CN104462293A - Search processing method and method and device for generating search result ranking model - Google Patents
Search processing method and method and device for generating search result ranking model Download PDFInfo
- Publication number
- CN104462293A CN104462293A CN201410706960.XA CN201410706960A CN104462293A CN 104462293 A CN104462293 A CN 104462293A CN 201410706960 A CN201410706960 A CN 201410706960A CN 104462293 A CN104462293 A CN 104462293A
- Authority
- CN
- China
- Prior art keywords
- user
- search results
- search
- click
- model
- 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
- 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
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)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides a search processing method and a method and device for generating a search result ranking model. The search processing method comprises the steps of receiving a search term of a user, acquiring a plurality of search result items according to the search term, extracting user model data comprising a plurality of user feature values of the user, acquiring the impact weights of the search result items from the search result ranking model with the search result items and the user feature values as input, ranking the search result items according to the impact weights, and sending the ranked search result items. By means of the search processing method and the method and device for generating the search result ranking model, the search result can conform to the search preference and habits of the user better, and the matching degree of the search results and the user requirement is improved.
Description
Technical field
The present invention relates to natural language processing technique field, particularly relate to the method and apparatus of a kind of search processing method, generation search results ranking model.
Background technology
Along with the development of internet, applications, search treatment technology is also day by day ripe.But existing search application is comparatively random to the process of Search Results, only considers the correlativity of Search Results and search word.
Such as, user's input " suspect's trackings " this search word is searched for, sort in the Search Results obtained first be the broadcasting link of this TV play, sort second be the encyclopaedia introduction of this TV play.
Concerning understand this TV play or had viewing history user, their demand finds to play link, but also there is the user not understanding this TV play, their demand is TV play introduction, in the Search Results using existing search treatment technology to export, sequence first be play link, and such Search Results do not meet searching preferences and the custom of this kind of user.
Summary of the invention
The object of the embodiment of the present invention is, provides the method and apparatus of a kind of search processing method, generation search results ranking model, thus makes Search Results more meet searching preferences and the custom of user, improves the matching degree of Search Results and user's request.
For achieving the above object, The embodiment provides a kind of search processing method, comprising: the search word receiving user; Multiple Search Results entry is obtained according to described search word; Extract the user model data of described user, described user model data comprise the value of multiple user characteristicses of described user; Using the value of described Search Results entry and described multiple user characteristics as input, obtain the weighing factor of each Search Results entry from search results ranking model; According to described weighing factor, described Search Results entry is sorted; Send the Search Results entry through sequence.
Embodiments of the invention additionally provide a kind of search process device, comprising: user search word receiver module, for receiving the search word of user; Search Results entry acquisition module, for obtaining multiple Search Results entry according to described search word; User model data extraction module, for extracting the user model data of described user, described user model data comprise the value of multiple user characteristicses of described user; Weighing factor acquisition module, for using the value of described Search Results entry and described multiple user characteristics as input, obtains the weighing factor of each Search Results entry from search results ranking model; Search Results entry order module, for sorting to described Search Results entry according to described weighing factor; Search Results entry sending module, for sending the Search Results entry through sequence.
Embodiments of the invention additionally provide a kind of method generating search results ranking model, comprising: set up using multiple user characteristics parameter as training characteristics to click and adjust power model; Obtain the eigenwert of multiple user characteristicses of at least one user; The training of power model is adjusted to described click, weighing factor Search Results entry sorted with the value learning each user characteristics with the eigenwert of described multiple user characteristics and one group of training sample.
Embodiments of the invention additionally provide a kind of device generating search results ranking model, comprising: click and adjust power model building module, adjust power model for setting up using multiple user characteristics parameter as training characteristics to click; Characteristic value acquisition module, for obtaining the eigenwert of multiple user characteristicses of at least one user; Weighing factor study module, for adjusting the training of power model to described click, the weighing factor sorted to Search Results entry with the value learning each user characteristics with the eigenwert of described multiple user characteristics and one group of training sample.
The method and apparatus of the search processing method that the embodiment of the present invention provides, generation search results ranking model, by combining the user model data set up in advance and the click trained tune power model data, Search Results entry is sorted to user, thus make Search Results more meet searching preferences and the custom of user, improve the matching degree of Search Results and user's request.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the method for the generation search results ranking model of the embodiment of the present invention one;
Fig. 2 is the schematic flow sheet of the search processing method of the embodiment of the present invention two;
Fig. 3 is one of application scenarios schematic diagram of the search processing method of the embodiment of the present invention two;
Fig. 4 is the application scenarios schematic diagram two of the search processing method of the embodiment of the present invention two;
Fig. 5 is the structural representation of the search process device of the embodiment of the present invention three;
Fig. 6 is the structural representation of the device of the generation search results ranking model of the embodiment of the present invention four.
Embodiment
Basic conception of the present invention is, clicks adjust power model in conjunction with multiple user characteristics and the training of Search Results labeled data; According to the Search Results entry of acquisition and multiple user characteristics values of search subscriber, power model is adjusted to obtain the weighing factor of each Search Results entry by the click of the user model set up in advance and training, according to this weighing factor, Search Results entry is sorted again, make Search Results more meet searching preferences and the custom of user, improve the matching degree of Search Results and user's request.
Be described in detail below in conjunction with the method and apparatus of accompanying drawing to a kind of search processing method of the embodiment of the present invention, generation search results ranking model.
Embodiment one
Fig. 1 is the schematic flow sheet of the method for the generation search results ranking model of the embodiment of the present invention one.The method of described generation search results ranking model comprises the steps:
Step 11: set up using multiple user characteristics parameter as training characteristics to click and adjust power model.
Step 12: the eigenwert obtaining multiple user characteristicses of at least one user.
Described user characteristics can comprise, but be not limited to, below at least one: Search Results repeats click, site preferences, user's sex, age, occupation, job intension, location, point of interest, consumption intention, consuming capacity, travel intent, health status and pregnant baby's situation.
According to exemplary embodiment of the present invention, the eigenwert of multiple user characteristicses of at least one user can be obtained from the user model database set up in advance.Foundation and the use of user model will be specifically described after a while.
Step 13: the training of power model is adjusted to click, weighing factor Search Results entry sorted with the value learning each user characteristics with the eigenwert of multiple user characteristics and one group of training sample.
Specifically, the eigenwert of multiple user characteristics and one group of training sample can be adjusted as click the input weighing model, wherein, training sample can be the training sample also marked from the search history data acquisition of at least one user.The click satisfaction data that each training sample can comprise historical search word, record and mark are clicked in user search.The value of each user characteristics adjusts to the weighing factor that Search Results entry sorts the output weighing model as click, the method based on above-mentioned machine learning generates search results ranking model.
By the method for this generation search results ranking model, search results ranking model can be trained in conjunction with the Search Results labeled data of user characteristics and user, thus the weighing factor of study can reflect searching preferences and the custom of user, contribute to the matching degree improving Search Results and user's request.
Embodiment two
Fig. 2 is the schematic flow sheet of the search processing method of the embodiment of the present invention two.Described method can be performed on such as search engine server.Described method comprises the steps:
Step 21: the search word receiving user.
Described search word can be the search word sent from client.Such as, user inputs " suspect's tracking " and searches in browser searches engine interface, and described search word is sent to search engine server by browser application.
Step 22: obtain multiple Search Results entry according to search word.
Search engine server can use search word to utilize existing search technique (such as, from web page index prepared in advance) to get multiple Search Results entry.
Step 23: the user model data extracting user, described user model data comprise the value of multiple user characteristicses of user.
Concrete, according to design of the present invention, by user access logs for each existing subscriber sets up user model.The user model data of each user comprise the value of multiple user characteristicses of user.As previously mentioned, described user characteristics can comprise, but be not limited to, below at least one: Search Results repeats click, site preferences, user's sex, age, occupation, job intension, location, point of interest, consumption intention, consuming capacity, travel intent, health status and pregnant baby's situation.
The mining process of user model data is described for " site preferences " this user characteristics.From subscriber network access daily record (such as duration 90 days), special preference whether can be had by counting user to the website that Search Results is clicked, method for digging contrasts with the click of other users to Search Results under same search word, such as same search word, the Search Results that major part user clicks belongs to X website, and the Search Results that a small amount of user clicks belongs to Y website, add up the click preference result that each user is all, obtain the preference of each user for each website.Specifically, namely excavate those " what click under a lot of search word is not popular click result; but prefer to certain specific websites some " user, such as the search word searching for movie name in a large number equally, major part user clicks the possibility of result and is encyclopaedia, likes strange skill, Sohu's high definition etc., and some user habit clicks bean cotyledon net after search movie name.
The mining process of user model data is described for " consumption intention " again.First be the process setting up consumption intention assessment model: the corelation behaviour daily record filtering out setting consumer field from the historical behavior daily record of each user, BMAT is carried out based on corelation behaviour daily record, determine corresponding buy move ahead into user behaviors log and corresponding buy after the user behaviors log of behavior, from the user behaviors log determined, select the user behaviors log meeting training data screening conditions as training sample, therefrom extract features training disaggregated model, obtain setting consumption intention assessment model corresponding to consumer field.Next is the process identifying consumption intention: the consumer field determining user to be identified, utilize the consumption intention assessment model that the consumer field determined is corresponding, the corelation behaviour daily record of user to be identified in nearly a period of time is classified, before the consumption obtaining user to be identified is intended that and buys or after buying.The aforementioned technology being intended to user characteristics digging user model data with consumption is disclosed in Chinese patent application 201310301375.7.
In addition, multiple prior art is also had to can be used for setting up described user model.
Step 24: using the value of Search Results entry and multiple user characteristics as input, obtain the weighing factor of each Search Results entry from search results ranking model.The training of described search results ranking model has been described in aforesaid embodiment one.
Step 25: the weighing factor according to obtaining sorts to Search Results entry.
After obtaining the weighing factor of each Search Results entry in step 24, just can be weighted Search Results entry according to weighing factor and to obtain score corresponding to each Search Results entry, according to score, multiple Search Results entry is sorted again, finally obtain the Search Results entry through sequence.
Step 26: send the Search Results entry through sequence.
By the search processing method of the present embodiment, the weighing factor of each Search Results entry can be obtained according to the value of Search Results entry and multiple user characteristics, according to this weighing factor, Search Results entry is sorted again, thus make Search Results more meet searching preferences and the custom of user, improve the matching degree of Search Results and user's request.
Below in conjunction with two concrete application scenarioss, further illustrate the embody rule of the embodiment of the present invention.
Fig. 3 is one of application scenarios schematic diagram of the search processing method of the embodiment of the present invention two, this is the Search Results entry through sequence sent for the user that repeatedly the acute title of searching television opens broadcasting link, embodies " Search Results repeats to click " this user characteristics to the impact of search results ranking.As shown in Figure 3, sorting in primary Search Results entry is the broadcasting link that suspect follows the trail of this TV play, because for understanding this TV play or having the user watching history, they find exactly the result demand of this search word and play link, and the Search Results entry through sequence using the search processing method of the present embodiment finally to send is concerning meeting most them.
Again such as, Fig. 4 is the application scenarios schematic diagram two of the search processing method of the embodiment of the present invention two, and this is not the user seeing novel but enter mhkc for repeatedly searching for novel name.Embody " site preferences " this user characteristics to the impact of search results ranking.As seen from Figure 4, sequence is the mhkc link of " dominating greatly " this novel in primary Search Results entry, find after to user's historical behavior data mining analysis, have certain customers to search for " dominating greatly " this word just to carry out chatting and commenting on this novel or exchange with online friend to enter " dominating greatly " mhkc, therefore for this kind of be not the user seeing that novel content is target, use the search processing method of the present embodiment the result dominating greatly mhkc can be advanceed to first, better must to meet consumers' demand.
Illustrational application scenarios is applied in the search application of PC end above, and the search processing method with sample embodiment also can be applicable in the search application of mobile terminal.
Embodiment three
Fig. 5 is the structural representation of the search process device of the embodiment of the present invention three.As shown in Figure 5, described search process device comprises:
User search word receiver module 31, for receiving the search word of user.
Search Results entry acquisition module 32, for obtaining multiple Search Results entry according to search word.
User model data extraction module 33, for extracting the user model data of user, user model data comprise the value of multiple user characteristicses of user.
Described user characteristics can comprise following at least one: Search Results repeats click, site preferences, user's sex, age, occupation, job intension, location, point of interest, consumption intention, consuming capacity, travel intent, health status and pregnant baby's situation.
Weighing factor acquisition module 34, for using the value of Search Results entry and multiple user characteristics as input, obtains the weighing factor of each Search Results entry from search results ranking model.
Search Results entry order module 35, for sorting to Search Results entry according to weighing factor.
Search Results entry sending module 36, for sending the Search Results entry through sequence.
By this search process device, the weighing factor of each Search Results entry can be obtained according to the value of Search Results entry and multiple user characteristics, according to this weighing factor, Search Results entry is sorted again, thus make Search Results more meet searching preferences and the custom of user, improve the matching degree of Search Results and user's request.
Alternatively, this search process device also comprises: click and adjust power model training module, carries out click adjust power model training for using the training sample of multiple mark to multiple user characteristics.
Described training sample can comprise historical search word, record is clicked in user search and click satisfaction.
Embodiment four
Fig. 6 is the structural representation of the device of the generation search results ranking model of the embodiment of the present invention four.As shown in Figure 6, the device of described generation search results ranking model comprises:
Clicking and adjust power model building module 41, adjusting power model for setting up using multiple user characteristics parameter as training characteristics to click.Described user characteristics can comprise following at least one: Search Results repeats click, site preferences, user's sex, age, occupation, job intension, location, point of interest, consumption intention, consuming capacity, travel intent, health status and pregnant baby's situation.
Characteristic value acquisition module 42, for obtaining the eigenwert of multiple user characteristicses of at least one user.
Preferably, characteristic value acquisition module 42 is for obtaining the eigenwert of multiple user characteristicses of at least one user described from the user model database set up in advance.
Weighing factor study module 43, for adjusting the training of power model to click, the weighing factor sorted to Search Results entry with the value learning each user characteristics with the eigenwert of multiple user characteristics and one group of training sample.The click satisfaction data that each training sample can comprise historical search word, record and mark are clicked in user search.Further, training sample is the training sample also marked from the search history data acquisition of at least one user.
By the device of this generation search results ranking model, search results ranking model can be generated, this model is utilized to sort to existing Search Results, thus make the Search Results through sequence more meet searching preferences and the custom of user, improve the matching degree of Search Results and user's request.
In several embodiment provided by the present invention, should be understood that, disclosed apparatus and method, can realize by another way.Such as, device embodiment described above is only schematic, and such as, the division of described module, is only a kind of logic function and divides, and actual can have other dividing mode when realizing.
In addition, each functional module in each embodiment of the present invention can be integrated in a processing module, also can be that the independent physics of modules exists, also can two or more module integrations in a module.Above-mentioned integrated module both can adopt the form of hardware to realize, and the form that hardware also can be adopted to add software function module realizes.
The above-mentioned integrated module realized with the form of software function module, can be stored in a computer read/write memory medium.Above-mentioned software function module is stored in a storage medium, comprising some instructions in order to make a computer equipment (can be personal computer, server, or the network equipment etc.) or processor (processor) perform the part steps of method described in each embodiment of the present invention.And aforesaid storage medium comprises: USB flash disk, portable hard drive, ROM (read-only memory) (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disc or CD etc. various can be program code stored medium.
The above; be only the specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, is anyly familiar with those skilled in the art in the technical scope that the present invention discloses; change can be expected easily or replace, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of described claim.
Claims (18)
1. a search processing method, is characterized in that, described method comprises:
Receive the search word of user;
Multiple Search Results entry is obtained according to described search word;
Extract the user model data of described user, described user model data comprise the value of multiple user characteristicses of described user;
Using the value of described Search Results entry and described multiple user characteristics as input, obtain the weighing factor of each Search Results entry from search results ranking model;
According to described weighing factor, described Search Results entry is sorted;
Send the Search Results entry through sequence.
2. method according to claim 1, it is characterized in that, described user characteristics comprise following at least one: Search Results repeats click, site preferences, user's sex, age, occupation, job intension, location, point of interest, consumption intention, consuming capacity, travel intent, health status and pregnant baby's situation.
3. method according to claim 2, is characterized in that, described method also comprises:
The training sample of multiple mark is used to carry out search results ranking model training to multiple user characteristics.
4. method according to claim 3, is characterized in that, described training sample comprises historical search word, record is clicked in user search and click satisfaction.
5. generate a method for search results ranking model, it is characterized in that, described method comprises:
Set up using multiple user characteristics parameter as training characteristics to click and adjust power model;
Obtain the eigenwert of multiple user characteristicses of at least one user;
The training of power model is adjusted to described click, weighing factor Search Results entry sorted with the value learning each user characteristics with the eigenwert of described multiple user characteristics and one group of training sample.
6. method according to claim 5, is characterized in that, the process of the eigenwert of multiple user characteristicses of described at least one user of acquisition comprises:
The eigenwert of multiple user characteristicses of at least one user described is obtained from the user model database set up in advance.
7. method according to claim 6, it is characterized in that, described user characteristics comprise following at least one: Search Results repeats click, site preferences, user's sex, age, occupation, job intension, location, point of interest, consumption intention, consuming capacity, travel intent, health status and pregnant baby's situation.
8. method according to claim 7, is characterized in that, the click satisfaction data that each described training sample comprises historical search word, record and mark are clicked in user search.
9. method according to claim 8, is characterized in that, described training sample is the training sample also marked from the search history data acquisition of at least one user described.
10. a search process device, is characterized in that, described device comprises:
User search word receiver module, for receiving the search word of user;
Search Results entry acquisition module, for obtaining multiple Search Results entry according to described search word;
User model data extraction module, for extracting the user model data of described user, described user model data comprise the value of multiple user characteristicses of described user;
Weighing factor acquisition module, for using the value of described Search Results entry and described multiple user characteristics as input, obtains the weighing factor of each Search Results entry from search results ranking model;
Search Results entry order module, for sorting to described Search Results entry according to described weighing factor;
Search Results entry sending module, for sending the Search Results entry through sequence.
11. devices according to claim 10, it is characterized in that, described user characteristics comprise following at least one: Search Results repeats click, site preferences, user's sex, age, occupation, job intension, location, point of interest, consumption intention, consuming capacity, travel intent, health status and pregnant baby's situation.
12. devices according to claim 11, is characterized in that, described device also comprises:
Click and adjust power model training module, for using the training sample of multiple mark, click is carried out to multiple user characteristics and adjust power model training.
13. devices according to claim 12, is characterized in that, described training sample comprises historical search word, record is clicked in user search and click satisfaction.
14. 1 kinds of devices generating search results ranking model, it is characterized in that, described device comprises:
Clicking and adjust power model building module, adjusting power model for setting up using multiple user characteristics parameter as training characteristics to click;
Characteristic value acquisition module, for obtaining the eigenwert of multiple user characteristicses of at least one user;
Weighing factor study module, for adjusting the training of power model to described click, the weighing factor sorted to Search Results entry with the value learning each user characteristics with the eigenwert of described multiple user characteristics and one group of training sample.
15. devices according to claim 14, is characterized in that, described characteristic value acquisition module is used for:
The eigenwert of multiple user characteristicses of at least one user described is obtained from the user model database set up in advance.
16. devices according to claim 15, it is characterized in that, described user characteristics comprise following at least one: Search Results repeats click, site preferences, user's sex, age, occupation, job intension, location, point of interest, consumption intention, consuming capacity, travel intent, health status and pregnant baby's situation.
17. devices according to claim 16, is characterized in that, the click satisfaction data that each described training sample comprises historical search word, record and mark are clicked in user search.
18. devices according to claim 17, is characterized in that, described training sample is the training sample also marked from the search history data acquisition of at least one user described.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410706960.XA CN104462293A (en) | 2014-11-27 | 2014-11-27 | Search processing method and method and device for generating search result ranking model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410706960.XA CN104462293A (en) | 2014-11-27 | 2014-11-27 | Search processing method and method and device for generating search result ranking model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104462293A true CN104462293A (en) | 2015-03-25 |
Family
ID=52908328
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410706960.XA Pending CN104462293A (en) | 2014-11-27 | 2014-11-27 | Search processing method and method and device for generating search result ranking model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104462293A (en) |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105069077A (en) * | 2015-07-31 | 2015-11-18 | 百度在线网络技术(北京)有限公司 | Search method and device |
CN105302903A (en) * | 2015-10-27 | 2016-02-03 | 广州神马移动信息科技有限公司 | Search method, apparatus and system and search result sequence adjustment basis determination method |
CN105472413A (en) * | 2015-11-19 | 2016-04-06 | 广州创维平面显示科技有限公司 | Television channel sorting method and system, and server |
CN106354856A (en) * | 2016-09-05 | 2017-01-25 | 北京百度网讯科技有限公司 | Enhanced deep neural network search method and device based on artificial intelligence |
CN106407346A (en) * | 2016-09-06 | 2017-02-15 | 百度在线网络技术(北京)有限公司 | Retrieval processing method and apparatus based on artificial intelligence |
CN106682018A (en) * | 2015-11-09 | 2017-05-17 | 百度在线网络技术(北京)有限公司 | Inquiry method and device |
WO2017084362A1 (en) * | 2015-11-18 | 2017-05-26 | 百度在线网络技术(北京)有限公司 | Model generation method, recommendation method and corresponding apparatuses, device and storage medium |
CN106777088A (en) * | 2016-12-13 | 2017-05-31 | 飞狐信息技术(天津)有限公司 | The method for sequencing search engines and system of iteratively faster |
CN107045505A (en) * | 2016-02-05 | 2017-08-15 | 百度在线网络技术(北京)有限公司 | A kind of method and apparatus for being used to inquire about |
CN107341187A (en) * | 2017-06-07 | 2017-11-10 | 努比亚技术有限公司 | Search processing method, device, equipment and computer-readable storage medium |
CN107532920A (en) * | 2015-04-15 | 2018-01-02 | 通腾导航技术股份有限公司 | The method for obtaining interest point data |
CN107832437A (en) * | 2017-11-16 | 2018-03-23 | 北京小米移动软件有限公司 | Audio/video method for pushing, device, equipment and storage medium |
CN108335137A (en) * | 2018-01-31 | 2018-07-27 | 北京三快在线科技有限公司 | Sort method and device, electronic equipment, computer-readable medium |
CN109299344A (en) * | 2018-10-26 | 2019-02-01 | Oppo广东移动通信有限公司 | The generation method of order models, the sort method of search result, device and equipment |
CN109791552A (en) * | 2016-10-28 | 2019-05-21 | 苹果公司 | It is resequenced using blended learning model to search result |
WO2020019562A1 (en) * | 2018-07-27 | 2020-01-30 | 天津字节跳动科技有限公司 | Search sorting method and device, electronic device, and storage medium |
CN110990701A (en) * | 2019-12-03 | 2020-04-10 | 掌阅科技股份有限公司 | Book searching method, computing device and computer storage medium |
CN111191133A (en) * | 2019-12-31 | 2020-05-22 | 口口相传(北京)网络技术有限公司 | Service search processing method, device and equipment |
CN111597444A (en) * | 2020-05-13 | 2020-08-28 | 北京达佳互联信息技术有限公司 | Searching method, searching device, server and storage medium |
CN112559895A (en) * | 2021-02-19 | 2021-03-26 | 深圳平安智汇企业信息管理有限公司 | Data processing method and device, electronic equipment and storage medium |
CN112818262A (en) * | 2021-01-28 | 2021-05-18 | 上海博泰悦臻网络技术服务有限公司 | Map POI searching method, system, device and medium based on user data |
CN113678147A (en) * | 2019-04-09 | 2021-11-19 | 基诺媒体株式会社 | Search method and information processing system |
CN115186163A (en) * | 2022-06-27 | 2022-10-14 | 北京百度网讯科技有限公司 | Training method and device of search result ranking model and search result ranking method and device |
WO2023138428A1 (en) * | 2022-01-19 | 2023-07-27 | 中兴通讯股份有限公司 | Search result sorting method, search system and computer-readable storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012054942A (en) * | 2011-09-20 | 2012-03-15 | Toshiba Corp | Program recommendation apparatus and program recommendation method |
CN102523511A (en) * | 2011-11-09 | 2012-06-27 | 中国传媒大学 | Network program aggregation and recommendation system and network program aggregation and recommendation method |
CN102810117A (en) * | 2012-06-29 | 2012-12-05 | 北京百度网讯科技有限公司 | Method and equipment for supplying search result |
CN103324675A (en) * | 2013-05-24 | 2013-09-25 | 崔吉平 | Internet individuation accurate information search and algorithm |
US20140025743A1 (en) * | 2012-07-18 | 2014-01-23 | Higher Challenge, Inc. | Apparatus and Method for Collaborative Content Browsing |
JP2014194723A (en) * | 2013-03-29 | 2014-10-09 | Jsol Corp | Advising system for promoting event preparation and advising method therefor |
-
2014
- 2014-11-27 CN CN201410706960.XA patent/CN104462293A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012054942A (en) * | 2011-09-20 | 2012-03-15 | Toshiba Corp | Program recommendation apparatus and program recommendation method |
CN102523511A (en) * | 2011-11-09 | 2012-06-27 | 中国传媒大学 | Network program aggregation and recommendation system and network program aggregation and recommendation method |
CN102810117A (en) * | 2012-06-29 | 2012-12-05 | 北京百度网讯科技有限公司 | Method and equipment for supplying search result |
US20140025743A1 (en) * | 2012-07-18 | 2014-01-23 | Higher Challenge, Inc. | Apparatus and Method for Collaborative Content Browsing |
JP2014194723A (en) * | 2013-03-29 | 2014-10-09 | Jsol Corp | Advising system for promoting event preparation and advising method therefor |
CN103324675A (en) * | 2013-05-24 | 2013-09-25 | 崔吉平 | Internet individuation accurate information search and algorithm |
Cited By (35)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107532920A (en) * | 2015-04-15 | 2018-01-02 | 通腾导航技术股份有限公司 | The method for obtaining interest point data |
CN105069077A (en) * | 2015-07-31 | 2015-11-18 | 百度在线网络技术(北京)有限公司 | Search method and device |
CN105302903A (en) * | 2015-10-27 | 2016-02-03 | 广州神马移动信息科技有限公司 | Search method, apparatus and system and search result sequence adjustment basis determination method |
WO2017071578A1 (en) * | 2015-10-27 | 2017-05-04 | 广州神马移动信息科技有限公司 | Searching method, apparatus and system, and method for determining search result order adjustment basis |
CN106682018A (en) * | 2015-11-09 | 2017-05-17 | 百度在线网络技术(北京)有限公司 | Inquiry method and device |
WO2017084362A1 (en) * | 2015-11-18 | 2017-05-26 | 百度在线网络技术(北京)有限公司 | Model generation method, recommendation method and corresponding apparatuses, device and storage medium |
CN105472413A (en) * | 2015-11-19 | 2016-04-06 | 广州创维平面显示科技有限公司 | Television channel sorting method and system, and server |
CN107045505B (en) * | 2016-02-05 | 2021-07-27 | 百度在线网络技术(北京)有限公司 | Method and device for query |
CN107045505A (en) * | 2016-02-05 | 2017-08-15 | 百度在线网络技术(北京)有限公司 | A kind of method and apparatus for being used to inquire about |
CN106354856A (en) * | 2016-09-05 | 2017-01-25 | 北京百度网讯科技有限公司 | Enhanced deep neural network search method and device based on artificial intelligence |
CN106407346A (en) * | 2016-09-06 | 2017-02-15 | 百度在线网络技术(北京)有限公司 | Retrieval processing method and apparatus based on artificial intelligence |
CN109791552A (en) * | 2016-10-28 | 2019-05-21 | 苹果公司 | It is resequenced using blended learning model to search result |
CN106777088A (en) * | 2016-12-13 | 2017-05-31 | 飞狐信息技术(天津)有限公司 | The method for sequencing search engines and system of iteratively faster |
CN107341187B (en) * | 2017-06-07 | 2021-08-10 | 努比亚技术有限公司 | Search processing method, device, equipment and computer storage medium |
CN107341187A (en) * | 2017-06-07 | 2017-11-10 | 努比亚技术有限公司 | Search processing method, device, equipment and computer-readable storage medium |
CN107832437A (en) * | 2017-11-16 | 2018-03-23 | 北京小米移动软件有限公司 | Audio/video method for pushing, device, equipment and storage medium |
CN107832437B (en) * | 2017-11-16 | 2021-03-02 | 北京小米移动软件有限公司 | Audio/video pushing method, device, equipment and storage medium |
WO2019148982A1 (en) * | 2018-01-31 | 2019-08-08 | 北京三快在线科技有限公司 | Sorting |
CN108335137A (en) * | 2018-01-31 | 2018-07-27 | 北京三快在线科技有限公司 | Sort method and device, electronic equipment, computer-readable medium |
WO2020019562A1 (en) * | 2018-07-27 | 2020-01-30 | 天津字节跳动科技有限公司 | Search sorting method and device, electronic device, and storage medium |
CN109299344A (en) * | 2018-10-26 | 2019-02-01 | Oppo广东移动通信有限公司 | The generation method of order models, the sort method of search result, device and equipment |
CN109299344B (en) * | 2018-10-26 | 2020-12-29 | Oppo广东移动通信有限公司 | Generation method of ranking model, and ranking method, device and equipment of search results |
CN113678147A (en) * | 2019-04-09 | 2021-11-19 | 基诺媒体株式会社 | Search method and information processing system |
CN110990701A (en) * | 2019-12-03 | 2020-04-10 | 掌阅科技股份有限公司 | Book searching method, computing device and computer storage medium |
CN110990701B (en) * | 2019-12-03 | 2022-11-15 | 掌阅科技股份有限公司 | Book searching method, computing device and computer storage medium |
CN111191133A (en) * | 2019-12-31 | 2020-05-22 | 口口相传(北京)网络技术有限公司 | Service search processing method, device and equipment |
CN111191133B (en) * | 2019-12-31 | 2023-12-01 | 口口相传(北京)网络技术有限公司 | Service search processing method, device and equipment |
CN111597444A (en) * | 2020-05-13 | 2020-08-28 | 北京达佳互联信息技术有限公司 | Searching method, searching device, server and storage medium |
CN111597444B (en) * | 2020-05-13 | 2024-03-05 | 北京达佳互联信息技术有限公司 | Searching method, searching device, server and storage medium |
CN112818262A (en) * | 2021-01-28 | 2021-05-18 | 上海博泰悦臻网络技术服务有限公司 | Map POI searching method, system, device and medium based on user data |
CN112818262B (en) * | 2021-01-28 | 2023-07-21 | 上海博泰悦臻网络技术服务有限公司 | Map POI searching method, system, equipment and medium based on user data |
CN112559895B (en) * | 2021-02-19 | 2021-05-18 | 深圳平安智汇企业信息管理有限公司 | Data processing method and device, electronic equipment and storage medium |
CN112559895A (en) * | 2021-02-19 | 2021-03-26 | 深圳平安智汇企业信息管理有限公司 | Data processing method and device, electronic equipment and storage medium |
WO2023138428A1 (en) * | 2022-01-19 | 2023-07-27 | 中兴通讯股份有限公司 | Search result sorting method, search system and computer-readable storage medium |
CN115186163A (en) * | 2022-06-27 | 2022-10-14 | 北京百度网讯科技有限公司 | Training method and device of search result ranking model and search result ranking method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104462293A (en) | Search processing method and method and device for generating search result ranking model | |
CN105335491B (en) | Behavior is clicked come to the method and system of user's Recommended Books based on user | |
CN101551806B (en) | Personalized website navigation method and system | |
CN102760138B (en) | Classification method and device for user network behaviors and search method and device for user network behaviors | |
CN105912669B (en) | Method and device for complementing search terms and establishing individual interest model | |
CN102760124B (en) | Pushing method and system for recommended data | |
CN105095187A (en) | Search intention identification method and device | |
US8209331B1 (en) | Context sensitive ranking | |
CN103886090A (en) | Content recommendation method and device based on user favorites | |
CN107862022B (en) | Culture resource recommendation system | |
CN104391999B (en) | Information recommendation method and device | |
CN104331459B (en) | A kind of network resource recommended method and device based on on-line study | |
Vu et al. | Temporal latent topic user profiles for search personalisation | |
CN102316130B (en) | A kind of behavior based on user judges the method and apparatus of the close and distant degree of itself and good friend | |
US20160210321A1 (en) | Real-time content recommendation system | |
CN105701216A (en) | Information pushing method and device | |
CN101685521A (en) | Method for showing advertisements in webpage and system | |
CN103838754B (en) | Information retrieval device and method | |
CN102073699A (en) | Method, device and equipment for improving search result based on user behaviors | |
CN104199872A (en) | Information recommendation method and device | |
CN104142999A (en) | Search result display method and device | |
CN106777282B (en) | The sort method and device of relevant search | |
CN103970754A (en) | Automatic article selection method and device | |
CN103383702A (en) | Method and system for recommending personalized news based on ranking of votes of users | |
CN104361092A (en) | Searching method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
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: 20150325 |