CN104143005B - A kind of related search system and method - Google Patents
A kind of related search system and method Download PDFInfo
- Publication number
- CN104143005B CN104143005B CN201410380639.7A CN201410380639A CN104143005B CN 104143005 B CN104143005 B CN 104143005B CN 201410380639 A CN201410380639 A CN 201410380639A CN 104143005 B CN104143005 B CN 104143005B
- Authority
- CN
- China
- Prior art keywords
- keyword
- search
- search word
- module
- correlation
- 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
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
Abstract
The invention provides a kind of related search system and method.Method comprises the following steps:A) search term is received, keyword and parameter is extracted;B) keyword and parameter are based on, candidate search word is screened;C) correlation between search term and candidate search word is calculated using neutral net language model, and obtains other feature correlations;D) weighted calculation to various features correlation is carried out, relevant search word result is obtained.System includes:Receive the receiving module of search term;Extract the keyword extracting module of keyword and keyword parameter;Store the search term database of candidate search word;The screening module of candidate search word is searched for by screening conditions of keyword parameter;Calculate the correlation calculations module of various features correlation;To the weighted calculation of various features relevance scores, the Fusion Module of relevant search word is obtained.Efficiently accurately relevant search can be realized with simple structure according to the present invention.
Description
Technical field
The present invention relates to search engine technique field, particularly a kind of related search system and method.
Background technology
With the development of internet, various services can be provided for client by Internet side.One of which is just
It is search service, that is, search engine is set in network side, when the search engine of network side receives the search that client is sent
After request, all words for matching the keyword that the searching request is carried stored in retrieval Internet side candidate data storehouse
There is provided to client for face index.In order to improve user's search experience degree using client, it is proposed that relevant search technology,
Be exactly network side search engine receive searching request after, not only retrieve Internet side candidate data storehouse in matching should
All literal indexes for the keyword that searching request is carried, the matching for also retrieving the candidate data place storage of Internet side should
The related or close literal index for the keyword that searching request is carried is there is provided to client, for searching further for for user.
However, existing relevant search technology is extracted using term frequency-inverse document frequency method to keyword, yet with
The accuracy for the keyword that this method is extracted is not high, have impact on the accuracy rate of the result of relevant search.
Accordingly, it would be desirable to which a kind of related search system and method, efficiently accurately relevant search is realized with simple structure.
The content of the invention
It is an object of the invention to provide a kind of related search system and method.
According to an aspect of the invention, there is provided a kind of method of relevant search, it is characterised in that including following step
Suddenly:A) search term is received, keyword and keyword parameter is extracted;B) keyword and the keyword parameter, screening are based on
Go out at least one candidate search word;C) calculated using neutral net language model between the search term and the candidate search word
Correlation, and obtain other feature correlations;D) weighted calculation to various features correlation is carried out, relevant search word is obtained
As a result.
Preferably, also include obtaining user equipment context information in the step a.
Preferably, the step a also includes obtaining subscriber identity information.
Preferably, the screening to the candidate search word is carried out by vertical search engine in the step b.
Preferably, the step c is calculated between current search word and candidate search word using neutral net language model
The method of correlation is as follows:C1 the keyword in the search term or the candidate search word) is extracted in the nerve net
Vector in network language model;C2 the model vector of the search term or the candidate search word) is calculated;C3 searched described in) calculating
The distance between model vector of rope word or the candidate search word.
Preferably, other described feature correlations include at least one of fraction below:It is keyword associated score, literal
Apart from fraction, searching times fraction, common appearance search fraction, physical distance fraction.
Preferably, step d comprises the following steps:
D1 the weight of various features relevance scores) is extracted;
D2) it is weighted, obtains the relevance scores of the candidate search word and the search term;
D3 fraction sequence will) be calculated;
D4 fraction highest one or more described candidate search words) are chosen as the relevant search word to return to user
Return result.
It is preferably based on the user equipment context information and various features relevance scores is configured with different weights, step
Rapid c is based on the user equipment context information, carries out the weighted calculation to various features relevance scores.
