CN111177571B - Value-driven dynamic recommendation system with multi-factor dimension space and multi-mesoscale fusion - Google Patents
Value-driven dynamic recommendation system with multi-factor dimension space and multi-mesoscale fusion Download PDFInfo
- Publication number
- CN111177571B CN111177571B CN202010032685.3A CN202010032685A CN111177571B CN 111177571 B CN111177571 B CN 111177571B CN 202010032685 A CN202010032685 A CN 202010032685A CN 111177571 B CN111177571 B CN 111177571B
- Authority
- CN
- China
- Prior art keywords
- user
- data
- module
- dimension
- time
- 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
- 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)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a value-driven dynamic recommendation system with multi-factor dimension space and multi-mesoscale fusion, which infers the purpose of a user through a data acquisition and processing module, a statistical learning module, a purpose analysis logic judgment module, a recommendation management module and a setting operation module of a control system, analyzes related steps according to the purpose of the user, substitutes a user thought cost calculation formula, takes a recommendation scheme with the minimum result as a final result to recommend the user, fuses the purpose of the user, minimizes the thought cost, maximizes the user comfort experience, and solves the problems of low comfort level, low efficiency and the like caused by the fact that the user usually spends a large amount of time and experiences to achieve the purpose.
Description
Technical Field
The invention discloses a value-driven dynamic recommendation system with multi-factor dimension space and multi-mesoscale fusion, and belongs to the field of informatization.
Background
At the moment of the technology change day by day, people can do a lot of things by using the mobile phone, but under the common condition, the intelligent degree of the mobile phone is low, the content suitable for the current purpose of the user cannot be recommended to the user in time, the user usually needs to spend a lot of time and energy to find a certain person needing to be contacted, or the user needs to complete subscription step by step according to the own needs when arranging a journey, and a large amount of irrelevant content needs to be read in the process, so the thinking cost is high, the efficiency is low, and the comfort feeling of the user is low. The method aims to collect and sort a large amount of data, reason out the purpose of the user, calculate the related recommendation scheme according to the purpose of the user, finally obtain the recommendation scheme meeting the minimum thought cost, improve the efficiency of handling things by the user and improve the comfort feeling of the user.
Disclosure of Invention
The technical problem is as follows: the intelligent degree of the mobile phone is low, content suitable for the current purpose of the user cannot be recommended to the user in time, the user usually needs to spend a lot of time and energy to find a certain person needing to be contacted, or the user needs to complete subscription step by step according to own needs when arranging a route, and a large amount of irrelevant content needs to be read in the process, so that the thinking cost is high, the efficiency is low, and the comfort feeling of the user is low.
The technical scheme is as follows: aiming at the problems, the invention designs a value-driven dynamic recommendation system with multi-factor dimensional space and multi-medium scale fusion, collects and arranges a large amount of data, and infers the purpose of a user.
The system structure is as follows:
the utility model provides a dynamic recommendation system that many mesoscale of multi-factor dimension space of value driven fuses which characterized in that fuses many mesoscale of multi-factor dimension space, optimizes the recommendation process under the value drive, carries out dynamic adjustment in real time, reduces user's thinking cost, and maximize user comfort experience, and specific step is:
s1: collecting information data of different dimensions of a user, and integrating known information;
s2: carrying out statistical learning and logic judgment on the acquired data to obtain the user purpose;
s3: and recommending according to the user purpose.
The system also comprises five modules, namely a data acquisition and processing module, a statistical learning module, a target analysis logic judgment module, a recommendation management module and a setting and running module.
Wherein step 1 further comprises using a data acquisition and processing module;
the data acquisition and processing module of the system comprises the following operation steps: the system carries out multi-dimensional data acquisition, and the multi-dimensional data comprises data in a time dimension, data in a space dimension, data in an event dimension and data in a basic information dimension, wherein dimensions are supposed to be distinguishable; after the user authorization, the data collected by the system takes time dimension and space dimension as marks and combines with the user basic information data, the user basic information data comprises the information of occupation, working time, working place and home address of the user, and the user basic data is crossed with the time dimension and the space dimension; the data in the event dimension is collected in the time dimension and the space dimension, and the time, the geographic position and the specific event activity process of starting and ending the activity event comprise that the user uses an APP or a small program, searches for a content keyword of a bar and clicks and views the content keyword when the user uses the equipment to perform the activity each time; according to the needs of users, the system integrates and processes known information data, and the method comprises the steps of summarizing the activities of the users into a table according to time, geographic positions and the activity processes of specific events of the users, classifying the activities into working time and non-working time according to time, classifying the activities into working addresses, home addresses and other addresses according to the geographic positions, marking the activities in the table, and determining keywords of the activities according to the activity processes of the specific time of the users, wherein the keywords are set by the users according to the conditions of the users; according to the reuse characteristic of data, the collected user basic information is not subjected to a new collection task unless the user selects to modify, and the data is reused when the related information is used.
