7a、y5Not less than 1.5a and y6When the value is more than or equal to 0.8a, c is 5 a; when y is2≥a、y3Not less than 1.7a and y5When the value is more than or equal to 1.5a, c is 4 a; when y is2A is not less than a and y5When the value is more than or equal to 1.5a, c is 3 a.
Preferably, when the job seeker selects and inputs the defined keywords to complete the creation of the user resume, or when the job seeker changes the resume and completes, the user resume is selected as the most target resume.
Preferably, when a new position vector is generated, the user resume corresponding to all resume vectors in the resume vector set is selected as the target resume.
The invention at least comprises the following beneficial effects:
firstly, respectively establishing input page frames which are used for inputting recruitment information and resume information and comprise defined keyword options, so that the recruitment information and the resume information can be conveniently obtained, meanwhile, matching mapping of the obtained resume information and the recruitment information is realized, and then a vote rejection method and a similarity method are adopted to screen the recruitment position and obtain a position recommendation list; the non-conforming items can be deleted in advance by excluding the position vectors according to position information, workplace information, salary information, position type information and academic information by adopting a ticket veto method, so that the calculation amount of a similarity algorithm is reduced, and the recommendation efficiency is improved;
and secondly, compared with a collaborative elimination recommendation algorithm, the method has the advantages of no cold start problem, good expandability and capability of meeting the recommendation needs of large-scale recruitment websites.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention.
Drawings
Fig. 1 is a schematic flowchart of a resume content-based job recommendation method according to one embodiment of the present invention;
FIG. 2 is a schematic diagram of a decision of a vote rejection method according to one embodiment of the present invention;
fig. 3 is a schematic flow chart of similarity calculation according to one embodiment of the present invention.
Detailed Description
The present invention is further described in detail below with reference to examples so that those skilled in the art can practice the invention with reference to the description.
As shown in fig. 1 to 3, the resume content-based job recommendation method includes the following steps:
step one, an input page frame which is used for inputting recruitment information and comprises a defined keyword option is established, namely a registration template I used by a recruiter when the recruiter publishes the recruitment information on a human resource website is established, the registration template I at least comprises 7 mother information frames which are used for selecting and inputting position information, workplace information, salary information, position type information, academic information, work experience information and industry information, wherein the 7 mother information frames comprise:
the positions are divided into a position category large category, a position category subclass and a position category subclass (namely position names) according to the relation contained in sequence, each position category large category comprises a plurality of position category subclasses, each position category subclass comprises a plurality of position names, a female information frame of the position information comprises three-level options which are respectively the position category large category, the position category subclass and the position names, the three-level options are all sub-options which can be selected independently, and the position information can be determined according to the position category classification of human resource websites such as intelligent joint recruitment and carefree fronthaul, for example:
TABLE 1 job Classification
The defined keyword options corresponding to the parent information frame of the work place information comprise two levels of options which are respectively a work place city (city) and a work place area;
the mother information frame of the salary information comprises two options of a salary lower limit and a salary upper limit, the defined keyword of each option is a money number which is arranged from small to large, the salary lower limit and the salary upper limit form a salary range, the mother information frame of the salary information is a selective filling item, and when a recruiter does not select and input information, or the salary upper limit-salary lower limit is more than or equal to 20k, the mother information frame of the salary information and the default salary information are 'interviews';
the defined keyword options corresponding to the parent information frame of the job position type information comprise: full-time, part-time, practice, campus;
when the academic information is input, a mother information frame of the academic information comprises two options which are a basic academic calendar and a skill academic calendar respectively, wherein the defined keywords of the basic academic calendar comprise unlimited junior middle school, senior high school, major specialty, subject, major and doctor, and the defined keywords of the skill academic calendar comprise middle school, middle school specialty, MBA and EMBA;
the defined keyword options corresponding to the parent information frame of the work experience information comprise: 0 (without limitation), 1 year or less, 1-3 years, 3-5 years, 5-7 years, 7-9 years, 10 years or more, and the working experience is the current time-the first time of participation;
the main information frame of the industry information comprises a main industry sub information frame and a secondary industry sub information frame, wherein defined keyword options of the main industry sub information frame and the secondary industry sub information frame are the same and are all classified options in the industry information;
step two, guiding the recruiter to select and input the defined keywords, acquiring the defined keywords input by the recruiter and the attribute of each defined keyword, and combining to generate the recruitment information corresponding to the recruitment position;
generating a position vector P for each recruitment position according to the recruitment informationi，PiPosition information (position category large category, position category small category, position name), workplace information (city, region), salary information, position type information, academic information, work experience information, industry information (main)Industry, minor industry)), among others, for example: the 'attribute' position information comprises three levels of options, namely three 'sub-attributes', 'sub-attribute' position category large class, 'sub-attribute' position category small class, 'sub-attribute' position name, 'sub-attribute' position category large class, and defined keywords corresponding to the 'sub-attribute' position category large class are<Energy | environmental protection | agriculture | finance>All position vectors form a position vector set P, P ═ P1、P2、…、Pi、…、Pm)；
Step two, establishing an input page frame for inputting resume information and comprising defined keyword options, namely a registration template II used by job seekers when resume filling is carried out on a corresponding human resource website, wherein the registration template II at least comprises 7 mother information frames for selecting and inputting position information, workplace information, salary information, position type information, academic information, work experience information and industry information, wherein,
the main information frame of the position information comprises two sub information frames, wherein one sub information frame comprises two-level options which are respectively a class of expected employment and a class of expected employment, the other sub information frame comprises three-level options which are respectively a class of position category, a class of position category and a position name, the option of the registration template II is expected to be the same as the established keyword option of the class of position category of the option of the registration template I, namely, the classification methods of the job categories are the same, the category of the job expected to be carried out corresponds to the category of the job, and the category of the job expected to be carried out are all multi-options, namely, the keywords which are selected and input corresponding to the major categories expected to be engaged in the job and the minor categories expected to be engaged in the job in the resume of the user comprise at least one keyword;
the parent information frame of the workplace information comprises two child information frames which are respectively expected workplace information and living place information, the expected workplace information comprises two-level options which are respectively expected work cities and areas, the living place information comprises two-level options which are respectively present living cities and areas, and the expected workplace information is a multi-option, namely the expected workplace information in the user resume corresponds to at least one group of selected and input keywords, and the keywords comprise the following steps: shanghai City, Huangpu area, Shanghai city, Pudong New area, and Tianjin, Heping area 3 groups;
the mother information frame of the salary information comprises two sub information frames which are respectively expected monthly salaries and position monthly salaries; the expected monthly salary and the post monthly salary both comprise two options of a salary lower limit and a salary upper limit, wherein a sub information frame of the expected monthly salary and a sub information frame of the post monthly salary are selective filling items, when a job seeker of the sub information frame of the expected monthly salary does not select and input information, default salary information is 'interview', and when a job seeker of the post monthly salary does not select and input information, default salary information is 'confidential';
the mother information frame of the job type information of the registration template II and the mother information frame of the job type information of the registration template I,
when the study calendar information is input, the mother information frame of the study calendar information of the registration template II is the same as that of the study calendar information of the registration template I, except that the basic study calendar has 'unlimited' changed into 'other' in the defined keyword options;
the difference between the parent information frame of the work experience information of the registration template II and the parent information frame of the work experience information of the registration template I is that the defined keyword option of the work experience information is changed from 'unlimited' to '0';
the main information frame of the industry information comprises two sub information frames which are respectively an expected industry and an industry category, the expected industry and the industry category are the same as the defined keyword options of the main industry, the expected industry is a plurality of options, namely, the user resume is provided with at least one keyword which is correspondingly selected and input by the expected industry;
in conclusion, the job category major category, the job category minor category, the job name, the job monthly salary and the industry category are all corresponding attributes of the latest job of the job seeker, and if not, the job category major category, the job category minor category, the job name, the job monthly salary and the industry category can not be filled;
guiding a job seeker to select and input defined keywords, acquiring the defined keywords input by the job seeker and the attributes of each defined keyword, and combining to generate resume information in a corresponding user resume, wherein the attributes comprise mother attributes and child attributes, and the mother attributes are attributes corresponding to 7 mother information frames, namely 'position information, workplace information, salary information, position type information, academic information, work experience information and industry information';