There is provided a kind of system of relevant search according to another aspect of the present invention, it is characterised in that the system includes
Receiving module, keyword extracting module, search term database, screening module, correlation calculations module, and Fusion Module, its
In, the receiving module is used to receive the search term from client, and by the search term to the keyword extracting module
Output;The keyword extracting module is used to extract keyword and keyword parameter, and by the keyword and keyword parameter
Export to the screening module;The search term database is used to store candidate search word;The screening module is used for described
In search term database, using the keyword parameter as screening conditions, candidate search word is searched out;The correlation calculations module
For calculating the correlation between the search term and the candidate search word using neutral net language model, and obtain other
Feature correlation fraction;The Fusion Module is used for the weighted calculation to various features relevance scores, obtains relevant search word
As a result.
Preferably, the user equipment context information of the receiving module also reading client, and by the user equipment feelings
Scape information is exported to the Fusion Module.
Preferably, the Fusion Module is primarily based on the user equipment context information and various features relevance scores is matched somebody with somebody
Different weights are put, then according to the weight, the weighted calculation to various features relevance scores is carried out, obtains the correlation and search
Rope word result.
According to a kind of related search system and method for the present invention, it can realize and be realized efficiently accurately with simple structure
Relevant search.
Brief description of the drawings
With reference to the accompanying drawing enclosed, the present invention more purpose, function and advantages will pass through the as follows of embodiment of the present invention
Description is illustrated, wherein:
Fig. 1 diagrammatically illustrates a kind of flow chart of related search method of the present invention.
Fig. 2 diagrammatically illustrates a kind of block diagram of related search system of the present invention.
Embodiment
By reference to one exemplary embodiment, the purpose of the present invention and function and the side for realizing these purposes and function
Method will be illustrated.However, the present invention is not limited to one exemplary embodiment as disclosed below;Can by multi-form come
It is realized.The essence of specification is only to aid in the detail of the various equivalent modifications Integrated Understanding present invention.
Hereinafter, embodiments of the invention will be described with reference to the drawings.In the accompanying drawings, identical reference represents identical
Or similar part, or same or like step.
Fig. 1 diagrammatically illustrates a kind of flow chart of related search method of the present invention.As shown in Figure 1:
Step 110, search term is received, keyword and keyword parameter is extracted.Wherein, extract keyword mode include with
Lower three kinds:Keyword is extracted by participle operation;Keyword is extracted by using knowledge base;Extracted by participle and knowledge base
Keyword.In addition, knowledge base (Knowledge Base) is structuring in knowledge engineering, easy to operate, easy utilized, comprehensively in a organized way
Knowledge cluster, be the need for being solved for a certain or some field questions, using certain or some knowledge representation modes in meter
The knowledge piece set interknited for storing, organize, managing and using in calculation machine memory.For example, search term sample is placed on
In one vertical search system, keyword and keyword parameter are carried out using past high frequency historical search word and knowledge base data
Extraction.
Keyword parameter refers to the corresponding attribute of each keyword, and the attribute includes but is not limited to category feature, title
Feature, behavioural characteristic, geographic location feature etc., the parameter that can be divided and identified for multiple attributes to keyword.
For example, the search term received is:" renting a house good inscription Tongcheng ".The keyword extracted using knowledge base is " good inscription
Tongcheng " and " renting a house ".The parameter of keyword " good inscription Tongcheng " is keyword classification (house property cell), position (geographical coordinate (longitude and latitude
Degree)) etc., the parameter such as keyword classification (house property behavior) of keyword " renting a house ".
According to another embodiment of the present invention, user equipment context information can also be obtained, user equipment context information is
User sends the facility information for the intelligent terminal that search term is utilized, and the intelligent terminal can such as smart mobile phone, flat board electricity
Brain, portable computer, palm digital assistants, intelligent wristwatch, desktop computer, digital glasses, electronic game machine and the panorama helmet
Any one in formula game machine etc..The user equipment information of acquisition include but is not limited to International Mobile Equipment Identity code (IMEI,
International Mobile Equipment Identity) and the information such as user agent (User Agent).
According to still another embodiment of the invention, for registered users, subscriber identity information (ID) can also be obtained, with
The analysis of personalized relevant search word is carried out in subsequent step.For example:User preference document is obtained from ID, user's purchase is obtained
Power level, history preference etc. are bought, so that further Optimizing Search.For example, in search term is the scene rented a house, the preference of user
To only focus on the source of houses in the residence of house type 3, then the analysis of personalized relevant search word can be further carried out accordingly.
Step 120, keyword and keyword parameter based on extraction filter out at least one candidate search word.Specifically,
Combination using above-mentioned keyword parameter is used as screening conditions, searches out at least one candidate search word.According to the present invention's
One embodiment, can carry out above-mentioned screening by vertical search engine.Screening conditions are for example:" local=Bei Jing &tag=rooms
Little Qu &tag=house properties behavior _ Zu Fang &lon=116.429741&lat=40.009287&distance=5km " is produced, specifically
Ground, " local:Beijing " represents that the keyword parameter of keyword is geographical position:Beijing;" tag=house properties cell " represents parameter
For label:House property cell;" tag=house properties behavior _ rent a house " expression parameter is label:House property behavior _ rent a house;" lon=
116.429741 " parameter longitude 116.429741 and parametric latitude are represented respectively with " lat=40.009287 ":
40.009287;" distance=5km " represents parameter distance scope:5km.Using above-mentioned all screening conditions, it can filter out
Geographical position is Beijing, and containing label " house property cell ", " house property behavior _ rent a house ", and containing (116.429741,
40.009287) search term of the longitude and latitude parameter in the range of 5km.
Step 130, calculate related between current search word and each candidate search word using neutral net language model
Property, and obtain other feature correlations.
The method for calculating the correlation between current search word and candidate search word using neutral net language model is as follows:
A) keyword in the keyword or candidate search word of current search word is extracted in neutral net language model
Vectorial (the hereinafter referred to as model vector of keyword).Wherein it is possible to participle and/or obtain current search word using knowledge base
Or the keyword in candidate search word.Furthermore, it is possible to using any one in the instruments such as word2vec, SENNA, HLBL, RNNLM
It is individual to train neutral net language model;
B) model vector of current search word or candidate search word is calculated.Specifically, for some search term, by it
In the model vector of each keyword be weighted summation, obtain the model vector of current search word or candidate search word.
C) the distance between model vector of current search word or candidate search word is calculated.Wherein, above-mentioned distance is to represent
Correlation, and above-mentioned distance can be COS distance or Euclidean distance etc..
In addition, other feature correlations are other features in addition to correlation of the current search word with candidate search word
At least one of correlation, including fraction below:Keyword associated score, it is literal apart from fraction, it is searching times fraction, common
There is search fraction, physical distance fraction.
Computational methods on other above-mentioned feature correlations are specific as follows.In the computational methods of keyword associated score
In, first participle, and by word2vec obtain term vector, then by addition of vectors, with (1) candidate search word calculate cosine away from
From;It is literal apart from fraction:Use editing distance algorithm, formula:1- (editing distance/total number of word);Physical distance fraction:Use warp
Latitude distance algorithm, formula:1-(longitude and latitude distance (rice)/preset distance (rice)) i.e. preset distances (rice)=5km=
5000m;It is common search fraction (number of times that 2 search terms were searched in same session) occur, drawn in statistical log, it is public
Formula:Occur maximum in searching times/candidate search word jointly and occur searching times jointly.
Step 140, the weighted calculation to various features relevance scores is carried out, relevant search word result is obtained.Specific bag
Include following steps:
A) weight of various features relevance scores is extracted;
B) it is weighted, obtains the relevance scores S of candidate search word and current search word, for example, S=keyword
Relevance scores * 0.15+ search term relevance scores * 0.1+ are literal common apart from fraction * 0.1+ searching times fraction * 0.15+
There is search fraction * 0.2+ physical distance fractions * 0.3;
C) fraction sequence will be calculated;
D) one or more candidate search words of fraction highest are chosen as relevant search word with to user's returning result.
According to one embodiment of present invention, various features relevance scores are configured not based on user equipment context information
Same weight, can be based on user equipment context information in step 140, carries out the weighting to various features relevance scores
Calculate.If for example, user equipment context information is certain mobile intelligent terminal (such as tablet personal computer), for the stronger spy of real-time
Levy fraction and assign larger weight, and the relatively reduced weight that less feature scores are associated with real-time.For example, in house property point
In class, when scene is mobile device, physical distance fraction can be important, and secondly other features are important has:Accessibility (is removed
Distance is outer, and whether house property is accessible to), whether real-time (can relate to intermediary/owner) now, and convenience (is matched somebody with somebody around house
Whether set is perfect) etc..
Fig. 2 diagrammatically illustrates a kind of block diagram of related search system of the present invention.As shown in Figure 2:The correlation of the present invention
Search system 200 includes receiving module 210, keyword extracting module 220, search term database 230, screening module 240, correlation
Property computing module 250, and Fusion Module 260.Each module is described in detail below:
Receiving module 210, for receiving the search word information from client, and by result to keyword extracting module
220 outputs.According to one embodiment of present invention, the user equipment context information of the also reading client of receiving module 210, and will
As a result exported to Fusion Module 260.
Keyword extracting module 220, is exported to screening module for extracting keyword and keyword parameter, and by result
240.Wherein, extracting the mode of keyword includes following three kinds:Keyword is extracted by participle operation;Carried by using knowledge base
Take keyword;Keyword is extracted by participle and knowledge base.
Search term database 230, for storing candidate search word.When aftermentioned screening module 240 carries out the sieve to search term
Search term database 230 can be accessed when selecting.
Screening module 240, in search term database 230, being joined with the keyword that keyword extracting module 220 is inputted
Number is screening conditions, searches out at least one candidate search word.According to one embodiment of present invention, vertical search can be passed through
Engine carries out above-mentioned screening.
Correlation calculations module 250, for calculating current search word and candidate search word using neutral net language model
Between correlation, and obtain other feature correlation fractions.Wherein, other feature correlations are except current search word and candidate
Other feature correlations outside the correlation of search term, including:It is keyword associated score, literal apart from fraction, searching times
Fraction, occur searching times, physical distance fraction jointly.
Fusion Module 260, for the weighted calculation to various features relevance scores, obtains relevant search word result.
According to one embodiment of present invention, Fusion Module 260 is primarily based on user equipment context information to various features
Relevance scores configure different weights, then according to the weight, carry out the weighted calculation to various features relevance scores, obtain
To relevant search word result.
According to a kind of related search system and method for the present invention, it can realize and be realized efficiently accurately with simple structure
Relevant search.
With reference to the explanation of the invention disclosed here and practice, other embodiment of the invention is for those skilled in the art
It all will be readily apparent and understand.Illustrate and embodiment is to be considered only as exemplary, of the invention true scope and purport is equal
It is defined in the claims.
Claims (8)
1. a kind of method of relevant search, it is characterised in that comprise the following steps:
A) search term is received, keyword and keyword parameter is extracted;
B) keyword and the keyword parameter are based on, the combination by the use of the keyword parameter is screened as screening conditions
Go out at least one candidate search word;
C) correlation between the search term and the candidate search word is calculated using neutral net language model, and obtains it
His feature correlation, wherein the correlation between the search term and the candidate search word is calculated by the following method:
C1) extract the keyword in the search term and the candidate search word in the neutral net language model to
Amount;
C2 the model vector of the search term and the candidate search word) is calculated;
C3 the distance between model vector of the search term and the candidate search word) is calculated;
D) based on user equipment context information various features relevance scores are configured with different weights, carries out including being based on user
The weighted calculation of the various features correlation of equipment context information, obtains relevant search word result, wherein the various features phase
The weighted calculation of closing property includes:
D1 the weight of various features relevance scores) is extracted;
D2) it is weighted, obtains the relevance scores of the candidate search word and the search term;
D3 fraction sequence will) be calculated;
D4) one or more described candidate search words of fraction highest are chosen to tie to return to user as the relevant search word
Really.
2. according to the method described in claim 1, it is characterised in that also include obtaining user equipment scene letter in the step a
Breath.
3. according to the method described in claim 1, it is characterised in that the step a also includes obtaining subscriber identity information.
