CN109299426A - A kind of recommended method and device of accurate top information - Google Patents
A kind of recommended method and device of accurate top information Download PDFInfo
- Publication number
- CN109299426A CN109299426A CN201811464098.0A CN201811464098A CN109299426A CN 109299426 A CN109299426 A CN 109299426A CN 201811464098 A CN201811464098 A CN 201811464098A CN 109299426 A CN109299426 A CN 109299426A
- Authority
- CN
- China
- Prior art keywords
- user
- information
- recommendation
- behavior
- browsing
- 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
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses the recommended methods and device of a kind of accurate top information, user accesses network by browser, obtain user behavior, it include: the advertising information of access, browse information and search key, and interested multiple information labels of user are determined according to the user behavior, in the time threshold of setting, multiple time points monitorings user behavior is set, for each information labels, different coefficients is arranged to a variety of behaviors of user, weighting obtains the recommendation of information labels, and the top information recommendation group for being directed to the user is determined according to the interested information labels of the user, and the top information recommendation group is sent to the user client.
Description
Technical field
The present invention relates to pushing away for the big data digging technology of field of computer technology more particularly to a kind of accurate top information
Recommend method and device.
Background technique
With the rapid development of internet, network technology reaches its maturity, and cost gradually reduces, in addition to traditional broadband industry
Business, major communication company additionally provide high speed and stablize also relatively cheap mobile network, be conducive to network and be widely applied to each row
Each industry both improves social productive forces, also facilitates people's lives.Under the promotion of software industry, IT application in enterprises degree
Also higher and higher, clothing, food, lodging and transportion -- basic necessities of life and the consumer entertainment of people is covered in the various networking products provided and service, greatly facilitates
People's lives mention for user if all big enterprises of e-commerce industry release mobile applications with their own characteristics one after another
For shoppings online functions such as convenient and fast goods browse, purchase payment, logistics distribution, after-sale services.But as e-commerce is advised
The continuous expansion of mould, the value volume and range of product of commodity increase rapidly, and a large amount of information becomes long-tail and is submerged in the ocean of data.
It to be quickly found out the commodity for oneself wanting to buy from the commodity of magnanimity, is not a simple thing, user is faced with " information mistake
Carry " the problem of.
Therefore, in the epoch of the various explosion of categories of information, the selection that people face is more and more, and selection is excessive, and information is super
It carries, also can usually make one to feel at a loss.In this case, recommended engine just starts to show technical advantage, exploits one's power.
Classified catalogue and search engine are the traditional solutions for alleviating problem of information overload.According to the attribute of commodity and spy
Sign classifies, and facilitates the lookup of user.However as quickling increase for type of merchandize and quantity, commodity classification can only
Cover the commodity of a small amount of prevalence.Search engine is the traditional technology for obtaining specified requirements information.User can be in shopping platform
Keyword of upper search commercial articles, such as title, brand, the material of commodity etc..By the retrieval of keyword, user can be inquired
Oneself desired each brand specifies the merchandise news of material, filters out quality-high and inexpensive by comparison while also complying with oneself demand
Commodity.But this needs the demand that specific keyword embodies user, the result of search is also to be limited in known to user
Within range of information, can not retrieve user may interested unknown merchandise news.It would therefore be desirable to have a systems will appreciate that
The interest preference and consumption habit of user, active recommended user is interested or the commodity that may need.Recommender system is met the tendency of
And give birth to, it does not need user and keyword is provided, the historical behavior data by analyzing user find the individual demand of user, actively
Recommend the commodity for being suitble to user, the commodity including being in long-tail solve the problems, such as information overload.
Recommender system passes through the various actions such as analysis user browsing, purchase, scoring using reasonable effective mining algorithm
Mass data excavates the consumption propensity of user, and recommended user has the commodity of potential purchasing demand, realizes precision marketing.For pushing away
Systematic research is recommended, either has positive meaning to the development or social progress of science and technology.
Recommender system is to solve the relatively effective method of information overload problem, it is according to the historical data analysis of user behavior
Interested commodity of user etc. are recommended user by the hobby etc. of user.Resnick and Varian was such in 1997
Define recommender system: " it is to provide merchandise news and suggestion to client using e-commerce website, and user is helped to determine purchase
Any product is bought, pseudo sale personnel help client to complete purchasing process ".The recommender system of e-commerce mainly includes candidate quotient
Product, recommended method and target user.Suitable commercial product recommending is filtered out to mesh from the candidate commodity of magnanimity using proposed algorithm
Mark user.
Summary of the invention
In view of this, the main purpose of the present invention is to provide one kind can accurately analyze the content that user wants browsing
Top information recommendation method, avoid user from being at a loss when in face of massive internet content.
The present invention is directed at least solve one of the technical problems existing in the prior art.For this purpose, disclosing a kind of accurate head
The recommended method of information, user pass through browser and access network, obtain user behavior, comprising: advertising information, the browsing of access
Information and search key, and determine according to the user behavior interested multiple information labels of user, setting when
Between in threshold value, being arranged multiple time points monitors the user behavior each information labels sets a variety of behaviors of user
Different coefficients is set, weighting obtains the recommendation of information labels, and determines needle according to the interested information labels of the user
The user client is sent to the top information recommendation group of the user, and by the top information recommendation group.
Preferably, different coefficients is arranged in a variety of behaviors to user, and the coefficient meets normal distribution.
Preferably, user is grouped according to the similarity of user behavior, determination is having the same with the user in group
Interest counts advertising information, browsing information and the search key of all users access in same group, accesses user
Information carries out interest-degree ranking, and the scheme of top score is selected to recommend user.
Preferably, it scores at least a set of top information recommendation scheme generated, the scoring is closed according to user
The time, time for hovering in top news of mouse, user for closing top news are browsing hits the time of information, whether user delivers comments
By, whether input in other search engines recommendation top information keyword, above-mentioned user's operation behavior is weighted
The accuracy for calculating top recommendation information, judges whether user is interested in the top information of recommendation, if loseing interest in, again
Recommend top information, until user is satisfied.
Preferably, the information in the top information recommendation group be by big data parallel computation according to cosine similarity into
Row collects, wherein cosine similarity, measures the similarity of two vectors, cosine value value model by calculating the cosine value of angle
Enclosing is [- 1,1], and angle is smaller, and cosine value tends to consistent closer to the direction of 1, two vector, and similarity is also higher, meter
Calculation method is as shown by the equation；
Sim (i, j)=cos (i, j)=u ∈ URu, i × Ru, ju ∈ UR2u, iu ∈ UR2u, j
Wherein sim (i, j) indicates the similarity of article i and article j, and Ru indicates scoring of the user u to article i, Ru, j table
Show scoring of the user u to article j, the scoring is based on based on user behavior.
Preferably, user's history behavioural characteristic is analyzed, user's history behavioural characteristic library is established, by calculating user's access
Advertising information, browsing information and search key, excavate the letter with user's browsing content most correlation with FP-Tree algorithm
Breath, obtains the same type information of above-mentioned user's browsing content most information of correlation by data mining.
The invention also discloses a kind of recommendation apparatus of accurate top information characterized by comprising user behavior is collected
Module, for obtaining user behavior, comprising: advertising information, browsing information and the search key of access；Data analysis module,
User demand is analyzed in conjunction with subscriber data for the action trail data according to user；Data-mining module, according to the use
Family behavior determines interested multiple information labels of user；Data match module, for being arranged in the time threshold of setting
Multiple time points monitor the user behavior and different coefficients are arranged to a variety of behaviors of user for each information labels,
Weighting obtains the recommendation of information labels, and the top news for being directed to the user is determined according to the interested information labels of the user
Information recommendation group.
Preferably, different coefficients is arranged in a variety of behaviors to user, and the coefficient meets normal distribution.
Preferably, score recommending module, for scoring at least a set of top information recommendation scheme generated, institute
Commentary point closes the time of top news according to user, the time that mouse hovers in top news, user be browsing hit information time,
Whether user make comments, whether inputted in other search engines recommendation top information keyword, above-mentioned user is grasped
The accuracy of top recommendation information is weighted as behavior, judges whether user is interested in the top information of recommendation, if
Lose interest in, then recommend top information again, until user is satisfied.
Preferably, user behavior memory module: for analyzing user's history behavioural characteristic, user's history behavioural characteristic is established
It is excavated with FP-Tree algorithm and user by calculating advertising information, the browsing information and search key of user's access in library
The information of browsing content most correlation obtains the same of above-mentioned user's browsing content most information of correlation by data mining
Type information.
Detailed description of the invention
From following description with reference to the accompanying drawings it will be further appreciated that the present invention.Component in figure is not drawn necessarily to scale,
But it focuses on and shows in the principle of embodiment.In the figure in different views, identical appended drawing reference is specified to be corresponded to
Part.
Fig. 1 is a kind of flow chart of the recommended method of accurate top information of the invention.
Fig. 2 is a kind of recommended method structure chart of accurate top information of the invention.
Specific embodiment
Embodiment one
System construction drawing of the invention shown in Fig. 2, a kind of recommendation side of specific accurate top information as shown in Figure 1
Method, user access network by browser, obtain user behavior, comprising: advertising information, browsing information and the search key of access
Word, and determine that interested multiple information labels of user are arranged more in the time threshold of setting according to the user behavior
A time point monitors the user behavior and each information labels is arranged different coefficients to a variety of behaviors of user, added
Power obtains the recommendation of information labels, and determines that the top news for the user is believed according to the interested information labels of the user
Recommendation group is ceased, and the top information recommendation group is sent to the user client.
Further, different coefficients is arranged in a variety of behaviors to user, and the coefficient meets normal distribution.
Further, user is grouped according to the similarity of user behavior, determining has phase with the user in group
Same interest counts advertising information, browsing information and the search key of all users access in same group, to user
Access information carries out interest-degree ranking, and the scheme of top score is selected to recommend user.
Further, score at least a set of top information recommendation scheme generated, the scoring according to
The time of top news is closed at family, time for hovering in top news of mouse, user are that browsing hits the time of information, whether user delivers
Comment, whether inputted in other search engines recommendation top information keyword, above-mentioned user's operation behavior is added
Power calculates the accuracy of top recommendation information, judges whether user is interested in the top information of recommendation, if loseing interest in, weighs
It is new to recommend top information, until user is satisfied.
Further, the information in the top information recommendation group is similar according to cosine by the parallel computation of big data
Degree is collected, wherein cosine similarity, measures the similarity of two vectors by calculating the cosine value of angle, cosine value takes
Value range is [- 1,1], and angle is smaller, and cosine value tends to consistent closer to the direction of 1, two vector, and similarity is also higher,
Its calculation method is as shown by the equation；
Sim (i, j)=cos (i, j)=u ∈ URu, i × Ru, ju ∈ UR2u, iu ∈ UR2u, j
Wherein sim (i, j) indicates the similarity of article i and article j, and Ru indicates scoring of the user u to article i, Ru, j table
Show scoring of the user u to article j, the scoring is based on based on user behavior.
Further, user's history behavioural characteristic is analyzed, user's history behavioural characteristic library is established, is visited by calculating user
Advertising information, browsing information and the search key asked, are excavated and user's browsing content most correlation with FP-Tree algorithm
Information, the same type information of above-mentioned user's browsing content most information of correlation is obtained by data mining.
It should be evident that in this embodiment, recommending the method for top information using two kinds, by the row for analyzing user
For data, the interested content of user's browsing content is grouped using while being grouped to user, by big data
Technology realizes the recommendation function of top information.
Embodiment two
Such as during user's shopping on the web, what is faced is dazzling commodity, from different suppliers and not
With brand, the material of product and workmanship be also it is very different, how to select to need to take much time to be browsed and compared,
It consumes more energy and removes history evaluation record of analysis commodity etc., embodiment two discloses one kind to solve the above-mentioned problems
The recommended method of accurate top news information, user access network by browser, obtain user behavior, comprising: the advertisement of access is believed
Breath browses information and search key, and interested multiple information labels of user are determined according to the user behavior, is setting
In fixed time threshold, multiple time points monitorings user behavior is set, for each information labels, to a variety of of user
Different coefficients is arranged in behavior, and weighting obtains the recommendation of information labels, and according to the interested information labels of the user
It determines the top information recommendation group for being directed to the user, and the top information recommendation group is sent to the user client.
Preferably, different coefficients is arranged in a variety of behaviors to user, and the coefficient meets normal distribution.
Preferably, user is grouped according to the similarity of user behavior, determination is having the same with the user in group
Interest counts advertising information, browsing information and the search key of all users access in same group, accesses user
Information carries out interest-degree ranking, and the scheme of top score is selected to recommend user.
Preferably, it scores at least a set of top information recommendation scheme generated, the scoring is closed according to user
The time, time for hovering in top news of mouse, user for closing top news are browsing hits the time of information, whether user delivers comments
By, whether input in other search engines recommendation top information keyword, above-mentioned user's operation behavior is weighted
The accuracy for calculating top recommendation information, judges whether user is interested in the top information of recommendation, if loseing interest in, again
Recommend top information, until user is satisfied.
Preferably, the information in the top information recommendation group be by big data parallel computation according to cosine similarity into
Row collects, wherein cosine similarity, measures the similarity of two vectors, cosine value value model by calculating the cosine value of angle
Enclosing is [- 1,1], and angle is smaller, and cosine value tends to consistent closer to the direction of 1, two vector, and similarity is also higher, meter
Calculation method is as shown by the equation；
Sim (i, j)=cos (i, j)=u ∈ URu, i × Ru, ju ∈ UR2u, iu ∈ UR2u, j
Wherein sim (i, j) indicates the similarity of article i and article j, and Ru indicates scoring of the user u to article i, Ru, j table
Show scoring of the user u to article j, the scoring is based on based on user behavior.
Preferably, user's history behavioural characteristic is analyzed, user's history behavioural characteristic library is established, by calculating user's access
Advertising information, browsing information and search key, excavate the letter with user's browsing content most correlation with FP-Tree algorithm
Breath, obtains the same type information of above-mentioned user's browsing content most information of correlation by data mining.
The invention also discloses a kind of recommendation apparatus of accurate top information characterized by comprising user behavior is collected
Module, for obtaining user behavior, comprising: advertising information, browsing information and the search key of access；Data analysis module,
User demand is analyzed in conjunction with subscriber data for the action trail data according to user；Data-mining module, according to the use
Family behavior determines interested multiple information labels of user；Data match module, for being arranged in the time threshold of setting
Multiple time points monitor the user behavior and different coefficients are arranged to a variety of behaviors of user for each information labels,
Weighting obtains the recommendation of information labels, and the top news for being directed to the user is determined according to the interested information labels of the user
Information recommendation group.
Preferably, different coefficients is arranged in a variety of behaviors to user, and the coefficient meets normal distribution.
Preferably, score recommending module, for scoring at least a set of top information recommendation scheme generated, institute
Commentary point closes the time of top news according to user, the time that mouse hovers in top news, user be browsing hit information time,
Whether user make comments, whether inputted in other search engines recommendation top information keyword, above-mentioned user is grasped
The accuracy of top recommendation information is weighted as behavior, judges whether user is interested in the top information of recommendation, if
Lose interest in, then recommend top information again, until user is satisfied.