generating a resume vector R for each user resume according to the resume informationj，RjJob information (job desired to be engaged in job major class, job desired to be engaged in job minor class, job category major class, job category minor class, job name), workplace information (desired job city, area, living city, area), salary information (desired monthly salary, job monthly salary), job type information, academic information, job experience information, industry information (desired to be engaged in industry, industry category)), where, for example: the 'mother attribute' position information comprises two child information boxes, wherein the two child information boxes correspond to 5 options, namely 5 'child attributes', 'position category class' expected to be engaged in the position category, 'child attributes', 'position category class,' child attributes 'position name' and 'child attributes' position name, and the 'position name' corresponds to a defined keyword<Environmental protection technical engineer>All resume vectors form a resume vector set R, R ═ R (R)1、R2、…、Rj、…、Rm)；
Step three, establishing a mapping relation between each component in the resume vector and each component in the position vector according to the same attribute;
step four, selecting a user resume as a target resume, traversing the position vector in the position vector set according to the resume vector of the target resume, eliminating the position vector by adopting a ticket veto method according to the position information, the workplace information, the salary information, the position type information and the academic information, and generating a candidate position vector set by using the rest position vector;
firstly, defining a position vector as a terminal B and a resume vector as a terminal C, and quantizing each defined keyword corresponding to the academic information to obtain: if the job vector is excluded by applying a vote method according to the job information, the workplace information, the salary information, the job type information and the academic information, without being limited to-1, other 0, junior-middle 1, senior-middle 2, major 3, the subject 4, major 5, doctor 6, junior 2, junior 3, MBA 5 and EMBA 6, the method of excluding the job vector by applying the vote method is as follows:
judging whether the component job names of the B terminal are contained in all job name sets covered by the component job information of the C terminal, and if not, excluding, wherein all the job name sets covered by the component job information of the C terminal comprise a job name I (all job names contained in subordinate categories of job categories expected to be engaged), a job name II (all job names contained in subordinate categories expected to be engaged), a job name III (all job names contained in subordinate categories of job categories major), a job name IV (all job names contained in subordinate categories of job categories minor), and a job name, wherein the C terminal represents components of the job information and further comprises deduplication processing before comparison and judgment with the B terminal, and due to other reasons that the job names of the job categories expected to be engaged may be equal to those of job categories, the job name data are duplicated, and duplicate data are deleted to perform deduplication processing;
if yes, judging whether the B-end component work place city is included in the component of the C-end component work place information, if not, excluding, wherein the corresponding keywords of the work place city are only considered to the city level one, for example, when the corresponding keywords of the B-end component work place information are 'Shanghai City, Huangpu district', the corresponding keywords of the C-end component work place information are 'Shanghai City, Changning district', 'Wuhan City, Wuchang district', the Shanghai City, Changning district are expected work cities and areas, the Wuhan City, the Wuchang district are current living cities and areas, if not, excluding;
if so, judging whether the lower salary limit represented by the B-end component salary information is smaller than the lower salary limit represented by the C-end component expected monthly salary and smaller than the lower salary limit represented by the C-end component position monthly salary, if so, excluding, wherein when the B-end component salary information is an 'interview', the B-end component salary is not excluded, when the C-end component expected monthly salary is an 'interview' and the position monthly salary is 'confidential', the C-end component expected monthly salary is an interview or the C-end component position monthly salary is confidential (the situation that the two occur at the same time is not included), and comparing the lower salary limit which is to be encountered definitely (whether the salary is an interview or the confidentiality) with the lower salary limit represented by the B-end component salary information;
if not, judging whether the position type represented by the B-terminal component position type information is the same as the position type represented by the C-terminal component position type information, and if not, excluding;
if yes, judging whether the quantized value corresponding to the component academic aptitude information of the B terminal is equal to-1, and if yes, storing the candidate position vector set;
if not, judging whether the quantized value corresponding to the B-end component academic record information-the quantized value corresponding to the C-end component academic record information is less than or equal to 2, if not, excluding, if so, storing in a candidate position vector set;
secondly, the selection conditions of the target resume (meeting the next point) are as follows:
the conditions are as follows: when the job seeker selects and inputs the defined keywords to complete the creation of the user resume, selecting the user resume as the target resume;
