CN109948014A - A kind of searching method and server - Google Patents
A kind of searching method and server Download PDFInfo
- Publication number
- CN109948014A CN109948014A CN201710735984.1A CN201710735984A CN109948014A CN 109948014 A CN109948014 A CN 109948014A CN 201710735984 A CN201710735984 A CN 201710735984A CN 109948014 A CN109948014 A CN 109948014A
- Authority
- CN
- China
- Prior art keywords
- search
- search term
- user
- rewriting
- term
- 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
- 238000000034 method Methods 0.000 title claims abstract description 48
- 230000006399 behavior Effects 0.000 claims description 22
- 230000015654 memory Effects 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 14
- 238000003860 storage Methods 0.000 claims description 10
- 230000003542 behavioural effect Effects 0.000 claims description 8
- 230000011218 segmentation Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 2
- 239000000047 product Substances 0.000 description 34
- 230000006870 function Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 8
- 206010049244 Ankyloglossia congenital Diseases 0.000 description 7
- 235000013399 edible fruits Nutrition 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 5
- 238000013519 translation Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000011017 operating method Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000005520 cutting process Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 101100481876 Danio rerio pbk gene Proteins 0.000 description 1
- 101100481878 Mus musculus Pbk gene Proteins 0.000 description 1
- 239000007795 chemical reaction product Substances 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 210000003127 knee Anatomy 0.000 description 1
- 230000005389 magnetism Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 229920001296 polysiloxane Polymers 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Abstract
This application provides a kind of searching method and servers, wherein this method comprises: obtaining user's search term;Determine rewriting search term similar with user's search term, wherein rewriting search term is the search term that similarity corresponding with the user's search term click behavior that behavior determines is clicked based on the historical search of multiple users;It scans for matching based on user's search term, obtains the first search matching result;It scans for matching based on search term is rewritten, obtains the second search matching result;First search matching result and the second search matching result are returned together as the search result based on user's search term.It is solved by the scheme of the application in the presence of existing searching method because of the technical issues of difference of semantic statement causes some some products for meeting user search intent that can not be called back, reach the recall rate for effectively improving product, improves the technical effect of search efficiency and accuracy.
Description
Technical field
The application belongs to Internet technical field more particularly to a kind of searching method and server.
Background technique
In e-commerce field, search is always the topic that people extremely pay close attention to.For electric business platform, it is desirable to user
It inputs after a search term (Query), available many search results are selected for user, in order to which user can select
To oneself interested thing, to improve the flow and probability of transaction of platform.
Existing search matching way is usually that the search term inputted to user segments, and is then matched, only
There is the segment to fit like a glove with the participle segment obtained after user's search term participle in the product title of product, these products are
It may show as the search result of user's search term.
However, the description language existed between some user's search terms and product title sometimes has differences, such as:
Query:nike sport footwear, product title: nike running shoe.In the case where user's search term is nike sport footwear, product mark
The product that nike sport footwear is still not present there are nike running shoe in topic would not be shown as matched search result.
It will lead to some products for meeting user search intent in this way, since cause will not conduct for the difference of description language
Search result is demonstrated.
In view of the above-mentioned problems, currently no effective solution has been proposed.
Summary of the invention
The application is designed to provide a kind of searching method and server, can more to meet user search intent
Product be called back as search result.
The application provides a kind of searching method and server is achieved in that
A kind of searching method, which comprises
Obtain user's search term;
Determine rewriting search term similar with user's search term, wherein the rewriting search term is based on multiple use
The search term that the similarity corresponding with user's search term that behavior determines clicks behavior is clicked in the historical search at family;
It scans for matching based on user's search term, obtains the first search matching result;
It scans for matching based on the rewriting search term, obtains the second search matching result;
The first search matching result and the second search matching result are searched for together as based on the user
The search result of word is returned.
A kind of server, including processor and for the memory of storage processor executable instruction, the processor
Execute following steps:
Obtain user's search term;
Determine rewriting search term similar with user's search term, wherein the rewriting search term is based on multiple use
The search term that the similarity corresponding with user's search term that behavior determines clicks behavior is clicked in the historical search at family;
It scans for matching based on user's search term, obtains the first search matching result;
It scans for matching based on the rewriting search term, obtains the second search matching result;
The first search matching result and the second search matching result are searched for together as based on the user
The search result of word is returned.
A kind of computer readable storage medium is stored thereon with computer instruction, and it is above-mentioned that described instruction is performed realization
The step of method.