The step 2 also comprises a statistical learning module and a target analysis logic judgment module;
the statistical learning module of the system comprises the following operation steps: inputting the data collected and processed by the data collecting and processing module into a statistical learning module, performing statistical learning according to different dimensions, for example, classifying the event data of the activity into working time and non-working time according to the time dimension and taking 30 days as a period, and then performing statistics of the times of the occurrence of the keywords of different activity purposes in the working time and the non-working time to obtain the probability of the occurrence of different keyword activities in different times, and also performing statistical mapping to a more subdivided time period [ t [ [ t ]1，t2]Probability p of occurrence of different keyword activitiesi[t1,t2]I represents different keywords, the statistical results are sorted from large to small according to the probability and are gathered into a table, and the probability p of different keyword activities in different geographic environments can be mapped by the same methodi(dj)I represents different keywords, dj represents different geographic environments, and statistical results are summarized into a table; as a result of statisticsImportant reference factors for the purpose of logic judgment of analysis;
the operation steps of the target analysis logic judgment module of the system are as follows: firstly, according to a summary table of different keyword occurrence probabilities divided by different dimensions obtained from a statistical learning module, the summary table is used as one of important conditions for judging user purposes to judge the possible purposes of the user, and pop for the possible purposes of the user is mapped by time dimension1Representing possible destination pop of users mapped by spatial dimensions2Denotes, pop1And pop2Is the event represented by the maximum probability in the statistical table; further, the current activity of the user, i.e. the APP or applet used by the user, is obtained in time, and the possible purpose of the user is further deduced, namely the pop3Represents; still further, search keywords by the user using APP or applets are obtained, the user's possible purpose being pop4Represents; the final possible purpose of the user is given by the formula: pop (point of Place)End=a*pop1+b*pop2+c*pop3+d*pop4+ α determination, popEndThe final possible purpose is obtained, a, b, c and d are influence factors and can also be used as weights, alpha is other influence factors and is used for determining the influence degree of the possible purpose of the user obtained through different steps, theoretically, 0 is more than or equal to a, b, c and d are less than or equal to 1, the clearer the content input by the user is, and the clearer the final purpose of the user can be obtained through reasoning.
Wherein, the step 3 also comprises a use recommendation management module and a setting operation module; the recommendation management aims at recommending a series of services for a user to shorten the distance between the user and the user, and the distance can be measured by the scale and the times of sliding a hand on a screen and the level of clicking;
the recommendation management module of the system comprises the following operation steps: the method comprises the following steps of carrying out recommendation management according to a final user purpose calculated by a purpose analysis logic judgment module, analyzing a possible final user purpose, analyzing what steps and associated contents need to be carried out to achieve the purpose, giving analysis steps according to logic according to organization, expanding the organization range step by step, listing the steps to be completed to achieve the user purpose, carrying out recommendation management on the basis, and forming different recommendation management schemes to form a network; the reduction of the thinking cost of the user is determined by the front and back sequence of recommendation management, if the content recommended by the system meets the user expectation, the thinking cost of the user is reduced, and the thinking cost of the user, which is the value, is represented by the formula: the distance _ value = e x + f y + g z + α, x represents the scale of the hand sliding on the screen, y represents the number of times of clicking the screen, z represents the level of clicking, α is other factors, e, f and g are influence factors and can also be used as weights; calculating, and taking a recommended management scheme with the minimum thinking cost as a final result;
the operation steps of the setting and running module of the system are as follows: and pushing the information to the user according to the recommended management scheme of the system in steps.
The method also comprises the steps of protecting the privacy of the user to the maximum extent, and collecting all data after the consent of the user; the more sufficient the user data obtained earlier, the more accurate the recommendation of the system.