condition two: when the job seeker changes the resume and finishes, selecting the user resume as the target resume;
condition (c): when a new role vector is generated, selecting the user resume corresponding to all resume vectors in the resume vector set as a target resume, and traversing the new role vector in the role vector set according to the resume vector of the target resume;
step five, calculating the similarity between resume information corresponding to a target resume and recruitment information corresponding to each position vector of a candidate position vector set according to position information, workplace information, salary information, academic information, work experience information and industry information, screening the first N recruitment positions with the highest similarity to the target resume to generate a position recommendation list, and recommending the position recommendation list to a job seeker, wherein in the recommendation process, the recommendation can be performed according to time, or according to a certain excitation point, or according to the matching of the time and the excitation point, batch recommendation is set at a certain time interval during the recommendation according to the time, multiple batches of recommendation information are not overlapped, and the excitation point can be set as the time when the user of the job seeker logs in a human resource website;
when the work experience information is input, the corresponding defined keyword options comprise: 0. defining the defined keywords corresponding to the work experience information as 1 gear, 2 gear, 3 gear, … gear and 7 gear from top to bottom in sequence for less than 1 year, 1-3 years, 3-5 years, 5-7 years, 7-9 years and more than 10 years;
the similarity between the resume information corresponding to the target resume and the recruitment information corresponding to the position vector in the candidate position vector set is Y (P)i,Rj)，Y(Pi,Rj)＝y1+y2+y3+y4+y5+y6+c，y1、y2、y3、y4、y5、y6Respectively representing the similarity of the target resume and the corresponding recruitment position on the position information, the workplace information, the salary information, the academic information, the work experience information and the industry information, wherein:
y1max＝3.5a，a>0, calculate y1When the value is taken, the calculation is carried out step by step from high to low, and the calculation comprises the following steps:
y is when the category subclass of the B-terminal component position is the same as one of the category subclass of the C-terminal component position and the expected employment subclass, and the name of the B-terminal component position is the same as that of the C-terminal component position1＝3.5a；
Y is the same as one of the category subclass of the B-end component position, the category subclass of the C-end component position and the category subclass of the expected employment, and the name of the B-end component position is the same as that of the C-end component position1＝3a；
Y is the same when the category of the B-side component position is the same as the category of the C-side component position is the same as one of the categories of the expected job jobs and the name of the B-side component position is the same as the name of the C-side component position1＝2.5a；
Y is the same when the category of the B-side component position is the same as one of the category of the C-side component position and one of the categories of the expected engaged position, and the name of the B-side component position is the same as that of the C-side component position1＝2a；
When the B-terminal component position name is the same as the C-terminal component position name, y1＝a；
When none of the above conditions is satisfied, y1＝0.5a；
y2max1.5a, calculate y2The step (2) is calculated from high to low step by step, and comprises the following steps:
when the working information of the B-side component is identical to that of the C-side component, y21.5a, namely a city and a region of the B terminal are the same as those of the C terminal, and the city and the region of the B terminal are the same as those of the C terminal;
when the information of the B-side component is identical to one of the information of the C-side component expected to work, y21.3a, namely a city and an area of the B terminal are the same as those of a desired working city and an area of the C terminal;
when the city of the working place of the B-terminal component is the same as the city of the existing living of the C-terminal component, y2A, namely a city at the B terminal is the same as one expected working city at the C terminal;
when none of the above conditions is satisfied, y2＝0；
y3maxCalculate y 2a3The step (2) is calculated from high to low step by step, and comprises the following steps:
when the salary information of the B-side component is an interview or the expected monthly salary of the C-side component is an interview and the monthly salary of the job position is confidential, y3＝a；
When the lower salary limit represented by the B-end component salary information is larger than or equal to the lower salary limit represented by the C-end component expected monthly salary, and the lower salary limit represented by the B-end component salary information is smaller than or equal to the upper salary limit represented by the C-end component expected monthly salary, y is smaller than or equal to the upper salary limit represented by the B-end component salary information3＝2a；
When the lower salary limit represented by the B-end component salary information is larger than or equal to the lower salary limit represented by the C-end component position monthly salary, and the lower salary limit represented by the B-end component salary information is smaller than or equal to the upper salary limit represented by the C-end component position monthly salary, y is smaller than or equal to the upper salary limit represented by the B-end component salary information3＝1.7a；
When the lower salary limit expressed by the | B-end component salary information-the lower salary limit expressed by the expected monthly salary of the C-end component is less than or equal to the upper salary limit expressed by the expected monthly salary of the | C-end component-the lower salary limit expressed by the expected monthly salary of the C-end component, y is3＝1.5a；