Searching method and server provided by the present application are not directly to be searched for user after getting user's search term
Word scans for result and recalls, but determines rewriting search term similar with user's search term, then will be searched for based on user
The search result of word and based on rewrite search term search result be returned together as search result, to solve existing
It is in the presence of searching method because the difference of semantic statement causes some some products for meeting user search intent can not be by
The technical issues of recalling has reached the recall rate for effectively improving product, improves search efficiency, so that more meeting user's search
The product of intention is called back as search result, effectively increases the search experience of user.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The some embodiments recorded in application, for those of ordinary skill in the art, in the premise of not making the creative labor property
Under, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of method flow diagram of embodiment of searching method provided by the present application;
Fig. 2 is the schematic diagram provided by the present application for determining similarity word;
Fig. 3 is that the similarity word provided by the present application based on model training determines flow chart;
Fig. 4 is existing searching method flow chart;
Fig. 5 is searching method flow diagram provided by the present application;
Fig. 6 is a kind of modular structure schematic diagram of embodiment of server provided by the present application;
Fig. 7 is a kind of modular structure schematic diagram of embodiment of searcher provided by the present application.
Specific embodiment
In order to make those skilled in the art better understand the technical solutions in the application, below in conjunction with the application reality
The attached drawing in example is applied, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described implementation
Example is merely a part but not all of the embodiments of the present application.Based on the embodiment in the application, this field is common
The application protection all should belong in technical staff's every other embodiment obtained without creative efforts
Range.
In order to enable search result can be matched as much as possible, it is existing mainly to have when matching search result
Following methods:
1) the search matching based on synonym
I.e., it is possible to a synonymicon be pre-established, after getting user's search term, from the synonymicon
Match the synonym of user's search term.Then it is obtained by the search result obtained based on user's search term and based on synonym
Search result is returned together as search result.
2) mode of translation study
That is, carrying out cutting and translation to user's search term, the candidate result of multiple cuttings and translation is obtained, machine is then passed through
The a little mode of device selects optimal search term to scan for as synonymic search word from multiple cuttings and translation result.
However, above-mentioned first way needs to pre-establish a synonymicon, and what is included in dictionary is generally exactly
Near synonym in some normal dictionaries, can not reflect whether similar from the angle of search, and be only approximation semantically.Second
Mode needs to carry out document translation, and accuracy is not high.
For altered in existing expanded search, in this example, it is contemplated that if user is inputting different user
After search term, there is similar click behavior, then the two user's search terms may be considered similar search term each other, such as
Fruit can rewrite user's search term by similar search term, then can be to avoid leading to some because of semantic difference
The product for appearing in search result can not appear in search result.
For example, searching request is " American-European sandals are flat ", then it is considered that if it is desired to will be with searching request kissing
The product of conjunction is all used as search result to recommend, and will lead to according to existing matching way of search and there is " Rome in product title
The product that " American-European sandals are flat " may be not present in shoes angle " is not present in search result.But in actual search behavior,
" American-European sandals are flat " and " calceus angle " the two search terms, it is similar that corresponding user, which clicks behavior,.
If for this purpose, in view of can be by the similar search term also conduct of " calceus angle " conduct " American-European sandals are flat "
Search term scans for matching, thus will be based on the search result that " calceus angle " obtains, and it is based on " American-European sandals are flat "
Obtained search result, all as input user's search term " American-European sandals are flat " when the search result that generates show, that
It can be searched to avoid above-mentioned because limitation present in matching search result causes some satisfactory products that can not be used as
Hitch fruit is shown.
Based on above-mentioned purpose, a kind of searching method is provided in this example, and Fig. 1 is a kind of herein described searching method one
The method flow diagram of a embodiment.Although this application provides as the following examples or method operating procedure shown in the drawings or dress
Structure is set, but based on routine or may include more or less in the method or device without creative labor
Operating procedure or modular unit.In the step of there is no necessary causalities in logicality or structure, the execution of these steps
Sequence or the modular structure of device are not limited to the embodiment of the present application description and execution shown in the drawings sequence or modular structure.It is described
Method or modular structure device in practice or end product in application, can be according to embodiment or shown in the drawings
Method or modular structure connection carry out sequence execution or parallel execution (such as the ring of parallel processor or multiple threads
Border or even distributed processing environment).
As shown in Figure 1, may include steps of:
Step 101: obtaining user's search term;
User's search term can be user and pass through the smart machines input such as similar mobile phone, computer.For example, user exists
In the search box of electric business platform input " American-European sandals are flat ", after clicking trigger function of search, " American-European sandals are flat " just
By as user's search term.When actually realizing, which, which not only can be, is inputted by text, can also be with
It is using voice input or icon-based programming input, two-dimensional code scanning input etc..These input and obtain user's search term
Mode can be all contemplated.
Step 102: determining rewriting search term similar with user's search term, wherein the rewriting search term is base
The search that the similarity corresponding with user's search term that behavior determines clicks behavior is clicked in the historical search of multiple users
Word;
This is similar to can be understood as expressing approximate search intention, for example, " American-European sandals are flat " and " calceus angle "
It is similar.For the angle of computer disposal, in order to realize that this similar matching determines, it can be using similarity feature
Mode.That is, obtaining the similarity feature of user's search term and the similarity feature of other search terms, it is then based on similarity spy
The Similarity matching of sign is determining to rewrite search terms with the matched one or more of user's search term.
In this example, behavior is clicked based on the historical search of multiple users and determines that rewriting similar with user's search term is searched for
Word does not need to pre-establish synonymicon, and accuracy is relatively high for existing expanded search mode, therefore,
The historical search based on multiple users in this example clicks behavior and determines that rewriting search term can effectively improve search and return the result
Quantity and accuracy, improve user experience.
Step 103: being scanned for according to user's search term and the rewriting search term, return to search result.
It scans for matching based on user's search term specifically, can be, obtains the first search matching result;It is based on
The rewriting search term scans for matching, and obtains the second search matching result;Then, by it is described first search matching result and
The second search matching result is returned together as the search result based on user's search term;
For example, behavioral data etc. can be clicked based on historical search, the similarity excavated between each Query is special
Sign.For example, it may be the similarity excavated between Query is special from dimensions such as the behavior of click, session (session situation), semantemes
Sign, for each search, the similarity feature excavated can be indicated are as follows: F1, F2, F3 ... Fn.
After having obtained the similarity feature of multiple Query, so that it may similarity feature and use based on each Query
The similarity feature of family search term, determines the similarity between each Query and user's search term.Determine similarity it
Afterwards, can extract the highest K Query of similarity be used for together with user's search term search for recall, or selection similarity
Query beyond preset threshold is used to search for together with user's search term and recall.
By taking " American-European sandals " as an example, as shown in Fig. 2, when carrying out rewriting search term matching, it is available and " American-European
The user of sandals " searches for the approximate word of click behavior as rewriting search term.For example, available or extraction " American-European sandals "
Similarity feature.Then, the similarity feature of other search terms is obtained, these similarity features are based on, can be determined and " Europe
The high several search terms of U.S. sandals " similarity-rough set, such as " intersecting sandals female bandage Rome ", " sandals summer bandage ", " sandals
Frenulum Rome ".So " it can will intersect sandals female bandage Rome ", " sandals summer bandage ", " sandals frenulum Rome " conduct " Europe
The rewriting search term of U.S. sandals " " will intersect sandals female bandage Rome ", " sandals summer bandage ", " sandals frenulum Rome " and " Europe
U.S. sandals " scan for matching together as search term, and obtained search after matching will be scanned for based on these search terms
As a result the search result all as " American-European sandals " is called back.
It in one embodiment, can be to based on " intersecting sandals female bandage Rome ", " sandals summer bandage ", " sandals
Frenulum Rome " and " American-European sandals " scan for matching obtained search result progress duplicate removal processing as search term, will go
Result after weight is recalled as final search result.
For another example can determine that the rewriting search term of " the sun-proof sunshading board of car " is " automobile heat insulation through the above way
Plate ", the rewriting search term of " baby ride toy " are " inflatable toys jump horse-vaulting ", and the rewriting search term of " upper and lower bunk bed " is
" two-berth basket cot bunk beds ", the rewriting search term of " God of Wealth's figure of buddha " are " the literary mammon ", " beggar's trousers summer tide trousers are for men "
Rewriting search term be " knee broken hole breathe out human relations jeans male ", the rewriting search term of " u disk mobile phone apple " is " memory flash vessel
iphone”。
It should be noted, however, that above-mentioned is in click behavior, session (session situation) dimension, semantic dimension etc.
As the dimension for excavating the similarity feature between Query, when actually realizing, can also be extracted from other dimensions similar
Feature is spent, such as: click the similarity etc. of the picture of product.
In one embodiment, it when matching obtains rewriting search term, can be through real-time clustering
Mode is also possible to using pre-establishing corresponding relationship, or pre-establishes what the mode of Matching Model carried out.It is specific to use
Which kind of mode can select according to actual needs.Above-mentioned several matching ways are described as follows below:
1) real-time clustering:
Real-time clustering can be and trigger when getting user's search term.Such as: user's input is " American-European
Sandals " are used as user's search term, obtain then the clusterings such as purchase data at this moment can be accessed by the history on electric business platform
To the similarity feature of " American-European sandals ", then also clustering obtains the similarity feature of other Query.It is then based on similar
It spends feature and carries out similarity mode, to determine and " American-European sandals " similar rewriting search term.
2) corresponding relationship is pre-established:
I.e., it is possible to pre-establish the corresponding rewriting search term of each search term.For example, can be in advance in electric business sample platform
Historical search is clicked behavioral data and is excavated, and determines the rewriting search term of each user's search term.It after determination, can be with
A concordance list either array set etc. is established, it is whether similar between each search term for storing, or storage
The similar rewriting search term of each search term.In this way after getting user's search term, can directly from the concordance list or
It is that the similar rewriting search term of user's search term is found out in data acquisition system.It can effectively improve the matched effect of search in this way
Rate.
3) Matching Model is established:
I.e., it is possible to which when establishing the Matching Model, training data be can be manually by establishing a Matching Model
The similar or dissimilar search term of multiple original user search words of mark, so that subsequent to the mode input one
After user's search term, it can go out search term similar with user's search term by the Model Matching.If training data
It is to be inputted according to similarity size, then it is similar with user's search term each to train obtained Matching Model that can identify
Similarity height between search term and user's search term, can be obtained in this way according to determining similarity height similarity compared with
High one or more search terms are used to search for together and recall.
For example, as shown in figure 3, Matching Model can be established in the following way:
S1: it is obtained from electric business platform and clicks behavioral data, session (after user logs in electric business platform, between a period of time
Every interior continuous search, click, the behavior sequences such as purchase), semantic dictionary.Then, based on click behavioral data, session, language
Adopted dictionary excavates the similarity feature between Query, can be expressed as f1, f2, f3.
S2: to each original Query (that is, user's search term), based on three above dimension (click behavioral data,
Session, semanteme) topK1 candidate Query of Feature Selection and duplicate removal processing is carried out, similar Query is generated to (q, q ').
S3: to each similar Query to (q, q '), original Query, the rewriting of the similar Query centering are extracted
Similarity feature of the Query and similar Query to (q, q '), wherein the similarity feature can include but is not limited to
The similarity feature for three positions stated.
S4: either can go out some original Query according to rule extraction at random, and from above-mentioned multiple similar
Query is manually labeled this n q ' to the corresponding n q ' of these original Query is randomly selected out in (q, q '), for example,
Mark are as follows: good (much like), fair (general similar), bad (dissmilarity).
S5: Matching Model is carried out by above-mentioned labeled data and characteristic (q, q ', label, f1, f2, f3 ...)
Training, can be ranked up all similar candidates q ' of each original Query by the Matching Model.
It can determine to be searched together with the similar Query of similar with original Query topK item based on above-mentioned Matching Model
Rope is recalled.
Above-mentioned is the explanation carried out with the search behavior in electric business field, however, above-mentioned extension is searched when actually realizing
The mode of rope can be applied not only to electric business field, can also be applied to the fields such as travelling search, browsing search, can be based on
The historical search of multiple users clicks behavior and determines rewriting search term, is searched together by user's search term and rewriting search term
Hitch fruit is recalled.
Above-mentioned searching and matching method is illustrated below with reference to a concrete scene, it should be noted, however, that the tool
Body embodiment does not constitute an undue limitation on the present application merely to the application is better described.
As shown in figure 4, being searched using user's search term as object matching is matched to be existing after obtaining user's search term
Hitch fruit.For example, in the case where user's search term is " American-European sandals ", to " American-European sandals " segmented to obtain " America and Europe " and
" sandals ".Matched based on the word segmentation result obtained after participle with product title, obtain in product title exist " America and Europe " and
The product of " sandals " is returned as search result, and the product for " America and Europe " and " sandals " is not present in product title can not conduct
Search result can not be called back so as to cause some products for meeting user search intent as search result.
As shown in figure 5, to scan for matched flow diagram using the searching method of this example, as shown in figure 5, user
Search term be " American-European sandals ", after getting user's search term, be not based on directly on user's search term " American-European sandals " into
Row matching, but first user's search term " American-European sandals " is rewritten, it obtains similar rewriting search term and " intersects sandals
Female's bandage Rome ", " sandals summer bandage ", " sandals frenulum Rome ".Then, sandals female bandage " is intersected with " American-European sandals " OR
Rome " OR " sandals summer bandage " OR " sandals frenulum Rome " scans for recalling.
By actual experiment it can be found that if only using " American-European sandals " as search term, the result searched for
For 77w or so, if with " American-European sandals " OR " intersecting sandals female bandage Rome " OR " sandals summer bandage " OR " sandals frenulum sieve
Horse " is used as search term, then the product quantity that can be recalled is 320w or so.It can be seen that passing through search provided by this example
Method can effectively improve the recall rate of product, improve search efficiency, so that the product for more meeting user search intent is made
It is called back for search result, effectively increases the search experience of user.
Embodiment of the method provided by example can be in server, computer cluster or similar arithmetic unit in the application
Middle execution.For operating on computer cluster, Fig. 6 is a kind of terminal of searching method of the embodiment of the present invention
Hardware block diagram.As shown in fig. 6, terminal 10 may include one or more (only showing one in figure) processors
102 (processing units that processor 102 can include but is not limited to Micro-processor MCV or programmable logic device FPGA etc.) are used
Memory 104 in storing data and the transmission module 106 for communication function.Those of ordinary skill in the art can manage
Solution, structure shown in fig. 6 are only to illustrate, and do not cause to limit to the structure of above-mentioned electronic device.For example, terminal 10
It may also include than shown in Fig. 6 more perhaps less component or with the configuration different from shown in Fig. 6.
Memory 104 can be used for storing the software program and module of application software, such as the short message in the embodiment of the present invention
Corresponding program instruction/the module of the sending method of breath, the software program that processor 102 is stored in memory 104 by operation
And module realizes the transmission of the short message of above-mentioned application program thereby executing various function application and data processing
Method.Memory 104 may include high speed random access memory, may also include nonvolatile memory, such as one or more magnetism
Storage device, flash memory or other non-volatile solid state memories.In some instances, memory 104 can further comprise phase
The memory remotely located for processor 102, these remote memories can pass through network connection to terminal 10.On
The example for stating network includes but is not limited to internet, intranet, local area network, mobile radio communication and combinations thereof.
Transmission module 106 is used to that data to be received or sent via a network.Above-mentioned network specific example may include
The wireless network that the communication providers of terminal 10 provide.In an example, transmission module 106 includes that a network is suitable
Orchestration (Network Interface Controller, NIC), can be connected by base station with other network equipments so as to
Internet is communicated.In an example, transmission module 106 can be radio frequency (Radio Frequency, RF) module,
For wirelessly being communicated with internet.
Above-mentioned searching request can be user's input, specifically, user can input searching request by client.Its
In, the client can be the terminal device or software that guest operation uses.Specifically, client can be intelligent hand
The terminal devices such as machine, tablet computer, laptop, desktop computer, smartwatch or other wearable devices.Certainly,
Client is also possible to the software that can be run in above-mentioned terminal device.Such as: mobile phone Taobao, Alipay or browser etc. are answered
Use software.
Referring to FIG. 7, it may include: the first acquisition that the searcher, which is applied in server, in Software Implementation
Unit 701, determination unit 702 and second acquisition unit 703.Wherein:
First acquisition unit 701, for obtaining user's search term;
Determination unit 702, for determining rewriting search term similar with user's search term, wherein the rewriting is searched
Rope word is the historical search click behavior determination based on multiple users;
Second acquisition unit 703, for being scanned for according to user's search term and the rewriting search term, return is searched
Hitch fruit.
In one embodiment, second acquisition unit 703 can scan for matching based on user's search term, obtain
To the first search matching result;It scans for matching based on the rewriting search term, obtains the second search matching result;Then,
By the first search matching result and the second search matching result together as the search based on user's search term
As a result it is returned;
In one embodiment, above-mentioned determination unit 702 can be used for obtaining the similarity spy of user's search term
Sign;Highest one or more rewrite of similarity between similarity feature and the similarity feature of user's search term is searched
Rope word is as search term similar with user's search term.
In one embodiment, above-mentioned determination unit 702 can be used for obtaining the similarity spy of user's search term
Sign;Similarity between similarity feature and the similarity feature of user's search term is exceeded one or more of preset threshold
A rewriting search term requests similar search term as with described search.
In one embodiment, above-mentioned similarity feature can include but is not limited at least one following dimensional characteristics:
Click behavioral data, session, semanteme.
In one embodiment, second acquisition unit 703 specifically can be used for searching user's search term and the rewriting
Rope word carries out word segmentation processing respectively and obtains participle segment;It scans for matching based on the participle segment, obtain in product title
There are the products of the participle segment;The product that matching is obtained is as search result.
In one embodiment, the search that generation unit 704 specifically can be used for obtaining based on user's search term
As a result the search result and based on the rewriting search term obtained carries out duplicate removal processing;The search knot that will be obtained after duplicate removal processing
Fruit, as search result.
Searching method and server provided by the present application are not directly to be searched for user after getting user's search term
Word scans for result and recalls, but determines rewriting search term similar with user's search term, then will be searched for based on user
The search result that word and rewriting search term match is returned together as search result, to solve existing search
The difference because of semantic statement in the presence of method causes some some products for meeting user search intent that can not be called back
The technical issues of, reach the recall rate for effectively improving product, search efficiency has been improved, so that more meeting user search intent
Product be called back as search result, effectively increase the search experience of user.
Although this application provides the method operating procedure as described in embodiment or flow chart, based on conventional or noninvasive
The labour for the property made may include more or less operating procedure.The step of enumerating in embodiment sequence is only numerous steps
One of execution sequence mode, does not represent and unique executes sequence.It, can when device or client production in practice executes
To execute or parallel execute (such as at parallel processor or multithreading according to embodiment or method shown in the drawings sequence
The environment of reason).
The device or module that above-described embodiment illustrates can specifically realize by computer chip or entity, or by having
The product of certain function is realized.For convenience of description, it is divided into various modules when description apparatus above with function to describe respectively.
The function of each module can be realized in the same or multiple software and or hardware when implementing the application.It is of course also possible to
Realization the module for realizing certain function is combined by multiple submodule or subelement.
Method, apparatus or module described herein can realize that controller is pressed in a manner of computer readable program code
Any mode appropriate is realized, for example, controller can take such as microprocessor or processor and storage can be by (micro-)
The computer-readable medium of computer readable program code (such as software or firmware) that processor executes, logic gate, switch, specially
With integrated circuit (Application Specific Integrated Circuit, ASIC), programmable logic controller (PLC) and embedding
Enter the form of microcontroller, the example of controller includes but is not limited to following microcontroller: ARC 625D, Atmel AT91SAM,
Microchip PIC18F26K20 and Silicone Labs C8051F320, Memory Controller are also implemented as depositing
A part of the control logic of reservoir.It is also known in the art that in addition to real in a manner of pure computer readable program code
Other than existing controller, completely can by by method and step carry out programming in logic come so that controller with logic gate, switch, dedicated
The form of integrated circuit, programmable logic controller (PLC) and insertion microcontroller etc. realizes identical function.Therefore this controller
It is considered a kind of hardware component, and hardware can also be considered as to the device for realizing various functions that its inside includes
Structure in component.Or even, it can will be considered as the software either implementation method for realizing the device of various functions
Module can be the structure in hardware component again.
Part of module in herein described device can be in the general of computer executable instructions
Upper and lower described in the text, such as program module.Generally, program module includes executing particular task or realization specific abstract data class
The routine of type, programs, objects, component, data structure, class etc..The application can also be practiced in a distributed computing environment,
In these distributed computing environment, by executing task by the connected remote processing devices of communication network.In distribution
It calculates in environment, program module can be located in the local and remote computer storage media including storage equipment.
As seen through the above description of the embodiments, those skilled in the art can be understood that the application can
It is realized by the mode of software plus required hardware.Based on this understanding, the technical solution of the application is substantially in other words
The part that contributes to existing technology can be embodied in the form of software products, and can also pass through the implementation of Data Migration
It embodies in the process.The computer software product can store in storage medium, such as ROM/RAM, magnetic disk, CD, packet
Some instructions are included to use so that a computer equipment (can be personal computer, mobile terminal, server or network are set
It is standby etc.) execute method described in certain parts of each embodiment of the application or embodiment.
Each embodiment in this specification is described in a progressive manner, the same or similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.The whole of the application or
Person part can be used in numerous general or special purpose computing system environments or configuration.Such as: personal computer, server calculate
Machine, handheld device or portable device, mobile communication terminal, multicomputer system, based on microprocessor are at laptop device
System, programmable electronic equipment, network PC, minicomputer, mainframe computer, the distribution including any of the above system or equipment
Formula calculates environment etc..
Although depicting the application by embodiment, it will be appreciated by the skilled addressee that the application there are many deformation and
Variation is without departing from spirit herein, it is desirable to which the attached claims include these deformations and change without departing from the application's
Spirit.
Claims (13)
1. a kind of searching method, which is characterized in that the described method includes:
Obtain user's search term;
Determine rewriting search term similar with user's search term, wherein the rewriting search term is based on multiple users
The search term that the similarity corresponding with user's search term that behavior determines clicks behavior is clicked in historical search;
It scans for matching based on user's search term, obtains the first search matching result;
It scans for matching based on the rewriting search term, obtains the second search matching result;
By the first search matching result and the second search matching result together as based on user's search term
Search result is returned.
2. the method according to claim 1, wherein determining rewriting search similar with user's search term
Word, comprising:
Obtain the similarity feature of user's search term;
By the highest one or more rewritings of the similarity between similarity feature and the similarity feature of user's search term
Search term is as rewriting search term similar with user's search term.
3. the method according to claim 1, wherein determining rewriting search similar with user's search term
Word, comprising:
Obtain the similarity feature of user's search term;
By the similarity between similarity feature and the similarity feature of user's search term beyond one of preset threshold or
Multiple rewriting search terms are as rewriting search term similar with user's search term.
4. according to the method in claim 2 or 3, which is characterized in that the similarity feature include following dimensional characteristics extremely
It is one of few: to click behavioral data, semanteme.
5. the method according to claim 1, wherein according to user's search term and the rewriting search term into
Row search, returns to search result, comprising:
Word segmentation processing is carried out to user's search term and the rewriting search term respectively and obtains participle segment;
Scan for matching based on the participle segment, obtain in product title there are it is described participle segment product;
The product that matching is obtained is as search result.
6. the method according to claim 1, wherein by the first search matching result and second search
Matching result is returned together as the search result based on user's search term, comprising:
Duplicate removal processing is carried out to the first search matching result and the second search matching result;
The search matching result that will be obtained after duplicate removal processing is returned as search result.
7. a kind of server, which is characterized in that including processor and for the memory of storage processor executable instruction, institute
It states processor and executes following steps:
Obtain user's search term;
Determine rewriting search term similar with user's search term, wherein the rewriting search term is based on multiple users
The search term that the similarity corresponding with user's search term that behavior determines clicks behavior is clicked in historical search;
It scans for matching based on user's search term, obtains the first search matching result;
It scans for matching based on the rewriting search term, obtains the second search matching result;
By the first search matching result and the second search matching result together as based on user's search term
Search result is returned.
8. server according to claim 7, which is characterized in that the processor is specifically used for obtaining user's search
The similarity feature of word;By similarity highest one between similarity feature and the similarity feature of user's search term
Or multiple rewriting search terms are as rewriting search term similar with user's search term.
9. server according to claim 7, which is characterized in that the processor is specifically used for obtaining user's search
The similarity feature of word;Similarity between similarity feature and the similarity feature of user's search term is exceeded into default threshold
One or more search terms of rewriting of value are as rewriting search term similar with user's search term.
10. server according to claim 8 or claim 9, which is characterized in that the similarity feature includes following dimensional characteristics
At least one: click behavioral data, semanteme.
11. server according to claim 7, which is characterized in that the processor is specifically used for searching for the user
Word and the rewriting search term carry out word segmentation processing respectively and obtain participle segment;It scans for matching based on the participle segment,
Obtain in product title there are it is described participle segment product;The product that matching is obtained is as search result.
12. server according to claim 7, which is characterized in that the processor searches for matching result for described first
It is returned with the second search matching result together as the search result based on user's search term, comprising:
Duplicate removal processing is carried out to the first search matching result and the second search matching result;
The search matching result that will be obtained after duplicate removal processing is returned as search result.
13. a kind of computer readable storage medium is stored thereon with computer instruction, described instruction, which is performed, realizes that right is wanted
The step of seeking any one of 1 to 6 the method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710735984.1A CN109948014A (en) | 2017-08-24 | 2017-08-24 | A kind of searching method and server |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710735984.1A CN109948014A (en) | 2017-08-24 | 2017-08-24 | A kind of searching method and server |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109948014A true CN109948014A (en) | 2019-06-28 |
Family
ID=67003898
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710735984.1A Pending CN109948014A (en) | 2017-08-24 | 2017-08-24 | A kind of searching method and server |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109948014A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110362749A (en) * | 2019-07-15 | 2019-10-22 | 北京百度网讯科技有限公司 | Search result recalls method, apparatus, server and medium |
CN110543593A (en) * | 2019-08-01 | 2019-12-06 | 北京三快在线科技有限公司 | Data processing method and device, electronic equipment and readable storage medium |
CN111782935A (en) * | 2020-05-12 | 2020-10-16 | 北京三快在线科技有限公司 | Information recommendation method and device, electronic equipment and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942279A (en) * | 2014-04-01 | 2014-07-23 | 百度(中国)有限公司 | Method and device for showing search result |
CN104598546A (en) * | 2014-12-21 | 2015-05-06 | 郭和友 | Synonym search system |
CN104834698A (en) * | 2015-04-27 | 2015-08-12 | 百度在线网络技术(北京)有限公司 | Information pushing method and device |
CN105224554A (en) * | 2014-06-11 | 2016-01-06 | 阿里巴巴集团控股有限公司 | Search word is recommended to carry out method, system, server and the intelligent terminal searched for |
US20160078086A1 (en) * | 2014-09-15 | 2016-03-17 | Google Inc. | Query Rewriting Using Session Information |
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 |
-
2017
- 2017-08-24 CN CN201710735984.1A patent/CN109948014A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942279A (en) * | 2014-04-01 | 2014-07-23 | 百度(中国)有限公司 | Method and device for showing search result |
CN105224554A (en) * | 2014-06-11 | 2016-01-06 | 阿里巴巴集团控股有限公司 | Search word is recommended to carry out method, system, server and the intelligent terminal searched for |
US20160078086A1 (en) * | 2014-09-15 | 2016-03-17 | Google Inc. | Query Rewriting Using Session Information |
CN104598546A (en) * | 2014-12-21 | 2015-05-06 | 郭和友 | Synonym search system |
CN104834698A (en) * | 2015-04-27 | 2015-08-12 | 百度在线网络技术(北京)有限公司 | Information pushing method and device |
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 |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110362749A (en) * | 2019-07-15 | 2019-10-22 | 北京百度网讯科技有限公司 | Search result recalls method, apparatus, server and medium |
CN110543593A (en) * | 2019-08-01 | 2019-12-06 | 北京三快在线科技有限公司 | Data processing method and device, electronic equipment and readable storage medium |
CN111782935A (en) * | 2020-05-12 | 2020-10-16 | 北京三快在线科技有限公司 | Information recommendation method and device, electronic equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112650907B (en) | Search word recommendation method, target model training method, device and equipment | |
CN105446973B (en) | The foundation of user's recommended models and application method and device in social networks | |
US9727584B2 (en) | Refining image annotations | |
CN110069650A (en) | A kind of searching method and processing equipment | |
CN111753198A (en) | Information recommendation method and device, electronic equipment and readable storage medium | |
US20150067505A1 (en) | System And Methods For User Curated Media | |
CN105027121A (en) | Indexing application pages of native applications | |
CN111949814A (en) | Searching method, searching device, electronic equipment and storage medium | |
CN109948014A (en) | A kind of searching method and server | |
US20210049354A1 (en) | Human object recognition method, device, electronic apparatus and storage medium | |
CN110674394B (en) | Knowledge graph-based information recommendation method, device and storage medium | |
CN107766399A (en) | For the method and system and machine readable media for image is matched with content item | |
CN110209810A (en) | Similar Text recognition methods and device | |
CN104809223A (en) | Method and device for supplying application content search result in application | |
Aydin et al. | Sail: Saliency-driven injection of aria landmarks | |
Jiang et al. | SLTFNet: A spatial and language-temporal tensor fusion network for video moment retrieval | |
KR101955920B1 (en) | Search method and apparatus using property language | |
CN113037925B (en) | Information processing method, information processing apparatus, electronic device, and readable storage medium | |
CN111666417B (en) | Method, device, electronic equipment and readable storage medium for generating synonyms | |
Aranda-Corral et al. | Reconciling knowledge in social tagging web services | |
CN117033799A (en) | Resource recommendation method, device, computer equipment and storage medium | |
CN111723280B (en) | Information processing method and device, storage medium and electronic equipment | |
CN112801053B (en) | Video data processing method and device | |
Zhang et al. | Ideagraph plus: A topic-based algorithm for perceiving unnoticed events | |
CN109408794A (en) | A kind of frequency dictionary method for building up, segmenting method, server and client side's equipment |
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: 20190628 |