CN113487295B - Off-line post recall method of part-time post recommendation system - Google Patents
Off-line post recall method of part-time post recommendation system Download PDFInfo
- Publication number
- CN113487295B CN113487295B CN202110842049.1A CN202110842049A CN113487295B CN 113487295 B CN113487295 B CN 113487295B CN 202110842049 A CN202110842049 A CN 202110842049A CN 113487295 B CN113487295 B CN 113487295B
- Authority
- CN
- China
- Prior art keywords
- post
- area
- city
- h3address
- recommendation system
- 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
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000000694 effects Effects 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000007115 recruitment Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/105—Human resources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- 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
- 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
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Economics (AREA)
- Remote Sensing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an off-line post recall method of a part-time post recommendation system, which comprises the following steps: s1, dividing all cities with part-time post distribution into regular hexagon h3 areas according to Uberh3api, wherein each city h3 area corresponds to the same city h3Address, and each post h3 area corresponds to one post h3Address; s2, calculating the distance between each city h3 area and the central point of each post h3 area in an off-line manner; s3, dividing the distance between the post h3 area and the city h3 area into a plurality of layers, wherein each layer is provided with n post data, so as to form a data pool; and S4, after the recommendation system receives the user request, generating h3Address of the user activity area in real time according to the longitude and latitude of the user, and recalling the post data of the corresponding data pool in the step S3 according to the h3Address of the user activity area. The invention can be suitable for the problem of recall of off-line posts in large data magnitude according to distance, reduces the calculation cost of a recommendation system and accelerates the recall efficiency.
Description
Technical Field
The invention relates to the field of part position recommendation application, in particular to an off-line position recall method of a part position recommendation system.
Background
On the one hand, with the surge of the Internet and the civilian entrepreneur, each industry faces industry upgrading, the competition of enterprises is focused on talent competition, and talent recruitment becomes the first major of each enterprise. The part job posts belong to the flexible employment category and are relatively special. The requirements of various types of posts, short working period, skills, experience and the like are greatly different from those of common full-time posts, so that different algorithm models are required to be built according to different user characteristics and post characteristics in order to achieve the best effect of recommending the full-time posts. The recall strategy is an important ring in the whole big data post recommendation, and the subsequent post filtering, sequencing and diversified alternate display all need to depend on a result set of the earlier post recall. The common recall mode of the off-line post is to calculate the distance between the longitude and latitude of the post and the longitude and latitude of the user, and then recall the post with a closer distance preferentially, but when the off-line post data set is larger, the calculation of the distance between the user and the post in real time each time inevitably causes the performance problem of inquiry.
Disclosure of Invention
The invention aims to provide an off-line post recall method of a part-time post recommendation system. The invention can be suitable for the problem of recall of off-line posts in large data magnitude according to distance, reduces the calculation cost of a recommendation system and accelerates the recall efficiency.
In order to solve the technical problems, the technical scheme provided by the invention is as follows: an off-line post recall method of a part-time post recommendation system comprises the following steps:
s1, dividing all cities with part-time post distribution into regular hexagon h3 areas according to Uber h3api, wherein each city h3 area corresponds to the same city h3Address, and each post h3 area corresponds to one post h3Address;
s2, calculating the distance between each city h3 area and the central point of each post h3 area in an off-line manner;
s3, dividing the distance between the post h3 area and the city h3 area into a plurality of layers, wherein each layer is provided with n post data, so as to form a data pool;
and S4, after the recommendation system receives the user request, generating h3Address of the user activity area in real time according to the longitude and latitude of the user, and recalling the post data of the corresponding data pool in the step S3 according to the h3Address of the user activity area.
In the off-line post recall method of the part-time post recommendation system, in step S1, h3Address is generated by using a geoToH3Address method through inputting latitude and longitude information and resolution information as parameters.
In the off-line post recall method of the part-time post recommendation system, in step S2, the distance between each city h3 area and the center point of each post h3 area is calculated by using a cartesian product method.