When none of the above conditions is satisfied, y3＝0.5a；
y4maxCalculating y as a4The step (2) is calculated from high to low step by step, and comprises the following steps:
when the quantized value corresponding to the B-terminal component basic academic record is equal to the quantized value corresponding to the C-terminal component basic academic record, y4＝a；
When the quantization value corresponding to the B-end component skill calendar is equal to the quantization value corresponding to the C-end component skill calendar, and the quantization value corresponding to the C-end component basic skill calendar-the quantization value corresponding to the B-end component skill calendar is less than or equal to 1, y is less than or equal to 14＝a；
When the B-terminal component skill calendar corresponding quantization value is equal to-1, y4＝0.5a；
When the quantized value corresponding to the C-end component basic academic calendar-the quantized value corresponding to the B-end component skill academic calendar is less than or equal to 1, y40.5 a; when none of the above conditions is satisfied, y4＝0；
y5maxCalculating y as a5The step (2) is calculated from high to low step by step, and comprises the following steps:
when the gear of the C-terminal component work experience information is equal to the gear represented by the B-terminal component work experience information, y5＝2a；
When the gear of the C-side component work experience information is 1 gear and the gear represented by the C-side component work experience information-B-side component work experience information is-1, or when the gear of the C-side component work experience information is not 1 gear and the gear represented by the C-side component work experience information-B-side component work experience information is 1 gear, y is5＝1.5a；
When the gear of the C-side component work experience information is 1 gear and the gear represented by the C-side component work experience information-the gear represented by the B-side component work experience information is-2, or when the gear of the C-side component work experience information is not 1 gear and the gear represented by the C-side component work experience information-the gear represented by the B-side component work experience information is-1, y is5＝a；
When none of the above conditions is satisfied, y5＝0.5a；
When the C-side component work experience information is 1 gear, see table 2 specifically:
TABLE 2
y6maxCalculating y as a6The step (2) is calculated from high to low step by step, and comprises the following steps:
when the B-side component major industry is the same as the C-side component desires to engage in one of the industries and the industry category, y is the same6＝a；
When the B-side component major industry is the same as the C-side component desires to engage in one of the industries and one of the industry categories,
y6＝0.8a；
when one of the B-side component minor industries is the same as one of the C-side component desires to engage in one of the industries and the industry category,
y6＝0.6a；
when one of the B-side component minor industries is the same as one of the C-side component expected to engage in one of the industries and one of the industry categories, y6＝0.4a；
When none of the above conditions is satisfied, y6＝0；
When y is2≥a、y3≥1.7a、y5Not less than 1.5a and y6When the value is more than or equal to 0.8a, c is 5 a; when y is2≥a、y3Not less than 1.7a and y5When the value is more than or equal to 1.5a, c is 4 a; when y is2A is not less than a and y5When the value is more than or equal to 1.5a, c is 3 a.
While embodiments of the invention have been described above, it is not limited to the applications set forth in the description and the embodiments, which are fully applicable in various fields of endeavor to which the invention pertains, and further modifications may readily be made by those skilled in the art, it being understood that the invention is not limited to the details shown and described herein without departing from the general concept defined by the appended claims and their equivalents.
Claims (3)
1. The resume content-based job recommendation method is characterized by comprising the following steps of:
acquiring the recruitment information of each recruitment position, and generating a position vector P for each recruitment position according to the recruitment informationiAll position vectors form a position vector set P, P ═ P1、P2、…、Pi、…、Pm)；
Obtaining the resume information in each user resume, and generating a resume vector R for each user resume according to the resume informationjAll resume vectors form a resume vector set R, R ═ R (R)1、R2、…、Rj、…、Rm) The job position information of each recruitment position and the resume information in each user resume respectively comprise keywords for representing the position information, the work place information, the salary information, the position type information, the academic information, the work experience information and the industry information;
selecting a user resume as a target resume, traversing a position vector centralized position vector according to the resume vector of the target resume, eliminating the position vector by adopting a ticket veto method according to position information, workplace information, salary information, position type information and academic information, and generating a candidate position vector set by using the rest position vector;
calculating the similarity of resume information corresponding to the target resume and recruitment information corresponding to each position vector of the candidate position vector set according to position information, work place information, salary information, academic information, work experience information and industry information, screening the first N recruitment positions with the highest similarity to the target resume to generate a position recommendation list, and recommending the position recommendation list to a job seeker;
the position information of the recruitment information of each recruitment position comprises a position category major class, a position category minor class and a position name; the industry information comprises major industry and minor industry; the work place information comprises a work place city and a work place area;
the position information of the resume information of each user resume comprises a major category expected to be engaged in the job, a minor category expected to be engaged in the job, a major category of the position, a minor category of the position and a name of the position; the workplace information comprises cities and areas of expected workplaces, living places and areas, and the salary information comprises expected monthly salaries and position monthly salaries; the industry information includes expected pursuit of an industry, an industry category; wherein, the job category major category, the job category minor category, the job name, the job monthly salary and the industry category are all corresponding attributes of the latest job of the job seeker;
respectively establishing input page frames which are used for inputting recruitment information and resume information and comprise defined keyword options;
the method for acquiring the recruitment information of each recruitment position comprises the following steps: guiding the recruiter to select and input the defined keywords, acquiring the defined keywords and the defined keyword attributes input by the recruiter, and combining to generate the recruitment information corresponding to the recruitment position;
the method for acquiring the resume information in each user resume comprises the following steps: guiding the job seeker to select and input the defined keywords, acquiring the defined keywords and the defined keyword attributes input by the job seeker, and combining to generate resume information in the corresponding user resume;
when the academic information is input, the corresponding defined keyword options comprise basic academic options and skill academic options, the basic academic options comprise junior high school, college, subject, master and doctor, the skill academic options comprise middle school, master, MBA and EMBA, the basic academic options corresponding to the recruitment information also comprise unlimited items, the basic academic options corresponding to the resume information also comprise others, and each defined keyword corresponding to the academic information is quantized: not limited to-1, other-0, junior-middle-1, senior-middle-2, major-3, parent-4, major-5, doctor-6, junior-2, junior-3, MBA-5, EMBA-6;
defining a position vector as a terminal B and a resume vector as a terminal C, and specifically, eliminating the position vector according to position information, workplace information, salary information, position type information and academic information by adopting a ticket veto method:
judging whether the component position name of the B terminal is contained in all position name sets covered by the component position information of the C terminal, and if not, excluding;
if yes, judging whether the B-end component work place city is contained in the component of the C-end component work place information, and if not, excluding;
if yes, judging whether the lower salary limit represented by the salary information of the B-end component is smaller than the lower salary limit represented by the expected monthly salary of the C-end component and smaller than the lower salary limit represented by the monthly salary of the position of the C-end component, and if yes, excluding;
if not, judging whether the position type represented by the B-terminal component position type information is the same as the position type represented by the C-terminal component position type information, and if not, excluding;
if yes, judging whether the quantized value corresponding to the component academic aptitude information of the B terminal is equal to-1, and if yes, storing the candidate position vector set;
if not, judging whether the quantized value corresponding to the B-end component academic record information-the quantized value corresponding to the C-end component academic record information is less than or equal to 2, if not, excluding, if so, storing in a candidate position vector set;
when the work experience information is input, the corresponding defined keyword options comprise: 0. defining the defined keywords corresponding to the work experience information as 1 gear, 2 gear, 3 gear, … gear and 7 gear from top to bottom in sequence for less than 1 year, 1-3 years, 3-5 years, 5-7 years, 7-9 years and more than 10 years;
the similarity between the resume information corresponding to the target resume and the recruitment information corresponding to the position vector in the candidate position vector set is Y (P)i,Rj)，Y(Pi,Rj)＝y1+y2+y3+y4+y5+y6+c，y1、y2、y3、y4、y5、y6Respectively representing the similarity of the target resume and the corresponding recruitment position on the position information, the workplace information, the salary information, the academic information, the work experience information and the industry information, wherein:
y1max＝3.5a，a>0, calculate y1When the value is taken, the calculation is carried out step by step from high to low, and the calculation comprises the following steps:
y is when the category subclass of the B-terminal component position is the same as one of the category subclass of the C-terminal component position and the expected employment subclass, and the name of the B-terminal component position is the same as that of the C-terminal component position1＝3.5a；
When the category of the B-end component position is subclass and the C-end component positionOne of the category subclass and one of the vocational subclasses expected to be engaged in, and the B-side component position name is the same as the C-side component position name, y1＝3a；
Y is the same when the category of the B-side component position is the same as the category of the C-side component position is the same as one of the categories of the expected job jobs and the name of the B-side component position is the same as the name of the C-side component position1＝2.5a；
Y is the same when the category of the B-side component position is the same as one of the category of the C-side component position and one of the categories of the expected engaged position, and the name of the B-side component position is the same as that of the C-side component position1＝2a；
When the B-terminal component position name is the same as the C-terminal component position name, y1＝a；
When none of the above conditions is satisfied, y1＝0.5a；
y2max1.5a, calculate y2The step (2) is calculated from high to low step by step, and comprises the following steps:
when the working information of the B-end component is identical to the expected working information of the C-end component, y2＝1.5a；
When the information of the B-side component is identical to one of the information of the C-side component expected to work, y2＝1.3a；
When the information of the working place of the B-end component is the same as that of the existing residential city of the C-end component, y2＝a；
When none of the above conditions is satisfied, y2＝0；
y3maxCalculate y 2a3The step (2) is calculated from high to low step by step, and comprises the following steps:
when the salary information of the B-side component is an interview or the expected monthly salary of the C-side component is an interview and the monthly salary of the job position is confidential, y3＝a；
When the lower salary limit represented by the B-end component salary information is larger than or equal to the lower salary limit represented by the C-end component expected monthly salary, and the lower salary limit represented by the B-end component salary information is smaller than or equal to the upper salary limit represented by the C-end component expected monthly salary, y is smaller than or equal to the upper salary limit represented by the B-end component salary information3＝2a；
When the lower salary limit represented by the B-end component salary information is larger than or equal to the lower salary limit represented by the C-end component position monthly salary, and the lower salary limit represented by the B-end component salary information is smaller than or equal to the upper salary limit represented by the C-end component position monthly salary, y is smaller than or equal to the upper salary limit represented by the B-end component salary information3＝1.7a；
When the lower salary limit expressed by the | B-end component salary information-the lower salary limit expressed by the expected monthly salary of the C-end component is less than or equal to the upper salary limit expressed by the expected monthly salary of the | C-end component-the lower salary limit expressed by the expected monthly salary of the C-end component, y is3＝1.5a；
When none of the above conditions is satisfied, y3＝0.5a；
y4maxCalculating y as a4The step (2) is calculated from high to low step by step, and comprises the following steps:
when the quantized value corresponding to the B-terminal component basic academic record is equal to the quantized value corresponding to the C-terminal component basic academic record, y4＝a；
When the quantization value corresponding to the B-end component skill calendar is equal to the quantization value corresponding to the C-end component skill calendar, and the quantization value corresponding to the C-end component basic skill calendar-the quantization value corresponding to the B-end component skill calendar is less than or equal to 1, y is less than or equal to 14＝a；
When the B-terminal component skill calendar corresponding quantization value is equal to-1, y4＝0.5a；
When the quantized value corresponding to the C-end component basic academic calendar-the quantized value corresponding to the B-end component skill academic calendar is less than or equal to 1, y4＝0.5a；
When none of the above conditions is satisfied, y4＝0；
When the gear of the C-terminal component work experience information is equal to the gear represented by the B-terminal component work experience information, y5＝2a；
When the gear of the C-side component work experience information is 1 gear and the gear represented by the C-side component work experience information-B-side component work experience information is-1, or when the gear of the C-side component work experience information is not 1 gear and the gear represented by the C-side component work experience information-B-side component work experience information is 1 gear, y is5＝1.5a；
When the gear of the C-end component work experience information is 1 gear, the C-end component work experience information is addedWhen the gear of the empirical information-B-side component work empirical information indicates the gear of-2, or when the gear of the C-side component work empirical information is not 1 gear and the gear of the C-side component work empirical information-B-side component work empirical information indicates the gear of-1, y is the same as5＝a；
When none of the above conditions is satisfied, y5＝0.5a；
y6maxCalculating y as a6The step (2) is calculated from high to low step by step, and comprises the following steps:
when the B-side component major industry is the same as the C-side component desires to engage in one of the industries and the industry category, y is the same6＝a；
When the B-side component major industry is the same as the C-side component desires to engage in one of the industries and one of the industry categories, y6＝0.8a；
When one of the B-side component minor industries is the same as one of the C-side component expected to engage in the industries and the industry category, y is the same6＝0.6a；
When one of the B-side component minor industries is the same as one of the C-side component expected to engage in one of the industries and one of the industry categories, y6＝0.4a；
When none of the above conditions is satisfied, y6＝0；
When y is2≥a、y3≥1.7a、y5Not less than 1.5a and y6When the value is more than or equal to 0.8a, c is 5 a; when y is2≥a、y3Not less than 1.7a and y5When the value is more than or equal to 1.5a, c is 4 a; when y is2A is not less than a and y5When the value is more than or equal to 1.5a, c is 3 a.
2. The job recommendation method based on resume content according to claim 1, wherein the user resume is selected as the target resume when the job seeker selects and inputs a predefined keyword to complete the user resume creation or when the job seeker changes the resume and completes.
3. The resume content-based job title recommendation method of claim 2, wherein when a new job title vector is generated, the user resumes corresponding to all the resume vectors in the resume vector set are selected as target resumes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910308204.4A CN110032681B (en) | 2019-04-17 | 2019-04-17 | Resume content-based job recommendation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910308204.4A CN110032681B (en) | 2019-04-17 | 2019-04-17 | Resume content-based job recommendation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110032681A CN110032681A (en) | 2019-07-19 |
| CN110032681B true CN110032681B (en) | 2022-03-15 |
Family
ID=67238778
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910308204.4A Active CN110032681B (en) | 2019-04-17 | 2019-04-17 | Resume content-based job recommendation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110032681B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111429082B (en) * | 2020-02-18 | 2023-04-07 | 北京网聘信息技术有限公司 | Bidirectional matching similarity calculation method for job positions and resumes of recruitment website |
| CN113627135B (en) * | 2020-05-08 | 2023-09-29 | 百度在线网络技术（北京）有限公司 | Recruitment post description text generation method, device, equipment and medium |
| CN111708929B (en) * | 2020-06-17 | 2023-05-30 | 北京字节跳动网络技术有限公司 | Information searching method, device, electronic equipment and storage medium |
| CN111858631A (en) * | 2020-07-16 | 2020-10-30 | 深圳刺猬教育科技有限公司 | Third-party platform delivery system |
| CN113570348A (en) * | 2021-09-26 | 2021-10-29 | 山东光辉人力资源科技有限公司 | Resume screening method |
| CN116523484B (en) * | 2023-06-29 | 2023-12-08 | 众科云(北京)科技有限公司 | Method and system for optimizing and collecting work platform data |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105893641A (en) * | 2016-07-01 | 2016-08-24 | 中国传媒大学 | Job recommending method |
| CN107590133A (en) * | 2017-10-24 | 2018-01-16 | 武汉理工大学 | The method and system that position vacant based on semanteme matches with job seeker resume |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100153290A1 (en) * | 2008-12-12 | 2010-06-17 | Paul Duggan | Methods of matching job profiles and candidate profiles |
| US10878381B2 (en) * | 2013-04-29 | 2020-12-29 | Monster Worldwide, Inc. | Identification of job skill sets and targeted advertising based on missing skill sets |
| CN106447285B (en) * | 2016-09-12 | 2020-06-12 | 北京大学 | Recruitment information matching method based on multi-dimensional domain key knowledge |
| CN108920544A (en) * | 2018-06-13 | 2018-11-30 | 桂林电子科技大学 | A kind of personalized position recommended method of knowledge based map |
| CN109558429B (en) * | 2018-11-16 | 2021-01-01 | 广东百城人才网络股份有限公司 | Talent service bidirectional recommendation method and system based on Internet big data |
-
2019
- 2019-04-17 CN CN201910308204.4A patent/CN110032681B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105893641A (en) * | 2016-07-01 | 2016-08-24 | 中国传媒大学 | Job recommending method |
| CN107590133A (en) * | 2017-10-24 | 2018-01-16 | 武汉理工大学 | The method and system that position vacant based on semanteme matches with job seeker resume |
Non-Patent Citations (2)
| Title |
|---|
| 基于移动就业信息服务的个性化推荐系统设计与实现;郑苗;《中国优秀硕士学位论文全文数据库信息科技辑》;20150815(第8期);第I138-1619页 * |
| 数据挖掘技术在人力资源市场中的应用研究;叶剑宇;《中国优秀硕士学位论文全文数据库信息科技辑》;20160315(第3期);第I138-5366页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110032681A (en) | 2019-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110032681B (en) | Resume content-based job recommendation method | |
| Salehi et al. | Personalized recommendation of learning material using sequential pattern mining and attribute based collaborative filtering | |
| US10157218B2 (en) | Author disambiguation and publication assignment | |
| Ying et al. | Collaborative deep ranking: A hybrid pair-wise recommendation algorithm with implicit feedback | |
| Balog et al. | Formal models for expert finding in enterprise corpora | |
| US10133807B2 (en) | Author disambiguation and publication assignment | |
| Nadimi-Shahraki et al. | Cold-start problem in collaborative recommender systems: Efficient methods based on ask-to-rate technique | |
| Boukhris et al. | A novel personalized academic venue hybrid recommender | |
| Nakata et al. | Twofold rough approximations under incomplete information | |
| Crescenzi et al. | Crowdsourcing for data management | |
| Gandhi et al. | A survey on recommendation system with collaborative filtering using big data | |
| CN112085158A (en) | Book recommendation method based on stack noise reduction self-encoder | |
| Zhao et al. | Multimedia retrieval via deep learning to rank | |
| Long et al. | Relational data clustering: models, algorithms, and applications | |
| Chen et al. | Hybrid deep collaborative filtering for job recommendation | |
| Sang et al. | Design and implementation of collaborative filtering approach for movie recommendation system | |
| CN110795640A (en) | Adaptive group recommendation method for compensating group member difference | |
| US9928291B2 (en) | Author disambiguation and publication assignment | |
| CN115630141A (en) | Scientific and technological expert retrieval method based on community query and high-dimensional vector retrieval | |
| Anand et al. | Information Retrieval in Computing Model | |
| Steven McMillan et al. | Paradigm shifts in industrial relations: A bibliometric and social network approach | |
| Kumar et al. | Comparative analysis of intersection algorithms on queries using precision, recall and F-score | |
| Patra et al. | Research on Digital Preservation: An empirical analysis | |
| Munz et al. | String coding in a generic framework | |
| Li | Application of Internet of Things technology in Human resource recommendation engine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 100102 unit 02-31, 5 / F, 5 / F, 10 Furong street, Chaoyang District, Beijing Patentee after: Beijing Wangpin Information Technology Co.,Ltd. Address before: 100102 unit 02-31, 5 / F, 5 / F, 10 Furong street, Chaoyang District, Beijing Patentee before: BEIJING WANGPIN CONSULTING Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |