CN109146562A - It is a kind of for retrieving the intelligent recommendation system that will be lost user - Google Patents
It is a kind of for retrieving the intelligent recommendation system that will be lost user Download PDFInfo
- Publication number
- CN109146562A CN109146562A CN201810917678.4A CN201810917678A CN109146562A CN 109146562 A CN109146562 A CN 109146562A CN 201810917678 A CN201810917678 A CN 201810917678A CN 109146562 A CN109146562 A CN 109146562A
- Authority
- CN
- China
- Prior art keywords
- user
- module
- intention
- intelligent recommendation
- 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.)
- 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
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- 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
- G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
- G06Q30/0224—Discounts or incentives, e.g. coupons or rebates based on user history
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This invention describes a kind of for retrieving the intelligent recommendation system that will be lost user, comprising: intention evaluation module, the module acquire the user behaviors log of user, and obtain intention score value of the user about registration behavior;Tactful configuration module, the module configure a variety of marketing incentives rules, and check each tactful bring effect statistics;Recommending module, the module generate according to the strategy protocol in the intention score value in intention evaluation module, and tactful configuration module and motivate and be presented to the user, and excitation user completes registration.Due to using above system framework, the present invention passes through intention evaluation module, tactful configuration module and recommending module, realize the functions such as user's assessment, matching strategy, intelligent recommendation, to have the beneficial effect for saving and promoting cost and promoting registration conversion ratio, user intent assessment simultaneously can provide significant data foundation for subsequent recommendation or other operations, establish solid foundation to excavate potential user.
Description
Technical field
The present invention relates to Internet technical field, more particularly to it is a kind of can be promoted conversion ratio for user will be lost
Intelligent recommendation system.
Background technique
With the development of technology and the improvement of people ' s living standards, have become must not in people's daily life for internet
One of the tool that can lack.
Nowadays, internet obtain objective mode be mainly by the dispensing of each advertising channel, allow user see company brand or
Then the exposing of content clicks advertisement by user, page is landed in the popularization of guidance user to company's site, and then in landing page
User's registration is guided to become company's site member.
Today of high speed development in internet, the cost that user obtains visitor is higher and higher, and promotion expense and day increase severely, so that mentioning
It rises registration conversion ratio and is equivalent to saving company cost, and create value.So how that hard-won website traffic is maximum
It is converted into degree the validated user of website, and promotes registration conversion ratio, becomes the important examination of current each company operation
Index.
Currently, mainly including the following prior art for how to promote traffic transformation rate:
1. promoting registration user volume by increasing advertisement dispensing.
2. shortening user's operation path, for example registration information needs to fill out more than ten and could complete to register originally, can shorten
Complete to register only to fill out cell-phone number and password, and other user informations are put into succeed in registration after allow user to improve letter again
Breath.
3. opening speed and the intention sense of optimization landing page, allow user that can quickly open the page, and generate dense
Interest converts to increase registration.
But there are still certain defects for the prior art, such as:
1. registration amount can be promoted really by increasing advertising input, but cost also greatly promotes, and conversion ratio is not
Certain proportional promotion.
2. by optimization registered paths, and promoting the user experience of the page, can be registered really with lift portion.But the skill
Art is directed to the stronger user of registration intention using conventional thinking, so final promotion effect is not obvious.
3. if the registration intention of user is not strong, it will usually which landing page is directly closed or jumped out in selection, and the prior art can not
Retrieve the loss of this kind of user.
Summary of the invention
In order to solve it is above-mentioned obtain that objective cost is big, effect is undesirable, can not retrieve the problems such as user, the present invention provides a kind of use
In retrieving the intelligent recommendation system that will be lost user, increase registration amount different from existing shellfish increasing simply, but
Intention can be guided not strong to the maximum extent or the user left that determined, and pass through on the basis of not increasing dispensing cost
The means such as excitation promote these users to be converted into website registration user, to promote conversion ratio, create maximum value.
It is above-mentioned a kind of for retrieving the intelligent recommendation system that will be lost user, comprising:
Intention evaluation module, the user behaviors log of the intention evaluation module acquisition user, and show that user goes about registration
For intention score value;
Tactful configuration module, the strategy configuration module configures a variety of marketing incentives rules, and checks that each strategy is brought
Effect statistics;
Recommending module, the recommending module is according in the intention score value in intention evaluation module, and tactful configuration module
Strategy protocol, generate motivate simultaneously be presented to the user, excitation user complete registration.
In above system, unique user's mark is generated for user when user accesses landing page in the intention evaluation module
Know, and the user identifier is implanted in user browser by landing page.
In above system, the intention evaluation module carries out score value assessment according to the user behaviors log of user, and to the row
Classification and filing processing are carried out for log.
In above system, the user behaviors log of global push user is carried out without a technology of burying using front end, so that log is circulated
Interface is crossed to pass to back-end server and be stored in data processing platform (DPP) after receipt.
In above system, the point value of evaluation of the intention evaluation module is 5 grades of score values.
In above system, the strategy protocol of the strategy configuration module configuration is to trigger primary effective or special time period
Interior same user triggers in vain again.
In above system, it is described strategy configuration module configuration strategy protocol include in limited time rob, page jump distribution and
Send certificate.
In above system, the trigger condition of the recommending module includes:
User occurs to close or jump out landing page line, or
Intention evaluation module monitors that the intention score value of user reaches the rule and policy of tactful configuration module configuration.
In above system, the recommending module is according to the intention score value of intention evaluation module and the plan of tactful configuration module
Slightly scheme, matches optimal policy scheme, and optimal policy scheme is transmitted to landing page by interface.
In above system, optimal policy scheme is presented to the user by the landing page, and user is motivated to complete registration.
The present invention by using intention evaluation module, tactful configuration module and recommending module, realize user's assessment,
With functions such as strategy, intelligent recommendations, have the advantage that
1. the present invention can custom-configure strategy protocol in backstage visualization interface, and can inquire each strategy side
Case bring effect, can flexible expansion.
2. intention evaluation module and recommending module are to be completely independent module, so as to not invade original system, but
It is only exported in the form of API (application programming interface), convenient for decoupling.
In turn, the present invention have it is following the utility model has the advantages that
Cost is promoted 1. saving, in the case where not increasing new cost, 4 percentage points of conversion ratio is promoted, is equivalent to saving
7% popularization input cost.
2. user intent is evaluated as subsequent recommendation or other operations provide significant data foundation, established to excavate potential user
Solid foundation.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the structural block diagram of the intelligent recommendation system of the embodiment of the present invention;
Fig. 2 is the workflow schematic diagram of the intention evaluation module in the intelligent recommendation system of the embodiment of the present invention.
Specific embodiment
With reference to the accompanying drawings and examples, further description of the specific embodiments of the present invention.Following embodiment is only
For clearly illustrating technical solution of the present invention, and not intended to limit the protection scope of the present invention.
As shown in Figure 1, this invention describes a kind of for retrieving the intelligent recommendation system that will be lost user, mainly include
Intention evaluation module, tactful configuration module and recommending module.Wherein:
Intention evaluation module:
The module is responsible for acquiring the user behaviors log of user, and obtains intention score value of the user about registration behavior.Work as user
When accessing landing page by equipment, intention evaluation module can be believed according to facility information, user sources ip, source advertising channel etc.
Breath, is generated for user a unique user identifier yrd_uid, for distinguishing different access users.Then, which is marked
Know yrd_uid and is implanted in the cookie of user browser by landing page.Operation behavior day of the later user in landing page
Will can be filed by user identifier yrd_uid.
The assessment foundation of intention evaluation module is the user behaviors log of user, so when collecting the user behaviors log of user,
Carry out classification and filing processing to it by intention evaluation module first, such as:
It is sorted by time, the user for counting the same yrd_uid has accessed the total degree of landing page, the day of connected reference
Number, the number when accessed are most high.For example classify by operation trace, the user of the same yrd_uid is counted in landing page point
Which button and input frame do not put, what content inputted, come the point of interest of counting user which, user more concerned with content
What is.By classification to User action log parsing and modeling, the intention of registering customers as is assessed as from 1 to 5 totally five grades
Not, and score value is bigger, and it is stronger to represent intention.
As shown in Fig. 2, intention evaluation module is mainly according to User action log, in addition algorithm model obtains user to integrate
Registration intention score value.
Wherein, the collection process of User action log pushes User action log without a technology overall situation is buried by front end, makes
Data processing platform (DPP) can be first put by obtaining the log stream that log stream is passed to back-end server, and be received by interface, at the data
Platform is preferably that (a kind of distributed post subscription message system of high-throughput, it can handle consumer's scale to Kafka
Everything flow data in website.For the daily record data and off-line analysis system as Hadoop, but require real-time
The limitation of processing, this is a feasible solution.The purpose of Kafka is by the loaded in parallel mechanism of Hadoop come unified
On line and offline Message Processing, also for providing real-time message by cluster).Log Source of the Kafka as collection,
Carry out classification processing for subsequent tag module, by multiple algorithm process, by data mart modeling at the finished product data of tape label, persistence
To in DB (Database, database).Specifically includes the following steps:
1. the full dose for carrying out log is collected, collected data are put into Kafka, as source data to be processed.
2. the different event and Else Rule that generate according to log configure, primary filtration, cleaning are carried out to log, extraction has
With data and it is stored in different redis-db, it is stand-by as preliminary working data.
3. the primary data in redis-db is passed through many algorithms operation, is processed into band according to algorithm model and demand
A kind of finished product data of tag attributes, then by the storage of finished product data persistence into DB.
Tactful configuration module:
The module is mainly responsible for a variety of incentive programs of configuration, and checks each tactful bring effect statistics.I.e., it is possible to root
Flexible strategy protocol is needed to configure according to business, including such as send certificate, send coin, a gift of money at Chinese New Year, robs in limited time, page jump distribution side
Case.And same user in the primary effective or special time period of triggering can be configured and trigger (same use in such as 3 days in vain again
It triggers again in vain at family).It is configured by strategy, not only reaches arousal effect, but also can prevent the user of some invalid brush pages from coming
Incentive policy is enjoyed, is had the characteristics that safely and flexibly.
Recommending module:
The recommending module is according to the strategy side in the intention score value in intention evaluation module, and tactful configuration module
Case generates and motivates and be presented to the user, and excitation user completes registration.
Recommending module in the present invention can't destroy the normal registration process of landing page, but only when user closes
Or when jumping out behavior or when the intention score value of system monitoring to user reach the rule and policy of setting, it can just trigger recommending module.
Meanwhile recommending module can match according to the strategy in the intention score value and tactful configuration module in intention evaluation module
It sets, to match current optimal policy scheme, and optimal policy scheme is transmitted to by landing page by interface, then will by landing page
Optimal policy scheme is presented to the user, to motivate user to complete registration.Specific step is as follows:
When user is when landing page reaches triggering rule, landing page can send one to the interface of recommending module and request, and
User identifier yrd_uid is reached into server-side.When recommending module receives request, according to the yrd_uid parameter received, first
Corresponding intention score value is taken out in intention evaluation module, then further according to intention score value, is found and is worked as from tactful configuration module
Preceding optimal policy scheme.After the combination of the parameter informations such as user identifier yrd_uid, intention score value, then will be optimal by interface
Strategy protocol returns to landing page, lands page for the content of the optimal policy scheme received and shows user, to motivate use
Family registration.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Any modification, equivalent replacement, improvement and so within principle, should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of for retrieving the intelligent recommendation system that will be lost user characterized by comprising
Intention evaluation module, the user behaviors log of the intention evaluation module acquisition user, and obtain user about registration behavior
Intention score value;
Tactful configuration module, the strategy configuration module configure a variety of marketing incentives rules, and check each tactful bring effect
Fruit statistics;
Recommending module, the recommending module is according to the plan in the intention score value in intention evaluation module, and tactful configuration module
Slightly scheme generates and motivates and be presented to the user, and excitation user completes registration.
2. intelligent recommendation system according to claim 1, which is characterized in that the intention evaluation module is fallen in user's access
When ground page, unique user identifier is generated for user, and the user identifier is implanted in user browser by landing page.
3. intelligent recommendation system according to claim 2, which is characterized in that row of the intention evaluation module according to user
Score value assessment is carried out for log, and classification and filing processing are carried out to the user behaviors log.
4. intelligent recommendation system according to claim 3, which is characterized in that carry out the overall situation without a technology of burying using front end and push away
The user behaviors log of user is sent, is put down so that log stream passes to back-end server by interface and is stored in data processing after receipt
Platform.
5. intelligent recommendation system according to claim 4, which is characterized in that the point value of evaluation of the intention evaluation module is
5 grades of score values.
6. intelligent recommendation system according to claim 1, which is characterized in that the strategy side of the strategy configuration module configuration
Case is to trigger same user in primary effective or special time period to trigger again in vain.
7. intelligent recommendation system according to claim 6, which is characterized in that the strategy side of the strategy configuration module configuration
Case includes robbing in limited time, page jump distribution and sending certificate.
8. intelligent recommendation system according to claim 1, which is characterized in that the trigger condition of the recommending module includes:
User occurs to close or jump out landing page line, or
Intention evaluation module monitors that the intention score value of user reaches the rule and policy of tactful configuration module configuration.
9. intelligent recommendation system according to claim 8, which is characterized in that the recommending module is according to intention evaluation module
Intention score value and tactful configuration module strategy protocol, match optimal policy scheme, and by interface by optimal policy
Scheme is transmitted to landing page.
10. intelligent recommendation system according to claim 9, which is characterized in that optimal policy scheme is in by the landing page
User is now given, and user is motivated to complete registration.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810917678.4A CN109146562A (en) | 2018-08-13 | 2018-08-13 | It is a kind of for retrieving the intelligent recommendation system that will be lost user |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810917678.4A CN109146562A (en) | 2018-08-13 | 2018-08-13 | It is a kind of for retrieving the intelligent recommendation system that will be lost user |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109146562A true CN109146562A (en) | 2019-01-04 |
Family
ID=64792799
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810917678.4A Pending CN109146562A (en) | 2018-08-13 | 2018-08-13 | It is a kind of for retrieving the intelligent recommendation system that will be lost user |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109146562A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111787345A (en) * | 2020-07-07 | 2020-10-16 | 北京达佳互联信息技术有限公司 | Interactive resource processing method and device based on network live broadcast room, server and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002149598A (en) * | 2000-11-07 | 2002-05-24 | Sony Corp | User register promoting system, information picture managing server, information picture managing program recording medium, information managing server, information managing program recording medium, client computer and client program recording medium |
CN105761093A (en) * | 2014-12-17 | 2016-07-13 | 华为技术有限公司 | Knowledge-space-based behavior result evaluation method and device |
CN107016592A (en) * | 2017-03-08 | 2017-08-04 | 美的集团股份有限公司 | Home appliance based on application guide page recommends method and apparatus |
CN107273436A (en) * | 2017-05-24 | 2017-10-20 | 北京京东尚科信息技术有限公司 | The training method and trainer of a kind of recommended models |
CN107369052A (en) * | 2017-08-29 | 2017-11-21 | 北京小度信息科技有限公司 | User's registration behavior prediction method, apparatus and electronic equipment |
-
2018
- 2018-08-13 CN CN201810917678.4A patent/CN109146562A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002149598A (en) * | 2000-11-07 | 2002-05-24 | Sony Corp | User register promoting system, information picture managing server, information picture managing program recording medium, information managing server, information managing program recording medium, client computer and client program recording medium |
CN105761093A (en) * | 2014-12-17 | 2016-07-13 | 华为技术有限公司 | Knowledge-space-based behavior result evaluation method and device |
CN107016592A (en) * | 2017-03-08 | 2017-08-04 | 美的集团股份有限公司 | Home appliance based on application guide page recommends method and apparatus |
CN107273436A (en) * | 2017-05-24 | 2017-10-20 | 北京京东尚科信息技术有限公司 | The training method and trainer of a kind of recommended models |
CN107369052A (en) * | 2017-08-29 | 2017-11-21 | 北京小度信息科技有限公司 | User's registration behavior prediction method, apparatus and electronic equipment |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111787345A (en) * | 2020-07-07 | 2020-10-16 | 北京达佳互联信息技术有限公司 | Interactive resource processing method and device based on network live broadcast room, server and storage medium |
CN111787345B (en) * | 2020-07-07 | 2022-04-01 | 北京达佳互联信息技术有限公司 | Interactive resource processing method and device based on network live broadcast room, server and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103559208B (en) | The ranking fraud detection method of application program and ranking fraud detection system | |
CN104636501B (en) | A kind of artificial instant translation system of multi-media network and method | |
US8156172B2 (en) | Monitoring and reporting enterprise data using a message-based data exchange | |
CN108205486A (en) | A kind of intelligent distributed call chain tracking based on machine learning | |
CN108009926A (en) | Method, information processor and readable storage medium storing program for executing for user's classification | |
CN107066512A (en) | A kind of user preference appraisal procedure and system based on Hadoop | |
CN102056351B (en) | Push service system and method | |
CN111708774B (en) | Industry analytic system based on big data | |
CN109670843A (en) | Data processing method, device, computer equipment and the storage medium of complaint business | |
CN101500017A (en) | Method for providing service based on flow and system thereof | |
CN112184302A (en) | Product recommendation method and device, rule engine and storage medium | |
CN108171527A (en) | Management System for Clients Information and method | |
CN106991577A (en) | A kind of method and device for determining targeted customer | |
CN109635089B (en) | Literature work novelty evaluation system and method based on semantic network | |
CN107767253A (en) | A kind of tax information management platform, method and system | |
CN110648172A (en) | Identity recognition method and system fusing multiple mobile devices | |
CN111489190A (en) | Anti-cheating method and system based on user relationship | |
CN107705227A (en) | A kind of network system for being used to provide law financial service | |
CN109146562A (en) | It is a kind of for retrieving the intelligent recommendation system that will be lost user | |
CN109544271A (en) | A kind of trade managing system | |
TWI448980B (en) | Information processing devices, information processing methods, and information processing programs | |
CN108764990A (en) | One kind sending out voice red packet software based on wechat small routine | |
CN103559210B (en) | The ranking fraud detection method and ranking fraud detection system of application program | |
CN108960307A (en) | A kind of proprietary protocol field format extracting method, device and server | |
CN108257042A (en) | A kind of task-driven type intelligent electricity selling business support system |
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: 20190104 |