In the off-line post recall method of the part-time post recommendation system, in the step S3, all posts are divided into k layers according to the distance from each city h3 on the basis of each city h3Address and the distance layer, wherein n post data closest to the city h3 area are stored in the first layer, and the more the layer is larger, the more the post data stored in the layer is far from the city h3 area.
Compared with the prior art, all cities distributed by part-time posts are divided into regular hexagon h3 areas according to Uber h3api, each city h3 area corresponds to the same city h3Address, each post h3 area corresponds to one post h3Address, the distance between each city h3 area and the central point of each post h3 area is calculated offline, and because users are distributed on each city area, namely, the h3 addresses generated according to the user activity areas are subsets of the cities h3 addresses, the distances between all city h3 areas and all off-line post h3 areas are obtained, and then the distances between each user activity area and all off-line posts are calculated in advance, so that huge expenditure caused by the fact that the data size under the scene of calculating the distances in real time according to the longitude and latitude of the users and the longitude and latitude of all posts is relatively large is omitted; meanwhile, when the off-line posts are many, the off-line posts with the designated number are required to be acquired at one time after the off-line posts are ordered according to the distance, instead of carrying out the recall of all the posts in a large batch, and the post h3 area and the city h3 area are required to be divided according to the levels. Therefore, the invention can be suitable for the problem of recall of off-line posts according to distance under the large data magnitude, reduces the calculation cost of a recommendation system and accelerates the recall efficiency.
Drawings
FIG. 1 is a schematic flow chart of the steps of the present invention;
fig. 2 is a schematic diagram of urban h3 area division in an embodiment of the invention.
Detailed Description
The invention is further illustrated by the following examples and figures, which are not intended to be limiting.
Examples: an off-line post recall method of a part-time post recommendation system, as shown in fig. 1, comprises the following steps:
s1, dividing all cities with part-time post distribution into regular hexagon h3 areas according to Uber h3api, wherein each city h3 area corresponds to the same city h3Address, and each post h3 area corresponds to one post h3Address; uber h3 is a grid-based spatial index, but unlike the normal rectangular grid index, each of his grids is regular hexagonal. The API divides each spatial region on the earth into regular hexagons with the corresponding resolution according to the resolution input by the user, and the larger the resolution is, the smaller the generated regular hexagons are, the more. Under the condition of certain resolution, the hexagonal area divided by the h3api is unchanged, and each hexagonal area corresponds to a unique h3Address. In this embodiment, the h3Address is generated by using the geoToH3Address method in the com.uber.h3core.h3core packet, where the input parameters are latitude and longitude information and resolution information, and under the condition of a certain resolution, the h3 addresses obtained when any one of the longitudes and latitudes on a certain hexagonal area is the same. It should be noted that, the detailed division mechanism about the Uber is not an innovation point of the present invention, and the present invention only applies the h3API of the Uber to perform hexagonal division on each city nationally; taking the Hangzhou city of FIG. 2 as an example, all hexagons covering the Hangzhou city in the above figure have a unique h3Address, referred to as the city h3Address. For the off-line posts, each post has a city and the longitude and latitude thereof, namely the off-line posts are distributed on each h3 area, and under a specific resolution (consistent with the resolution of generating the city h3 Address), the unique h3Address representing the area of the off-line post can be obtained through the longitude and latitude of the off-line post, which is called post h3Address.
S2, calculating the Distance between each city h3 area and the central point of each post h3 area in an off-line mode by using a Cartesian product solving method, in the embodiment, uber also provides an api for calculating the Distance between the central points of two hexagonal areas under the same resolution, and only a com.uber.h3core.H3Core h3Distance method (only the Cartesian product solving method is adopted in the method) is required to be called, wherein the method is called as h3Address of two hexagonal areas. The distance between the two h3 center points can be easily calculated in advance by using the api for calculating the distance between the two h3 center points;
s3, dividing the distance between the post h3 area and the city h3 area into a plurality of layers, wherein each layer is provided with n post data, so as to form a data pool; in this embodiment, each city h3Address and distance level are used as keys or bases, all the posts are divided into k levels according to the distance from the city h3, wherein n post data closest to the city h3 area are stored in the first level, the more the level is, the more the post data stored in the first level is far away from the city h3 area, namely, the n post data closest to the city h3 area are stored in the first level, and the n post data furthest from the city h3 area are stored in the kth level.
S4, after the recommendation system receives the user request, h3 addresses of a user active area are generated in real time according to the longitude and latitude of the user, and post data of corresponding data pools in the step S3 are recalled according to the h3 addresses of the user active area, namely, the distances between all post h3 and the center point of the city h3 area are pre-ordered, the user active area is distributed on the city h3 area, and the user h3 is a subset of the city h3 necessarily, so that when h3api of Uber is used for generating h3 addresses for the longitude and latitude of the user, as long as the resolution of the h3 addresses is consistent, post pools (data pools) which are pre-calculated for the h3 addresses and are arranged in sequence according to the distances in advance can be precisely recalled, and the post data in the post pools (data pools) can be quickly recalled.
In summary, the invention can save the huge expense caused by the large data volume in the scene of calculating the distance in real time according to the longitude and latitude of the user and the longitude and latitude of all posts, and simultaneously ensures that the user can recall when the number of posts is large. Therefore, the invention can be suitable for the problem of recall of off-line posts according to distance under the large data magnitude, reduces the calculation cost of a recommendation system and accelerates the recall efficiency.
Claims (4)
1. An off-line post recall method of a part-time post recommendation system is characterized in that: the method comprises the following steps:
s1, dividing all cities with part-time post distribution into regular hexagon h3 areas according to Uber h3api, wherein each city h3 area corresponds to the same city h3Address, and each post h3 area corresponds to one post h3Address;
s2, calculating the distance between each city h3 area and the central point of each post h3 area in an off-line manner;
s3, dividing the distance between the post h3 area and the city h3 area into a plurality of layers, wherein each layer is provided with n post data, so as to form a data pool;
and S4, after the recommendation system receives the user request, generating h3Address of the user activity area in real time according to the longitude and latitude of the user, and recalling the post data of the corresponding data pool in the step S3 according to the h3Address of the user activity area.
2. The off-line post recall method of a part-time post recommendation system of claim 1, wherein: in step S1, h3Address is generated by using the geoToH3Address method by inputting latitude and longitude information and resolution information as parameters.
3. The off-line post recall method of a part-time post recommendation system of claim 1, wherein: in step S2, a cartesian product method is used to calculate the distance between each city h3 area and the center point of each post h3 area.
4. The off-line post recall method of a part-time post recommendation system of claim 1, wherein: in step S3, based on each city h3Address and distance level, all the posts are divided into k levels according to the distance from the city h3 in reverse order, wherein n post data closest to the city h3 area are stored in the first level, and the more the level is, the more the post data stored in the first level is far from the city h3 area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110842049.1A CN113487295B (en) | 2021-07-26 | 2021-07-26 | Off-line post recall method of part-time post recommendation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110842049.1A CN113487295B (en) | 2021-07-26 | 2021-07-26 | Off-line post recall method of part-time post recommendation system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113487295A CN113487295A (en) | 2021-10-08 |
| CN113487295B true CN113487295B (en) | 2024-01-19 |
Family
ID=77943618
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110842049.1A Active CN113487295B (en) | 2021-07-26 | 2021-07-26 | Off-line post recall method of part-time post recommendation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113487295B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110998568A (en) * | 2018-03-28 | 2020-04-10 | 北京嘀嘀无限科技发展有限公司 | Navigation determination system and method for embarkable vehicle seeking passengers |
| US10762455B1 (en) * | 2016-11-28 | 2020-09-01 | Blue Yonder Group, Inc. | System and method of schedule optimization for long-range staff planning |
| CN111797093A (en) * | 2020-05-13 | 2020-10-20 | 中国科学院软件研究所 | Discrete global grid structure generation method and rapid unit positioning method |
| CN112367531A (en) * | 2020-10-30 | 2021-02-12 | 腾讯科技（深圳）有限公司 | Video stream display method, processing method and related equipment |
| CN112527867A (en) * | 2020-12-18 | 2021-03-19 | 重庆师范大学 | Non-agricultural employment post supply capacity identification method, storage device and server |
| CN112860554A (en) * | 2021-02-07 | 2021-05-28 | 杭州弧途科技有限公司 | Part-time post recommendation system based on multi-algorithm strategy bucket-dividing test |
| CN113127741A (en) * | 2021-04-29 | 2021-07-16 | 杭州弧途科技有限公司 | Cache method for reading and writing data of mass users and posts in part-time post recommendation system |
-
2021
- 2021-07-26 CN CN202110842049.1A patent/CN113487295B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10762455B1 (en) * | 2016-11-28 | 2020-09-01 | Blue Yonder Group, Inc. | System and method of schedule optimization for long-range staff planning |
| CN110998568A (en) * | 2018-03-28 | 2020-04-10 | 北京嘀嘀无限科技发展有限公司 | Navigation determination system and method for embarkable vehicle seeking passengers |
| CN111797093A (en) * | 2020-05-13 | 2020-10-20 | 中国科学院软件研究所 | Discrete global grid structure generation method and rapid unit positioning method |
| CN112367531A (en) * | 2020-10-30 | 2021-02-12 | 腾讯科技（深圳）有限公司 | Video stream display method, processing method and related equipment |
| CN112527867A (en) * | 2020-12-18 | 2021-03-19 | 重庆师范大学 | Non-agricultural employment post supply capacity identification method, storage device and server |
| CN112860554A (en) * | 2021-02-07 | 2021-05-28 | 杭州弧途科技有限公司 | Part-time post recommendation system based on multi-algorithm strategy bucket-dividing test |
| CN113127741A (en) * | 2021-04-29 | 2021-07-16 | 杭州弧途科技有限公司 | Cache method for reading and writing data of mass users and posts in part-time post recommendation system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113487295A (en) | 2021-10-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lian et al. | A novel particle swarm optimization algorithm for permutation flow-shop scheduling to minimize makespan | |
| CN109597965B (en) | Data processing method, system, terminal and medium based on deep neural network | |
| CN1637764A (en) | Workflow system and method | |
| CN105701108A (en) | Information recommendation method, information recommendation device and server | |
| CN102027468A (en) | Provisioning a geographical image for retrieval | |
| CN111652378A (en) | Learning to select vocabulary of category features | |
| CN111832869A (en) | Vehicle scheduling method and device, electronic equipment and storage medium | |
| CN105528395A (en) | Method and system for recommending potential consumers | |
| CN110928576A (en) | Convolution processing method and device of convolutional neural network and storage medium | |
| CN113487295B (en) | Off-line post recall method of part-time post recommendation system | |
| CN110110899A (en) | Prediction technique, adaptive learning method and the electronic equipment of acquisition of knowledge degree | |
| CN110287415A (en) | A kind of content recommendation method, device and calculate equipment | |
| CN111723306A (en) | Sequence position recommendation method based on geographical perception | |
| CN105718147A (en) | Input method panel enabling method and device and input method and input method system | |
| CN116151600B (en) | Maintenance method, device, computer equipment and storage medium for shared vehicle | |
| CN116503098A (en) | Mining method, mining device, computer equipment and storage medium for shared vehicle station | |
| CN109978353B (en) | Big data crowd sensing incentive method facing position-adjustable user | |
| CN1256670C (en) | Application server sample pool dynamic configuration method based on real number coding genetic calculation | |
| CN109754805A (en) | A kind of the voice input method and platform of production operation process | |
| CN106909612B (en) | Method and device for processing following behavior data | |
| CN113641471B (en) | Soft load scheduling method, device, equipment and medium based on genetic algorithm model | |
| CN113886709A (en) | Recommendation method, device and equipment for business products and storage medium | |
| WO2021126076A1 (en) | Methods and systems for recommendation using a neural network | |
| CN115759306B (en) | Analog wander method, device, electronic equipment and storage medium | |
| CN110609850A (en) | Information determination method, electronic equipment and computer storage medium |
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 |