Preferably, user behavior memory module: for analyzing user's history behavioural characteristic, user's history behavioural characteristic is established
It is excavated with FP-Tree algorithm and user by calculating advertising information, the browsing information and search key of user's access in library
The information of browsing content most correlation obtains the same of above-mentioned user's browsing content most information of correlation by data mining
Type information.
In example 2, recommendation apparatus is a multi-source information device, is connect with multiple information, shopping platform, is used
User behavior collection and memory module or log acquisition module are using log acquisition module to user behavior analysis
When, the behavior that user occurs on a web browser will be captured by log acquisition module and pass to user behavior memory module.Recommend
System excavates the browsing preference of user, the top information recommendation inventory of generation according to the user behavior in user behavior storage system
It will be presented to the user by page presentation user client.
Although describing the present invention by reference to various embodiments above, but it is to be understood that of the invention not departing from
In the case where range, many changes and modifications can be carried out.Therefore, be intended to foregoing detailed description be considered as it is illustrative and
It is unrestricted, and it is to be understood that following following claims (including all equivalents) is intended to limit spirit and model of the invention
It encloses.The above embodiment is interpreted as being merely to illustrate the present invention rather than limit the scope of the invention.It is reading
After the content of record of the invention, technical staff can be made various changes or modifications the present invention, these equivalence changes and
Modification equally falls into the scope of the claims in the present invention.
Claims (10)
1. a kind of recommended method of accurate top information, user access network by browser, obtain user behavior, comprising: visit
The advertising information asked, browsing information and search key, and determine according to the user behavior interested multiple letters of user
Label is ceased, in the time threshold of setting, being arranged multiple time points monitors the user behavior, for each information labels,
Different coefficients is arranged to a variety of behaviors of user, weighting obtains the recommendation of information labels, and emerging according to the sense of the user
The information labels of interest determine the top information recommendation group for being directed to the user, and the top information recommendation group is sent to the user
Client.
2. a kind of recommended method of accurate top information as described in claim 1, further comprises: described to a variety of of user
Different coefficients is arranged in behavior, and the coefficient meets normal distribution.
3. a kind of recommended method of accurate top information as claimed in claim 2, further comprises: according to the phase of user behavior
User is grouped like degree, is determined with user's interest having the same in group, to the wide of all users access in same group
It accuses information, browsing information and search key to be counted, interest-degree ranking is carried out to user access information, selects top score
Scheme recommend user.
4. a kind of recommended method of accurate top information as claimed in claim 3, further comprises: to generated at least one
It covers top information recommendation scheme to score, the scoring closes the time of top news according to user, mouse hovers in top news
Whether time, user are that browsing hits the time of information, whether user makes comments, input and recommend in other search engines
The accuracy of top recommendation information is weighted to above-mentioned user's operation behavior, judges user for the keyword of top information
It is whether interested in the top information of recommendation, if loseing interest in, recommend top information again, until user is satisfied.
5. a kind of recommended method of accurate top information as claimed in claim 4, further comprises: the top news information recommendation
Information in group is to be collected by the parallel computation of big data according to cosine similarity, wherein cosine similarity, passes through calculating
The cosine value of angle measures the similarities of two vectors, and cosine value value range is [- 1,1], and angle is smaller, and cosine value more connects
It is bordering on 1, the direction of two vectors tends to consistent, and similarity is also higher, and calculation method is as shown by the equation；
Sim (i, j)=cos (i, j)=u ∈ URu, i × Ru, ju ∈ UR2u, iu ∈ UR2u, j
Wherein sim (i, j) indicates the similarity of article i and article j, and Ru indicates scoring of the user u to article i, and Ru, j indicate to use
Scoring of the family u to article j, the scoring is based on based on user behavior.
6. a kind of recommended method of accurate top information as claimed in claim 5, further comprises: analysis user's history behavior
Feature establishes user's history behavioural characteristic library, by calculating the advertising information, browsing information and search key of user's access,
The information with user's browsing content most correlation is excavated with FP-Tree algorithm, it is clear to obtain above-mentioned user by data mining
Look at content most correlation information same type information.
7. a kind of recommendation apparatus of accurate top information characterized by comprising user behavior collection module is used for obtaining
Family behavior, comprising: advertising information, browsing information and the search key of access；Data analysis module, for the row according to user
User demand is analyzed in conjunction with subscriber data for track data；Data-mining module determines user's according to the user behavior
Interested multiple information labels；Data match module, for multiple time point monitorings institute in the time threshold of setting, to be arranged
User behavior is stated, for each information labels, different coefficients is arranged to a variety of behaviors of user, weighting obtains information labels
Recommendation, and according to the interested information labels of the user determine be directed to the user top information recommendation group.
8. a kind of recommendation apparatus of accurate top information as claimed in claim 7, further comprises: described to a variety of of user
Different coefficients is arranged in behavior, and the coefficient meets normal distribution.
9. a kind of recommendation apparatus of accurate top information as claimed in claim 8, further comprise: scoring recommending module is used
Score in at least a set of top information recommendation scheme generated, the scoring closed according to user top news time,
Whether time that mouse hovers in top news, user are that browsing hits the time of information, whether user makes comments, search at other
Index holds up the keyword for the top information that middle input is recommended, and top recommendation information is weighted to above-mentioned user's operation behavior
Accuracy, judge whether user interested in the top information of recommendation, if loseing interest in, recommend top information again, directly
Until user is satisfied.
10. a kind of recommendation apparatus of accurate top information as claimed in claim 9, further comprise: user behavior stores mould
Block: for analyzing user's history behavioural characteristic, user's history behavioural characteristic library is established, the advertisement by calculating user's access is believed
Breath, browsing information and search key, excavate the information with user's browsing content most correlation with FP-Tree algorithm, lead to
It crosses data mining and obtains the same type information of above-mentioned user's browsing content most information of correlation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811464098.0A CN109299426A (en) | 2018-11-30 | 2018-11-30 | A kind of recommended method and device of accurate top information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811464098.0A CN109299426A (en) | 2018-11-30 | 2018-11-30 | A kind of recommended method and device of accurate top information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109299426A true CN109299426A (en) | 2019-02-01 |
Family
ID=65142349
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811464098.0A Pending CN109299426A (en) | 2018-11-30 | 2018-11-30 | A kind of recommended method and device of accurate top information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109299426A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110225374A (en) * | 2019-06-13 | 2019-09-10 | 北京连屏科技有限公司 | The user information acquiring and processing method of Interactive Internet TV |
| CN110297966A (en) * | 2019-04-24 | 2019-10-01 | 上海易点时空网络有限公司 | Content recommendation method and device for community's class application program |
| CN112764855A (en) * | 2021-01-20 | 2021-05-07 | 恒瑞通（福建）信息技术有限公司 | Block chain based government affair information automatic display method, equipment and medium |
| CN113360761A (en) * | 2021-06-11 | 2021-09-07 | 北京百度网讯科技有限公司 | Information flow recommendation method and device, electronic equipment and computer-readable storage medium |
| CN113836423A (en) * | 2021-09-29 | 2021-12-24 | 上海陆道动美科技有限公司 | Information recommendation method based on user use behaviors |
| CN114612126A (en) * | 2021-12-08 | 2022-06-10 | 江苏众亿国链大数据科技有限公司 | Information pushing method based on big data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108205682A (en) * | 2016-12-19 | 2018-06-26 | 同济大学 | It is a kind of for the fusion content of personalized recommendation and the collaborative filtering method of behavior |
| CN108615177A (en) * | 2018-04-09 | 2018-10-02 | 武汉理工大学 | Electric terminal personalized recommendation method based on weighting extraction interest-degree |
| CN108647996A (en) * | 2018-04-11 | 2018-10-12 | 中山大学 | A kind of personalized recommendation method and system based on Spark |
-
2018
- 2018-11-30 CN CN201811464098.0A patent/CN109299426A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108205682A (en) * | 2016-12-19 | 2018-06-26 | 同济大学 | It is a kind of for the fusion content of personalized recommendation and the collaborative filtering method of behavior |
| CN108615177A (en) * | 2018-04-09 | 2018-10-02 | 武汉理工大学 | Electric terminal personalized recommendation method based on weighting extraction interest-degree |
| CN108647996A (en) * | 2018-04-11 | 2018-10-12 | 中山大学 | A kind of personalized recommendation method and system based on Spark |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110297966A (en) * | 2019-04-24 | 2019-10-01 | 上海易点时空网络有限公司 | Content recommendation method and device for community's class application program |
| CN110225374A (en) * | 2019-06-13 | 2019-09-10 | 北京连屏科技有限公司 | The user information acquiring and processing method of Interactive Internet TV |
| CN110225374B (en) * | 2019-06-13 | 2021-08-06 | 贵州广电新媒体产业发展有限公司 | User information collecting and processing method for interactive network television |
| CN112764855A (en) * | 2021-01-20 | 2021-05-07 | 恒瑞通（福建）信息技术有限公司 | Block chain based government affair information automatic display method, equipment and medium |
| CN113360761A (en) * | 2021-06-11 | 2021-09-07 | 北京百度网讯科技有限公司 | Information flow recommendation method and device, electronic equipment and computer-readable storage medium |
| CN113836423A (en) * | 2021-09-29 | 2021-12-24 | 上海陆道动美科技有限公司 | Information recommendation method based on user use behaviors |
| CN114612126A (en) * | 2021-12-08 | 2022-06-10 | 江苏众亿国链大数据科技有限公司 | Information pushing method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109299426A (en) | A kind of recommended method and device of accurate top information | |
| CN105224699B (en) | News recommendation method and device | |
| Lo et al. | Understanding behaviors that lead to purchasing: A case study of pinterest | |
| Wen et al. | A hybrid approach for personalized recommendation of news on the Web | |
| TWI636416B (en) | Method and system for multi-phase ranking for content personalization | |
| CN103544216B (en) | The information recommendation method and system of a kind of combination picture material and keyword | |
| Xiao et al. | Time-ordered collaborative filtering for news recommendation | |
| Zhang et al. | Personalised online sales using web usage data mining | |
| US20070214133A1 (en) | Methods for filtering data and filling in missing data using nonlinear inference | |
| CN105426528A (en) | Retrieving and ordering method and system for commodity data | |
| CN107256512A (en) | One kind house-purchase personalized recommendation method and system | |
| EP2188712A2 (en) | Recommendation systems and methods | |
| CN102073717A (en) | Home page recommending method for orienting vertical e-commerce website | |
| Mishra et al. | Research problems in recommender systems | |
| Ma et al. | Measuring the coverage and redundancy of information search services on e-commerce platforms | |
| Yin et al. | Exploring social activeness and dynamic interest in community-based recommender system | |
| CN111310046A (en) | Object recommendation method and device | |
| Wang | A personalized collaborative recommendation approach based on clustering of customers | |
| Zhao et al. | Personalized recommendation by exploring social users’ behaviors | |
| CN110827044A (en) | Method and device for extracting user interest mode | |
| Zhou et al. | Model and Implementation of E-commerce Recommendation System Based on User Clustering | |
| Trappey et al. | Using Web Mining and Perceptual Mapping to Support Customer-Oriented Product Positions Designs | |
| WO2008032037A1 (en) | Method and system for filtering and searching data using word frequencies | |
| Zeng et al. | Location-aware real-time recommender systems for brick-and-mortar retailers | |
| Hu et al. | Utilizing users' tipping points in E-commerce Recommender systems |
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: 20190201 |