The invention has the following beneficial effects:
1. the multi-factor dimension space and multi-medium scale are fused, the recommendation process is optimized under the drive of value, dynamic adjustment is carried out in real time, the thinking cost of a user is reduced, and the comfort experience of the user is maximized;
2. through data support, the purpose of a user is determined, the thinking cost of the user is quantified, the optimal scheme recommended by the user is determined, and the comfort level and the efficiency of the user are improved;
3. the privacy of the user is protected to the maximum extent, and all data acquisition is agreed by the user.
Drawings
FIG. 1 is a detailed flow diagram of a value-driven dynamic recommendation system for multi-dimensional spatial multi-mesoscale fusion.
The specific implementation mode is as follows:
a value-driven dynamic recommendation system for multi-factor dimension space multi-medium scale fusion is characterized in that the multi-factor dimension space multi-medium scale fusion is carried out, a recommendation process is optimized under the drive of a value, dynamic adjustment is carried out in real time, thinking cost of a user is reduced, user comfort experience is maximized, and specific steps of the scheme are explained by examples:
example one: the user worked as a corporate finance and routinely performed financial posting checks on wednesday at 10:30 am, followed by reporting in the corporate group at 11:30 noon.
Collecting information data of different dimensions of a user, and integrating known information;
after user authorization, multi-dimensional data are collected, which are respectively:
the time dimension is as follows: wednesday morning at 10: 45;
spatial dimension: a corporate office;
basic information dimension: the occupation is company finance, the financial statement check is routinely carried out at 10:30 am on wednesday, and then the statement is carried out in a company group at noon of 11:30 noon;
the system also comprises all data information collected by the user since the user uses the system, the data collected by the system takes time dimension and space dimension as marks and combines with user basic information data, the user basic information data comprises information of career, working time, working place and home address of the user, and at the moment, user basic data is crossed with the time dimension and the space dimension; the data in the event dimension is collected in the time dimension and the space dimension, and the time, the geographic position and the specific event activity process of starting and ending the activity event comprise that the user uses an APP or a small program, searches for a content keyword of a bar and clicks and views the content keyword when the user uses the equipment to perform the activity each time;
according to the user requirement, the system carries out the integration processing work of the known information data: summarizing the activities of a user into a table according to time, geographic positions and the activity process of specific events of the user, classifying the activities into working time and non-working time according to the time, classifying the activities into working addresses, home addresses and other addresses according to the geographic positions, marking the activities in the table, and determining a keyword of an activity target according to the activity process of the specific time of the user, wherein the keyword is set by the user according to the self condition; according to the reuse characteristic of data, the collected user basic information is not subjected to a new collection task unless the user selects to modify, and the data is reused when the related information is used.
Carrying out statistical learning and logic judgment on the acquired data to obtain the user purpose;
inputting the data collected and processed by the data collecting and processing module into a statistical learning module, performing statistical learning according to different dimensions, for example, classifying the event data of the activity into working time and non-working time according to the time dimension and taking 30 days as a period, and then performing statistics of the times of the occurrence of the keywords of different activity purposes in the working time and the non-working time to obtain the probability of the occurrence of different keyword activities in different times, and also performing statistical mapping to a more subdivided time period [ t [ [ t ]1，t2]Probability p of occurrence of different keyword activitiesi[t1,t2]I represents different keywords, the statistical results are sorted from large to small according to the probability and are gathered into a table, and the probability p of different keyword activities in different geographic environments can be mapped by the same methodi(dj)I represents different keywords, dj represents different geographic environments, and statistical results are summarized into a table; the statistical result is used as an important reference factor for the purpose of analyzing logic judgment;
the user sets keywords of 10:30 in the morning of Wednesday as financial statistics report, the time of using the mobile phone by the user is 10:45 in the morning of Wednesday, the probability of the keywords being financial statistics report is 85 percent, and the probability is taken as a possible destination pop1(ii) a The user is positioned as a company office, and the probability that the keyword is work is 95 percent as known by a spatial dimension probability statistical table, so that the probability is taken as a pop for possible purpose2(ii) a Further, the user opens the mobile phone WeChat, and the inference user aims at reporting work, which is called pop3Represents; and (3) carrying out user final purpose reasoning according to the contents, and substituting into a user final possible purpose formula: pop (point of Place)End=a*pop1+b*pop2+c*pop3+d*pop4And the + alpha, a, b, c and d are influence factors and can also be used as weights, alpha is other influence factors, and the final purpose of the user is calculated to open the WeChat company group for reporting.
Recommending according to the user purpose:
the user aims to open a WeChat company group for reporting work, and a series of operations to be completed are as follows: opening the WeChat, finding a company group, finding a report file, reporting and closing the WeChat; substituting the scale of sliding on the screen, the number of times of clicking the screen and the clicked-in level for completing the series of work into a user thinking cost source _ value formula: the method comprises the following steps that (1) the result of thinking cost of a user is obtained by determining the value of thought _ value = e x + f y + g z + alpha, x represents the scale of sliding a hand on a screen, y represents the number of times of clicking the screen, z represents the level of clicking, alpha is other factors, e, f and g are influence factors and can also be used as weights, and the result of thinking cost of the user is obtained, and the recommendation management scheme with the minimum thinking cost is taken as the final result;
and finally, pushing the information to the user according to the recommended management scheme of the system.
If the user report completion time is 11:30, the probability that the keyword is noon break is 93% as known from the time dimension probability statistical table, and the probability is taken as the pop of possible purpose1(ii) a The user is positioned as a company office, and the probability that the keyword is noon break is 95 percent as known by a spatial dimension probability statistical table and is taken as a possible destination pop2(ii) a Further, the user opens the ordering software to reason about the user's intent to eat, by pop3Represents; and (3) carrying out user final purpose reasoning according to the contents, and substituting into a user final possible purpose formula: pop (point of Place)End=a*pop1+b*pop2+c*pop3+d*pop4+ α, a, b, c, d are influence factors, which can also be used as weights, α is other influence factors, and the final purpose of the user is calculated to be eating.
The final purpose of the user is to eat, and the related work is as follows: opening the ordering software, selecting a restaurant, selecting a set meal, filling an address, paying, and closing the ordering software; substituting the scale of sliding on the screen, the number of times of clicking the screen and the clicked-in level for completing the series of work into a user thinking cost source _ value formula: the method comprises the following steps that (1) the result of thinking cost of a user is obtained by determining the value of thought _ value = e x + f y + g z + alpha, x represents the scale of sliding a hand on a screen, y represents the number of times of clicking the screen, z represents the level of clicking, alpha is other factors, e, f and g are influence factors and can also be used as weights, and the result of thinking cost of the user is obtained, and the recommendation management scheme with the minimum thinking cost is taken as the final result;
and finally, pushing the menu to the user according to the recommended management scheme of the system, automatically starting the menu ordering software through a user mobile phone interface, finding out the house and the signboard of the restaurant concerned by the user, automatically filling the address for payment after the user selects the menu, and closing the menu ordering software after the menu ordering software is successful.
Claims (1)
1. The utility model provides a dynamic recommendation system that many mesoscales of multifactor dimension space fuse based on value drive, its characterized in that fuses many mesoscales of multifactor dimension space, optimizes the recommendation process under the value drive, carries out dynamic adjustment in real time, reduces user's thinking cost, and maximize user comfort experiences specifically does:
s1: designing a data acquisition and processing module facing multi-factor dimension space and multi-mesoscale, acquiring information data of different dimensions of a user, and performing fusion processing;
s2: the design statistical learning module carries out statistical learning on the acquired data, the occurrence probability of different keywords divided by different dimensions is used, and the purpose of the user is deduced according to the probability through the purpose analysis logic judgment module so as to realize logic judgment;
s3: designing a recommendation management module, analyzing according to the user purpose, realizing value-driven recommendation based on the user thinking cost, and finally recommending by setting an operation module;
it is characterized in that step S1 also includes using a data acquisition and processing module;
the data acquisition and processing module of the system comprises the following operation steps: the system carries out multi-dimensional data acquisition, and the multi-dimensional data comprises data in a time dimension, data in a space dimension, data in an event dimension and data in a basic information dimension, wherein dimensions are supposed to be distinguishable; after the user authorization, the data collected by the system takes time dimension and space dimension as marks and combines with the user basic information data, the user basic information data comprises the information of occupation, working time, working place and home address of the user, and the user basic data is crossed with the time dimension and the space dimension; the data in the event dimension is collected in the time dimension and the space dimension, and the time, the geographic position and the specific event activity process of starting and ending the activity event comprise that the user uses an APP or a small program, searches for a content keyword of a bar and clicks and views the content keyword when the user uses the equipment to perform the activity each time; according to the needs of users, the system integrates and processes known information data, and the method comprises the steps of summarizing the activities of the users into a table according to time, geographic positions and the activity processes of specific events of the users, classifying the activities into working time and non-working time according to time, classifying the activities into working addresses, home addresses and other addresses according to the geographic positions, marking the activities in the table, and determining keywords of the activities according to the activity processes of the specific time of the users, wherein the keywords are set by the users according to the conditions of the users; according to the reuse characteristic of data, the acquired user basic information is not subjected to a new acquisition task unless the user selects to modify, and the data is reused when the related information is used;
the method is characterized in that the step S2 also comprises a statistical learning module and a target analysis logic judgment module;
the statistical learning module of the system comprises the following operation steps: inputting the data collected and processed by the data collecting and processing module into a statistical learning module, performing statistical learning according to different dimensions, for example, classifying the event data of the activity into working time and non-working time according to the time dimension and taking 30 days as a period, and then performing statistics of the times of the occurrence of the keywords of different activity purposes in the working time and the non-working time to obtain the probability of the occurrence of different keyword activities in different times, and also performing statistical mapping to a more subdivided time period [ t [ [ t ]1，t2]Probability p of occurrence of different keyword activitiesi[t1,t2]I represents different keywords, the statistical results are sorted from large to small according to the probability and are gathered into a table, and the probability p of different keyword activities in different geographic environments can be mapped by the same methodi(dj)I represents different keywords, dj represents different geographic environments, and statistical results are summarized into a table; the statistical result is used as an important reference factor for the purpose of analyzing logic judgment;
wherein the purpose analysis logic of the system judges the module operationThe method comprises the following steps: firstly, according to a summary table of different keyword occurrence probabilities divided by different dimensions obtained from a statistical learning module, the summary table is used as one of important conditions for judging user purposes to judge the possible purposes of the user, and pop for the possible purposes of the user is mapped by time dimension1Representing possible destination pop of users mapped by spatial dimensions2Denotes, pop1And pop2Is the event represented by the maximum probability in the statistical table; further, the current activity of the user, i.e. the APP or applet used by the user, is obtained in time, and the possible purpose of the user is further deduced, namely the pop3Represents; still further, search keywords by the user using APP or applets are obtained, the user's possible purpose being pop4Represents; the final possible purpose of the user is given by the formula: pop (point of Place)End=a*pop1+b*pop2+c*pop3+d*pop4+ α determination, popEndThe final possible purpose is obtained, a, b, c and d are influence factors and can also be used as weights, alpha is other influence factors and is used for determining the influence degree of the possible purpose of the user obtained through different steps, a, b, c and d are more or less than or equal to 0 and less than or equal to 1, the clearer the content input by the user is, the clearer the final purpose of the user can be obtained through reasoning;
the method is characterized in that the step S3 also comprises a use recommendation management module and a setting operation module; the recommendation management aims at recommending a series of services for a user to shorten the distance between the user and the user, and the distance can be measured by the scale and the times of sliding a hand on a screen and the level of clicking;
the recommendation management module of the system comprises the following operation steps: the method comprises the following steps of carrying out recommendation management according to a final user purpose calculated by a purpose analysis logic judgment module, analyzing a possible final user purpose, analyzing what steps and associated contents need to be carried out to achieve the purpose, giving analysis steps according to logic according to organization, expanding the organization range step by step, listing the steps to be completed to achieve the user purpose, carrying out recommendation management on the basis, and forming different recommendation management schemes to form a network; the reduction of the thinking cost of the user is determined by the front and back sequence of recommendation management, if the content recommended by the system meets the user expectation, the thinking cost of the user is reduced, and the thinking cost of the user, which is the value, is represented by the formula: the distance _ value = e x + f y + g z + α, x represents the scale of the hand sliding on the screen, y represents the number of times of clicking the screen, z represents the level of clicking, α is other factors, e, f and g are influence factors and can also be used as weights; calculating, and taking a recommended management scheme with the minimum thinking cost as a final result;
the operation steps of the setting and running module of the system are as follows: and pushing the information to the user according to the recommended management scheme of the system in steps.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010032685.3A CN111177571B (en) | 2020-01-13 | 2020-01-13 | Value-driven dynamic recommendation system with multi-factor dimension space and multi-mesoscale fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010032685.3A CN111177571B (en) | 2020-01-13 | 2020-01-13 | Value-driven dynamic recommendation system with multi-factor dimension space and multi-mesoscale fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111177571A CN111177571A (en) | 2020-05-19 |
| CN111177571B true CN111177571B (en) | 2020-11-27 |
Family
ID=70650977
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010032685.3A Active CN111177571B (en) | 2020-01-13 | 2020-01-13 | Value-driven dynamic recommendation system with multi-factor dimension space and multi-mesoscale fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111177571B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106803190A (en) * | 2017-01-03 | 2017-06-06 | 北京掌阔移动传媒科技有限公司 | A kind of ad personalization supplying system and method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101747303B1 (en) * | 2011-04-11 | 2017-06-15 | 삼성전자주식회사 | Method for recommendation of application, mobile terminal thereof and communication system thereof |
| US20150073932A1 (en) * | 2013-09-11 | 2015-03-12 | Microsoft Corporation | Strength Based Modeling For Recommendation System |
| CN104063429A (en) * | 2014-06-11 | 2014-09-24 | 深圳德协保税电子商务有限公司 | Predicting method for user behavior in e-commerce |
| CN108230007B (en) * | 2017-11-28 | 2021-09-10 | 北京三快在线科技有限公司 | User intention identification method and device, electronic equipment and storage medium |
| CN109634488A (en) * | 2018-12-24 | 2019-04-16 | 上海碳蓝网络科技有限公司 | A kind of method and apparatus using entrance preferentially applied for rendering |
| CN110334284B (en) * | 2019-04-03 | 2021-07-30 | 厦门大学 | Novel recommendation method integrating multi-factor decision making |
-
2020
- 2020-01-13 CN CN202010032685.3A patent/CN111177571B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106803190A (en) * | 2017-01-03 | 2017-06-06 | 北京掌阔移动传媒科技有限公司 | A kind of ad personalization supplying system and method |
Non-Patent Citations (1)
| Title |
|---|
| 企业社会责任匹配性与消费者品牌态度：条件化间接效果暨调节效应研究;袁海霞;《中国博士学位论文全文数据库 经济与管理科学辑》;20140915;J152-38 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111177571A (en) | 2020-05-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7693818B2 (en) | UserRank: ranking linked nodes leveraging user logs | |
| US6934687B1 (en) | Computer architecture and method for supporting and analyzing electronic commerce over the world wide web for commerce service providers and/or internet service providers | |
| CN112766550B (en) | Random forest-based power failure sensitive user prediction method, system, storage medium and computer equipment | |
| CN104572449A (en) | Automatic test method based on case library | |
| CN101872354A (en) | Dynamic sustainability search engine | |
| CN111177544A (en) | Operation system and method based on user behavior data and user portrait data | |
| CN110210684B (en) | Grain processing scheme optimization method, device, equipment and storage medium | |
| CN112149352B (en) | Prediction method for marketing activity clicking by combining GBDT automatic characteristic engineering | |
| CN109768869A (en) | A kind of traffic forecast method, system and computer storage medium | |
| DEAN TING et al. | Product and process cost estimation with fuzzy multi-attribute utility theory | |
| CN105740434A (en) | Network information scoring method and device | |
| US20040139035A1 (en) | System and method for integration of value-added product costs | |
| CN111177571B (en) | Value-driven dynamic recommendation system with multi-factor dimension space and multi-mesoscale fusion | |
| CN110633401A (en) | Prediction model of store data and establishment method thereof | |
| CN105787800A (en) | Intelligent social platform potential contact retrieval device, system and method | |
| CN116090702B (en) | ERP data intelligent supervision system and method based on Internet of things | |
| JP4780668B2 (en) | Traffic analysis model construction method, apparatus, construction program, and storage medium thereof | |
| CN112860899B (en) | Label generation method and device, computer equipment and computer readable storage medium | |
| CN114445134A (en) | Big data marketing method, device and medium | |
| CN105843785A (en) | Data custom calculation statement generation method embedded with organizational management level | |
| Naseri et al. | Online operational budgeting of electricity distribution network based on integration strategic and operational planning by Swot & AHP methods | |
| CN114185957B (en) | Intelligent mining method suitable for power big data service requirements | |
| Huang | An Analysis of Artificial Intelligence Engineering Cost Information Management Platform | |
| CN105574350A (en) | Event prediction method | |
| CN111597097B (en) | Big data processing method and 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |