CN109034869A - Real-time recommendation system and method based on similar audient - Google Patents
Real-time recommendation system and method based on similar audient Download PDFInfo
- Publication number
- CN109034869A CN109034869A CN201810671631.4A CN201810671631A CN109034869A CN 109034869 A CN109034869 A CN 109034869A CN 201810671631 A CN201810671631 A CN 201810671631A CN 109034869 A CN109034869 A CN 109034869A
- Authority
- CN
- China
- Prior art keywords
- unit
- real
- data
- time
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0255—Targeted advertisements based on user history
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/06—Asset management; Financial planning or analysis
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Game Theory and Decision Science (AREA)
- Theoretical Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Operations Research (AREA)
- Technology Law (AREA)
- Human Resources & Organizations (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses the real-time recommendation system and methods based on similar audient, the system includes off-line module, Real time capable module, business recommended server, assess all historical datas of user, the behavior of user is learnt by machine learning, and assessment of the completion to user in second grade, so as to recommend related service to user in time, and it can be truly realized objective effective, avoid problems brought by human intervention, make to recommend more preferably accurate, service more preferably customizes, conversion ratio is higher, for product/service recommendation company, not only business development can be made more accurate, also greatly reduce expansion cost, the more rapid and convenient for becoming the popularization of product is more accurate, it allows terminal user to enjoy simultaneously and is more suitable their products & services.The present invention can be used in the real-time recommendation of financial product and the real-time recommendation of other electric business products and ad system or other similar system.
Description
Technical field
The present invention relates to big datas, the technical field of data analysis and data mining, more particularly to based on similar audient's
Real-time recommendation system and method.
Background technique
The abundant of financial derivative product provide the user selection abundant, and multifarious selection also increases making for user
Suitable selection can not be often made when user faces these selections with difficulty, so that the development of financial business is affected,
Also counteract that user enjoys the good service that it should be enjoyed.
For this problem, the common practice of present industry is that financing corporation sets up huge business department, by business people
The business that member promotes to user and customized user needs.There are also big problems for this.On the one hand, the development one of this business
As can only be carried out under line, time-consuming and laborious, waste of resource.On the other hand, due to a large amount of manual intervention, exist many subjective not true
Factor is determined, so that user can not really obtain oneself most desirable service.In addition to this, between consumer and financial service subsidiary
Communication problem and trust problem are generated, shaping up for business is also just hindered.
Technical aspect, the appearance of network recommendation system is already present in some e-commerce and application scenarios are recommended in advertisement
In.These systems can generally handle a large amount of user data, and the feature of the action learning user by user, periodically generate mould
Type, and product either advertisement is promoted to user based on the model of generation.But this model is generally based on mass data,
So its data analysis process generally requires long time, for example several hours were differed by several days.It is answered when the model of generation
When using real-time system, the behavior of user may have occurred that huge variation, so that recommendation resource is wasted, also with regard to nothing
Method, which provides a user, more accurately to be serviced.Even in addition, in e-commerce or web advertisement recommender system, recommender system
Generally still be artificially defined some rules and the behavior for being used to match a new user based on business experience, decide whether to
It recommends complementary goods or advertisement.Here an experience and subjectivity problem are equally existed, because the definition of rule is to rely on
Experience, and the subjectivity of people and experience difference directly limit the validity of the rule of definition.
Similar, this occasion for needing real-time recommendation, the prior art cannot meet user well and product/service mentions
For quotient for the demand of real-time, accurate recommendation.
Summary of the invention
For overcome the deficiencies in the prior art, the purpose of the present invention is to provide the real-time recommendation systems based on similar audient
And method, it is intended to solve the problems, such as the prior art can not in real time, accurately to user's recommendation or service.
The purpose of the present invention is implemented with the following technical solutions:
A kind of real-time recommendation system based on similar audient, including off-line module, Real time capable module, business recommended server,
Wherein:
Off-line module includes the HDFS unit being linked in sequence, model compiler, model data library unit;HDFS unit also with
The connection of real time data library unit;
Real time capable module includes the data access module being linked in sequence, feature extraction unit, decision engine unit, real time data
Library unit;Rule database unit is connect with model compiler;Feature extraction unit, decision engine unit are also compiled with model respectively
Translate device connection;
Business recommended server is connect with real time data library unit;
HDFS unit timing is loaded into seed data and rule from real time data library unit;Model compiler is obtained from HDFS unit
All seed datas and all historical datas of these seed datas are taken, feature are extracted, to these seed datas and its extraction
Feature out carries out intellectual analysis, extracts common trait, is compiled into model and is stored in model data library unit, and to rule
Rule in Database Unit is deleted or is updated;
When business recommended server receives service request, the real-time behavioral data of user by data access module into
Enter feature extraction unit and carries out feature extraction;The historical behavior data of real time data library unit extraction user;Decision engine unit
In conjunction with the real-time behavioral data and historical behavior data of user, synthesis is carried out to user from model data library unit stress model and is commented
Estimate operation, operation result is returned into real time data library unit;The fortune that business recommended server calls decision engine unit returns
Calculate result.
On the basis of the above embodiments, it is preferred that further include API server;
API server is connect with HDFS unit, rule database unit respectively;
API server receives seed data and custom rule, HDFS unit is sent by seed data, by customized rule
Then it is sent to rule database unit.
On the basis of above-mentioned any embodiment, it is preferred that further include the tracker server being connect with HDFS unit;
The operation result includes recommendation;
Business recommended server records behavior of the user corresponding with service request to recommendation;
Tracker server tracks user and tracking result is stored in HDFS unit.
On the basis of above-mentioned any embodiment, it is preferred that off-line module is run in Hadoop framework;Real time capable module exists
It is run on Apache Heron framework.
A kind of real-time recommendation method based on similar audient, comprising:
Off-line step:
HDFS unit timing is loaded into seed data and rule from real time data library unit;Model compiler is obtained from HDFS unit
All seed datas and all historical datas of these seed datas are taken, feature are extracted, to these seed datas and its extraction
Feature out carries out intellectual analysis, extracts common trait, is compiled into model and is stored in model data library unit, and to rule
Rule in Database Unit is deleted or is updated;
Real time steps:
When business recommended server receives service request, the real-time behavioral data of user by data access module into
Enter feature extraction unit and carries out feature extraction;The historical behavior data of real time data library unit extraction user;Decision engine unit
In conjunction with the real-time behavioral data and historical behavior data of user, synthesis is carried out to user from model data library unit stress model and is commented
Estimate operation, operation result is returned into real time data library unit;The fortune that business recommended server calls decision engine unit returns
Calculate result.
On the basis of the above embodiments, it is preferred that the off-line step further includes;
API server receives seed data and custom rule, HDFS unit is sent by seed data, by customized rule
Then it is sent to rule database unit.
On the basis of above-mentioned any embodiment, it is preferred that the real time steps further include:
The metadata of the real-time behavioral data of user and operation result are stored in real-time data base list by business recommended server
Member, and real-time synchronization gives HDFS unit.
On the basis of above-mentioned any embodiment, it is preferred that the real time steps further include:
The operation result includes recommendation;
Business recommended server records behavior of the user corresponding with service request to recommendation;
Tracker server tracks user and tracking result is stored in HDFS unit.
On the basis of above-mentioned any embodiment, it is preferred that the user to the behavior of recommendation include click and/or
It fills in a form application.
On the basis of above-mentioned any embodiment, it is preferred that model compiler to the compiling of model include compiling completely and
Incremental compilation.
Compared with prior art, the beneficial effects of the present invention are:
The invention discloses the real-time recommendation system and method based on similar audient, which includes off-line module, in real time
Module, business recommended server assess all historical datas of user, the behavior of user are learnt by machine learning, and in second grade
Interior assessment of the completion to user, so as to recommend related service to user in time, and can be truly realized it is objective effectively, avoid people
To intervene brought problems, keep recommendation more preferably accurate, service more preferably customizes, and conversion ratio is higher, for product/service
For recommendation company, not only business development can be made more accurate, also greatly reduce expansion cost, become the popularization of product
More rapid and convenient is more accurate, while allowing terminal user to enjoy and being more suitable their products & services.The present invention can
For the real-time recommendation of financial product and the real-time recommendation of other electric business products and ad system or other similar system.
Detailed description of the invention
Present invention will be further explained below with reference to the attached drawings and examples.
Fig. 1 shows a kind of structural representation of real-time recommendation system based on similar audient provided in an embodiment of the present invention
Figure;
Fig. 2 shows a kind of process signals of the real-time recommendation method based on similar audient provided in an embodiment of the present invention
Figure.
Specific embodiment
In the following, being described further in conjunction with attached drawing and specific embodiment to the present invention, it should be noted that not
Under the premise of conflicting, new implementation can be formed between various embodiments described below or between each technical characteristic in any combination
Example.
Specific embodiment one
As shown in Figure 1, the embodiment of the invention provides a kind of real-time recommendation system based on similar audient, including offline mould
Block, Real time capable module, business recommended server, in which:
Off-line module includes the HDFS unit being linked in sequence, model compiler, model data library unit;HDFS unit also with
The connection of real time data library unit;
Real time capable module includes the data access module being linked in sequence, feature extraction unit, decision engine unit, real time data
Library unit;Rule database unit is connect with model compiler;Feature extraction unit, decision engine unit are also compiled with model respectively
Translate device connection;
Business recommended server is connect with real time data library unit;
HDFS unit timing is loaded into seed data and rule from real time data library unit;Model compiler is obtained from HDFS unit
All seed datas and all historical datas of these seed datas are taken, feature are extracted, to these seed datas and its extraction
Feature out carries out intellectual analysis, extracts common trait, is compiled into model and is stored in model data library unit, and to rule
Rule in Database Unit is deleted or is updated;
When business recommended server receives service request, the real-time behavioral data of user by data access module into
Enter feature extraction unit and carries out feature extraction;The historical behavior data of real time data library unit extraction user;Decision engine unit
In conjunction with the real-time behavioral data and historical behavior data of user, synthesis is carried out to user from model data library unit stress model and is commented
Estimate operation, operation result is returned into real time data library unit;The fortune that business recommended server calls decision engine unit returns
Calculate result.
The embodiment of the present invention assesses all historical datas of user, the behavior of user is learnt by machine learning, and in second grade
Interior assessment of the completion to user, so as to recommend related service to user in time, and can be truly realized it is objective effectively, avoid people
To intervene brought problems, keep recommendation more preferably accurate, service more preferably customizes, and conversion ratio is higher, for product/service
For recommendation company, not only business development can be made more accurate, also greatly reduce expansion cost, become the popularization of product
More rapid and convenient is more accurate, while allowing terminal user to enjoy and being more suitable their products & services.The present invention is implemented
Example can be used in the real-time recommendation of financial product and pushing away in real time for other electric business products and ad system or other similar system
It recommends.
Preferably, the embodiment of the present invention can also include API server;API server respectively with HDFS unit, regular number
It is connected according to library unit;API server receives seed data and custom rule, sends HDFS unit for seed data, will be certainly
Definition rule is sent to rule database unit.The advantage of doing so is that by API server by seed data and customized rule
It is then stored in HDFS unit and rule database unit respectively.
Preferably, the embodiment of the present invention can also include the tracker server connecting with HDFS unit;The operation knot
Fruit includes recommendation;Business recommended server records behavior of the user corresponding with service request to recommendation;tracker
Server tracks user and tracking result is stored in HDFS unit.The advantage of doing so is that can be to user for pushing away
The feedback and subsequent behavior for recommending content are recorded, and implementation model itself is improved and improved.
Preferably, off-line module can be run in Hadoop framework;Real time capable module can be in Apache Heron framework
Upper operation.
This system is divided into off-line data collecting and data analysis and online data processing two large divisions.Since it is desired that very much
Computing resource, acquisition, analysis and the model foundation of data generally carry out offline, and the result analyzed is used directly for online
System, to effectively improve the reaction time of online service.Off-line system and on-line system in this way combines, and off-line system is regular
Operation handles historical data, and on-line system handles the flow data of user, and combines the processing result of off-line system, final to provide
The solution of one real-time recommendation.
Result is synchronized to Real time capable module and is loaded directly into use by off-line module periodic operation.Off-line module is on Hadoop
Periodic operation can extract all historical datas of all seed datas He these seed datas, extract feature, pass through model point
Analysis deletes or updates rule, and seed data and its corresponding rule are updated by API service.Whenever there is new rule
It is loaded into memory, model compiler can read in seed data required for dependency rule from HDFS, and combine this rule, carry out
Feature extraction and model compilation.Model compilation is compiled so be classified as compiling completely again with increment due to takeing a long time
It translates.Real time capable module can periodically obtain compiled model from model compiler and be loaded into memory, use for decision engine.
The embodiment of the present invention to model compilation process without limitation, it is preferred that model compilation may include prelisting for model
It translates and formally compiles two parts.The precompile of model mainly includes being loaded into seed data and rule etc. from real time data library unit
Metadata then by taking all relevant seed datas from HDFS, then obtains the historical data of all seed users, extracts
Feature, and send result to by message queue the formal collector of model.The formal collector of model is to these kinds
Child user and its feature extracted carry out intellectual analysis, extract common trait, are compiled into model, and be stored in model database
Unit.In this way, Real time capable module can be loaded into these compiled models in real time.
On the other hand, the model that Real time capable module directly utilizes offline service to generate, assesses the data accessed in real time
Calculate, quickly provide as a result, and Real-time Metadata and operation result are stored in real time data library unit, and be synchronized in real time
HDFS unit is used for business recommended server.Real time capable module handles the behavioral data of user in real time, passes through data access module
It is linked into our real-time processing data system, carries out the feature extraction of current-user data, and by real-time data base, mention
The historical behavior data at family are taken, then by decision engine, comprehensive assessment is carried out to user, calculated result is write back into active user
Database recommends business server calls for industry.
For the recommendation request that business recommended server receives, Real time capable module can read user by live user database
Information makes financial product and recommends to determine, the products & services of customization are recommended terminal user.On the other hand, terminal is used
Behavior of the family to recommendation, than such as whether clicking, if application etc. of filling in a form can also be chased after by tracker server
Track, and result is stored in HDFS, itself improving and improving for model.
It can be carried out in real time by distributed disappearance queue, such as Kafka between HDFS unit and real time data library unit
It is synchronous.In this way, the result generated can be supplied to online service use.
In above-mentioned specific embodiment one, the real-time recommendation system based on similar audient is provided, it is corresponding,
The application also provides the real-time recommendation method based on similar audient.Since embodiment of the method is substantially similar to system embodiment, institute
To describe fairly simple, related place illustrates referring to the part of system embodiment.Embodiment of the method described below is only
It is only illustrative.
Specific embodiment two
As shown in Fig. 2, the embodiment of the invention provides a kind of real-time recommendation methods based on similar audient, comprising:
Off-line step S101:
HDFS unit timing is loaded into seed data and rule from real time data library unit;Model compiler is obtained from HDFS unit
All seed datas and all historical datas of these seed datas are taken, feature are extracted, to these seed datas and its extraction
Feature out carries out intellectual analysis, extracts common trait, is compiled into model and is stored in model data library unit, and to rule
Rule in Database Unit is deleted or is updated;
Real time steps S102:
When business recommended server receives service request, the real-time behavioral data of user by data access module into
Enter feature extraction unit and carries out feature extraction;The historical behavior data of real time data library unit extraction user;Decision engine unit
In conjunction with the real-time behavioral data and historical behavior data of user, synthesis is carried out to user from model data library unit stress model and is commented
Estimate operation, operation result is returned into real time data library unit;The fortune that business recommended server calls decision engine unit returns
Calculate result.
The embodiment of the present invention assesses all historical datas of user, the behavior of user is learnt by machine learning, and in second grade
Interior assessment of the completion to user, so as to recommend related service to user in time, and can be truly realized it is objective effectively, avoid people
To intervene brought problems, keep recommendation more preferably accurate, service more preferably customizes, and conversion ratio is higher, for product/service
For recommendation company, not only business development can be made more accurate, also greatly reduce expansion cost, become the popularization of product
More rapid and convenient is more accurate, while allowing terminal user to enjoy and being more suitable their products & services.The present invention is implemented
Example can be used in the real-time recommendation of financial product and pushing away in real time for other electric business products and ad system or other similar system
It recommends.
Preferably, the off-line step S101 can also include;API server receives seed data and custom rule,
HDFS unit is sent by seed data, sends rule database unit for custom rule.The advantage of doing so is that passing through
Seed data and custom rule are stored in HDFS unit and rule database unit by API server respectively.
Preferably, the real time steps S102 can also include: business recommended server by the real-time behavioral data of user
Metadata and operation result be stored in real time data library unit, and real-time synchronization gives HDFS unit.The advantage of doing so is that saving
Metadata and operation result facilitate and subsequent are targetedly optimized.
Preferably, it includes recommendation that the real time steps S102, which can also include: the operation result,;Business recommended clothes
Business device records behavior of the user corresponding with service request to recommendation;Tracker server tracks user and incites somebody to action
Tracking result is stored in HDFS unit.The advantage of doing so is that can to user for recommendation feedback and subsequent behavior into
Row record, implementation model itself are improved and are improved.
Preferably, the user may include click and/or application of filling in a form to the behavior of recommendation.
The embodiment of the present invention to model compiler to the compiling of model without limitation, it is preferred that model compiler is to model
Compiling may include complete compiling and incremental compilation.Model compilation is due to takeing a long time, it is possible to be divided into completely
Compiling and incremental compilation.
The present invention is from using in purpose, and in efficiency, the viewpoints such as progressive and novelty are illustrated, the practical progress having
Property, oneself meets the function that Patent Law is emphasized and promotes and use important document, and more than the present invention explanation and attached drawing are only of the invention
Preferred embodiment and oneself, the present invention is not limited to this, therefore, it is all constructed with the present invention, device, wait the approximations, thunder such as levy
With, i.e., all according to equivalent replacement made by present patent application range or modification etc., the patent application that should all belong to of the invention is protected
Within the scope of shield.
It should be noted that in the absence of conflict, the feature in embodiment and embodiment in the present invention can phase
Mutually combination.Although present invention has been a degree of descriptions, it will be apparent that, in the item for not departing from the spirit and scope of the present invention
Under part, the appropriate variation of each condition can be carried out.It is appreciated that the present invention is not limited to the embodiments, and it is attributed to right and wants
The range asked comprising the equivalent replacement of each factor.It will be apparent to those skilled in the art that can as described above
Various other corresponding changes and deformation are made in technical solution and design, and all these change and deformation is all answered
Within this is belonged to the protection scope of the claims of the invention.
Claims (10)
1. a kind of real-time recommendation system based on similar audient, which is characterized in that pushed away including off-line module, Real time capable module, business
Recommend server, in which:
Off-line module includes the HDFS unit being linked in sequence, model compiler, model data library unit;HDFS unit is also and in real time
Database Unit connection;
Real time capable module includes the data access module being linked in sequence, feature extraction unit, decision engine unit, real-time data base list
Member;Rule database unit is connect with model compiler;Feature extraction unit, decision engine unit also respectively with model compiler
Connection;
Business recommended server is connect with real time data library unit;
HDFS unit timing is loaded into seed data and rule from real time data library unit;Model compiler obtains institute from HDFS unit
There are seed data and all historical datas of these seed datas, extracts feature, to these seed datas and its extract
Feature carry out intellectual analysis, extract common trait, be compiled into model and be stored in model data library unit, and to regular data
Rule in library unit is deleted or is updated;
When business recommended server receives service request, the real-time behavioral data of user enters spy by data access module
It levies extraction unit and carries out feature extraction;The historical behavior data of real time data library unit extraction user;Decision engine unit combines
The real-time behavioral data and historical behavior data of user carries out comprehensive assessment fortune to user from model data library unit stress model
It calculates, operation result is returned into real time data library unit;The operation knot that business recommended server calls decision engine unit returns
Fruit.
2. the real-time recommendation system according to claim 1 based on similar audient, which is characterized in that further include API service
Device;
API server is connect with HDFS unit, rule database unit respectively;
API server receives seed data and custom rule, sends HDFS unit for seed data, custom rule is sent out
It is sent to rule database unit.
3. the real-time recommendation system according to claim 1 or 2 based on similar audient, which is characterized in that further include with
The tracker server of HDFS unit connection;
The operation result includes recommendation;
Business recommended server records behavior of the user corresponding with service request to recommendation;
Tracker server tracks user and tracking result is stored in HDFS unit.
4. the real-time recommendation system according to claim 1 or 2 based on similar audient, which is characterized in that off-line module exists
It is run in Hadoop framework;Real time capable module is run on Apache Heron framework.
5. a kind of real-time recommendation method based on similar audient characterized by comprising
Off-line step:
HDFS unit timing is loaded into seed data and rule from real time data library unit;Model compiler obtains institute from HDFS unit
There are seed data and all historical datas of these seed datas, extracts feature, to these seed datas and its extract
Feature carry out intellectual analysis, extract common trait, be compiled into model and be stored in model data library unit, and to regular data
Rule in library unit is deleted or is updated;
Real time steps:
When business recommended server receives service request, the real-time behavioral data of user enters spy by data access module
It levies extraction unit and carries out feature extraction;The historical behavior data of real time data library unit extraction user;Decision engine unit combines
The real-time behavioral data and historical behavior data of user carries out comprehensive assessment fortune to user from model data library unit stress model
It calculates, operation result is returned into real time data library unit;The operation knot that business recommended server calls decision engine unit returns
Fruit.
6. the real-time recommendation method according to claim 5 based on similar audient, which is characterized in that the off-line step is also
Including;
API server receives seed data and custom rule, sends HDFS unit for seed data, custom rule is sent out
It is sent to rule database unit.
7. the real-time recommendation method according to claim 5 or 6 based on similar audient, which is characterized in that the real-time step
Suddenly further include:
The metadata of the real-time behavioral data of user and operation result are stored in real time data library unit by business recommended server, and
Real-time synchronization gives HDFS unit.
8. the real-time recommendation method according to claim 5 or 6 based on similar audient, which is characterized in that the real-time step
Suddenly further include:
The operation result includes recommendation;
Business recommended server records behavior of the user corresponding with service request to recommendation;
Tracker server tracks user and tracking result is stored in HDFS unit.
9. the real-time recommendation method according to claim 5 or 6 based on similar audient, which is characterized in that the user couple
The behavior of recommendation includes click and/or application of filling in a form.
10. the real-time recommendation method according to claim 5 or 6 based on similar audient, which is characterized in that model compiler
Compiling to model includes complete compiling and incremental compilation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810671631.4A CN109034869A (en) | 2018-06-26 | 2018-06-26 | Real-time recommendation system and method based on similar audient |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810671631.4A CN109034869A (en) | 2018-06-26 | 2018-06-26 | Real-time recommendation system and method based on similar audient |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109034869A true CN109034869A (en) | 2018-12-18 |
Family
ID=64611037
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810671631.4A Pending CN109034869A (en) | 2018-06-26 | 2018-06-26 | Real-time recommendation system and method based on similar audient |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109034869A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103345698A (en) * | 2013-07-09 | 2013-10-09 | 焦点科技股份有限公司 | Personalized recommendation method based on cloud processing mode and applied in e-business environment |
CN104021483A (en) * | 2014-06-26 | 2014-09-03 | 陈思恩 | Recommendation method for passenger demands |
CN105488662A (en) * | 2016-01-07 | 2016-04-13 | 北京歌利沃夫企业管理有限公司 | Bi-directional recommendation-based online recruitment system |
CN106126641A (en) * | 2016-06-24 | 2016-11-16 | 中国科学技术大学 | A kind of real-time recommendation system and method based on Spark |
-
2018
- 2018-06-26 CN CN201810671631.4A patent/CN109034869A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103345698A (en) * | 2013-07-09 | 2013-10-09 | 焦点科技股份有限公司 | Personalized recommendation method based on cloud processing mode and applied in e-business environment |
CN104021483A (en) * | 2014-06-26 | 2014-09-03 | 陈思恩 | Recommendation method for passenger demands |
CN105488662A (en) * | 2016-01-07 | 2016-04-13 | 北京歌利沃夫企业管理有限公司 | Bi-directional recommendation-based online recruitment system |
CN106126641A (en) * | 2016-06-24 | 2016-11-16 | 中国科学技术大学 | A kind of real-time recommendation system and method based on Spark |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106126641B (en) | A kind of real-time recommendation system and method based on Spark | |
Cavallo et al. | The billion prices project: Using online prices for measurement and research | |
CN106296305A (en) | Electric business website real-time recommendation System and method under big data environment | |
Shevade et al. | Oil palm plantations in Peninsular Malaysia: Determinants and constraints on expansion | |
CN104166668B (en) | News commending system and method based on FOLFM models | |
CN101226557B (en) | Method for processing efficient relating subject model data | |
JP5913467B2 (en) | Messaging composition using the context provided by the consumer | |
US20170091810A1 (en) | Brand engagement touchpoint attribution using brand engagement event weighting | |
CN102033883A (en) | Method, device and system for improving data transmission speed of website | |
CN105183727A (en) | Method and system for recommending book | |
CN102609533A (en) | Kernel method-based collaborative filtering recommendation system and method | |
CN105989074A (en) | Method and device for recommending cold start through mobile equipment information | |
CN103345698A (en) | Personalized recommendation method based on cloud processing mode and applied in e-business environment | |
CN102750334A (en) | Agricultural information accurate propelling method based on data mining (DM) | |
US20140180799A1 (en) | Techniques for optimizing the impact of video content on electronic commerce sales | |
CN104134108A (en) | Sales data analysis method of electronic commerce website | |
CN108122153A (en) | Personalized recommendation method based on cloud computing tupe under e-commerce environment | |
CN110880124A (en) | Conversion rate evaluation method and device | |
CN110009416A (en) | A kind of system based on big data cleaning and AI precision marketing | |
KR102163412B1 (en) | Method for providing business establishing information report | |
CN110297990A (en) | The associated detecting method and system of crowdsourcing marketing microblogging and waterborne troops | |
Smyrnaios | How does news infomediation operate online? The examples of Google and Facebook | |
CN109190027A (en) | Multi-source recommended method, terminal, server, computer equipment, readable medium | |
CN108536570A (en) | The method, apparatus and system that data direct broadcasting room gray scale pressure is surveyed | |
KR100641335B1 (en) | Internet keyword advertising method using secondary keyword, and Internet keyword advertising system thereof |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181218 |