4. according to the method described in claim 1, it is characterised in that carried out in the step b by vertical search engine to institute
State the screening of candidate search word.
5. according to the method described in claim 1, it is characterised in that other described feature correlations are included in fraction below extremely
Few one kind:Keyword associated score, it is literal apart from fraction, searching times fraction, common there is search fraction, physical distance point
Number.
6. a kind of system of relevant search, it is characterised in that the system includes receiving module, keyword extracting module, search
Word database, screening module, correlation calculations module, and Fusion Module, wherein,
The receiving module is used to receive the search term from client, and by the search term to the keyword extracting module
Output;
The keyword extracting module is used to extract keyword and keyword parameter, and the keyword and keyword parameter is defeated
Go out to the screening module, at least one candidate search is filtered out by the use of the combination of the keyword parameter as screening conditions
Word;
The search term database is used to store candidate search word;
The screening module is used in the search term database, using the keyword parameter as screening conditions, searches out time
Select search term;
The correlation calculations module is used to calculate the search term and the candidate search word using neutral net language model
Between correlation, and obtain other feature correlation fractions;And
The Fusion Module is used to configure various features relevance scores based on user equipment context information different weights, enters
Row obtains relevant search word result to the weighted calculation of various features relevance scores.
7. system according to claim 6, it is characterised in that the user equipment feelings of the receiving module also reading client
Scape information, and the user equipment context information is exported to the Fusion Module.
8. system according to claim 7, it is characterised in that the Fusion Module is primarily based on the user equipment scene
Information configures different weights to various features relevance scores, then according to the weight, carries out to various features correlation point
Several weighted calculations, obtains the relevant search word result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410380639.7A CN104143005B (en) | 2014-08-04 | 2014-08-04 | A kind of related search system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410380639.7A CN104143005B (en) | 2014-08-04 | 2014-08-04 | A kind of related search system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104143005A CN104143005A (en) | 2014-11-12 |
CN104143005B true CN104143005B (en) | 2017-09-12 |
Family
ID=51852179
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410380639.7A Active CN104143005B (en) | 2014-08-04 | 2014-08-04 | A kind of related search system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104143005B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI811708B (en) * | 2021-06-21 | 2023-08-11 | 信義房屋股份有限公司 | Family Member Induction Device Based on Search Behavior |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104572836A (en) * | 2014-12-10 | 2015-04-29 | 百度在线网络技术(北京)有限公司 | Method and device for confirming comprehensive relevancy of candidate inquiry sequence |
CN106326311B (en) * | 2015-07-03 | 2019-09-20 | 北京国双科技有限公司 | The method and apparatus of processing data information |
KR101694727B1 (en) * | 2015-12-28 | 2017-01-10 | 주식회사 파수닷컴 | Method and apparatus for providing note by using calculating degree of association based on artificial intelligence |
CN107292690A (en) * | 2016-03-31 | 2017-10-24 | 杨舜凯 | A kind of thing platform implementation method easy transboundary and platform based on neutral net |
CN107562761A (en) * | 2016-06-30 | 2018-01-09 | 阿里巴巴集团控股有限公司 | A kind of information-pushing method and device |
CN106776763B (en) * | 2016-11-22 | 2019-12-06 | 北京云知声信息技术有限公司 | Destination searching method and device |
CN106777282B (en) * | 2016-12-29 | 2018-07-13 | 百度在线网络技术(北京)有限公司 | The sort method and device of relevant search |
CN106933962A (en) * | 2017-02-06 | 2017-07-07 | 涂正富 | A kind of film micro area network insertion and vertical search precise positioning obtain mesh calibration method |
CN106919702B (en) * | 2017-02-14 | 2020-02-11 | 北京时间股份有限公司 | Keyword pushing method and device based on document |
CN108804443A (en) * | 2017-04-27 | 2018-11-13 | 安徽富驰信息技术有限公司 | A kind of judicial class case searching method based on multi-feature fusion |
CN107220384B (en) * | 2017-06-29 | 2019-11-15 | 北京拉勾科技有限公司 | A kind of search word treatment method based on correlation, device and calculate equipment |
WO2019084867A1 (en) * | 2017-11-02 | 2019-05-09 | 深圳前海达闼云端智能科技有限公司 | Automatic answering method and apparatus, storage medium, and electronic device |
CN112395408B (en) * | 2020-11-19 | 2023-11-07 | 平安科技(深圳)有限公司 | Stop word list generation method and device, electronic equipment and storage medium |
CN116455861B (en) * | 2023-04-17 | 2023-11-24 | 福建数智闽政科技有限公司 | Big data-based computer network security monitoring system and method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7849078B2 (en) * | 2006-06-07 | 2010-12-07 | Sap Ag | Generating searchable keywords |
CN102722498A (en) * | 2011-03-31 | 2012-10-10 | 北京百度网讯科技有限公司 | Search engine and implementation method thereof |
CN103377232A (en) * | 2012-04-25 | 2013-10-30 | 阿里巴巴集团控股有限公司 | Headline keyword recommendation method and system |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103678668A (en) * | 2013-12-24 | 2014-03-26 | 乐视网信息技术(北京)股份有限公司 | Prompting method of relevant search result, server and system |
CN103744940B (en) * | 2013-12-31 | 2017-05-31 | 北京百度网讯科技有限公司 | A kind of method and apparatus that recommended keywords are determined based on geographic position of information releaser |
-
2014
- 2014-08-04 CN CN201410380639.7A patent/CN104143005B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7849078B2 (en) * | 2006-06-07 | 2010-12-07 | Sap Ag | Generating searchable keywords |
CN102722498A (en) * | 2011-03-31 | 2012-10-10 | 北京百度网讯科技有限公司 | Search engine and implementation method thereof |
CN103377232A (en) * | 2012-04-25 | 2013-10-30 | 阿里巴巴集团控股有限公司 | Headline keyword recommendation method and system |
Non-Patent Citations (2)
Title |
---|
"Approximate Keyword Search in Web Search Engines";Sun Wu et al.;《2006 1st International Conference on Digital Information Management》;20070604;第404-411页 * |
"百度相关搜索功能的分析语应用";李佳 等;《情报搜索》;20110831(第8期);第99-101页 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI811708B (en) * | 2021-06-21 | 2023-08-11 | 信義房屋股份有限公司 | Family Member Induction Device Based on Search Behavior |
Also Published As
Publication number | Publication date |
---|---|
CN104143005A (en) | 2014-11-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104143005B (en) | A kind of related search system and method | |
CN107688823B (en) | A kind of characteristics of image acquisition methods and device, electronic equipment | |
Zheng et al. | Collaborative location and activity recommendations with gps history data | |
CN103279868B (en) | A kind of method and apparatus of automatic identification swindle order | |
CN104123332B (en) | The display methods and device of search result | |
US8315953B1 (en) | Activity-based place-of-interest database | |
CN102236663B (en) | Query method, query system and query device based on vertical search | |
CN107463704A (en) | Searching method and device based on artificial intelligence | |
CN105302810A (en) | Information search method and apparatus | |
TWI398824B (en) | A provider, a provider, a provider, and a computer-readable recording medium that memorizes its program | |
CN109582969A (en) | Methodology for Entities Matching, device and electronic equipment | |
CN101479728A (en) | Visual and multi-dimensional search | |
CN105045799A (en) | Searchable index | |
CN105205689A (en) | Method and system for recommending commercial tenant | |
CN103593425A (en) | Preference-based intelligent retrieval method and system | |
US20210133271A1 (en) | Uniform resource identifier encoding | |
CN109284498A (en) | Self-service cabinet recommendation method, self-service cabinet recommendation device and electronic device | |
CN104516980B (en) | The output method and server system of search result | |
KR20170131924A (en) | Method, apparatus and computer program for searching image | |
CN103309869A (en) | Method and system for recommending display keyword of data object | |
CN110019650A (en) | Method, apparatus, storage medium and the electronic equipment of search associational word are provided | |
CN106919588A (en) | A kind of application program search system and method | |
WO2019133206A1 (en) | Search engine for identifying analogies | |
CN111191133B (en) | Service search processing method, device and equipment | |
CN103617221B (en) | Software recommendation method and software recommendation system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |