CN110020101A - The restoring method of search scene, device and system in real time - Google Patents

The restoring method of search scene, device and system in real time Download PDF

Info

Publication number
CN110020101A
CN110020101A CN201710747385.1A CN201710747385A CN110020101A CN 110020101 A CN110020101 A CN 110020101A CN 201710747385 A CN201710747385 A CN 201710747385A CN 110020101 A CN110020101 A CN 110020101A
Authority
CN
China
Prior art keywords
search
word string
scene
personalized
initial
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.)
Granted
Application number
CN201710747385.1A
Other languages
Chinese (zh)
Other versions
CN110020101B (en
Inventor
张勇乔
李剑
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Taobao China Software Co Ltd
Original Assignee
Alibaba Group Holding Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Priority to CN201710747385.1A priority Critical patent/CN110020101B/en
Publication of CN110020101A publication Critical patent/CN110020101A/en
Application granted granted Critical
Publication of CN110020101B publication Critical patent/CN110020101B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management

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

This application discloses a kind of restoring method of scene of search in real time, a kind of restoring method including scene of search in real time, it comprises determining that the search scene that needs reappear, the initial search word string of corresponding described search scene is mutually merged with personalized adjusting parameter, forms reproducible search word string;Search is executed according to the reproducible search word string, returns to the search scene of reproduction.The application provides a kind of reduction apparatus of scene of search in real time simultaneously.Using method provided by the present application, in real time individual search process, the changing factor that will affect search screening and sequencer procedure is recorded, and makes it possible accurately to reappear personalized search results.Due to just having carried out relative recording during personalized search, by related art method, the real-time reproduction to the whole network user's humongous search behavior can be realized in the case where mass data using this method.

Description

The restoring method of search scene, device and system in real time
Technical field
This application involves real-time search fields, and in particular to a kind of restoring method of the scene of search in real time.The application is simultaneously It is related to the reduction apparatus of scene of search in real time a kind of, a kind of scene of search in real time goes back original system and a kind of scene of search in real time also Former electronic equipment and storage equipment.
Background technique
Existing web search technology has been realized in real time individual.For example, currently used real time individual search Technology is the real time individual based on search pora and igraph.It is that a user is directed to one by key with user under the technology The once-through operation (for example clicking, collection ...) of a object (such as commodity) is abstracted as an event, captures the whole network user's in real time Up-to-date event is synchronized to igraph, and in combination with the historical events got before system, driving executes all user's dimensions Algoritic module, the real time individual data of output user's dimension, user's dimension of real-time update on-line storage (such as igraph) Individuation data, to influence the personalized search results of the user.
In many instances, need to restore the actual conditions of certain search;But due to real time individual search technique It influences, after user feedback itself search result goes wrong, is gone wrong by original user's request URL come reappearing user Scene is no longer feasible.
For example, user's request URL is only used only with regard to nothing when the unreasonable situation of search results ranking occur and needing to solve Method obtains original search result.Such as: the various of seller's user feedback can not find oneself shop commodity or sequence in the search Rearward, not various generate in the search undesirably (mark is not pair or price is not right, and postage is incorrect etc.) of buyer's feedback Commodity;If not considering the influence of real time individual search, the problem of user is reflected cannot be correctly understood.
Likewise, different user needs to transmit search knot between each other under the influence of the search of above-mentioned real time individual Fruit will also be difficult to realize.For example, A buyer has carried out the search of some commodity on Taobao website, and it is desirable that by search result point B buyer is enjoyed, user search request URL is then supplied to B;In A buyer, the sharing that he provides should enable B client Enough see the commodity that he searches.But due to search for be real time individual search, so search result according to B the case where It is adjusted, the user's request URL for causing B to provide using A can not but obtain result identical with A;Therefore, real time individual Search, which hampers, shares search result between different search users, for electric business, reduces the sharing enjoyment of buyer, detracts User experience.
Such issues that prior art solves in the following manner:
By the commodity id in offline logs in the commodity of the search of represent layer reappearing user as a result, restoring search result Commodity snapshot;
In the search scene of data Layer part reappearing user by way of the secondary inquiry of the request URL in offline logs, Restore commodity score and commodity mark;
By the way that necessity search log is stored and analyzed, in the complete search scene of full link reappearing user, reduction Every search result.
The above-mentioned prior art cannot accurately and rapidly restore personalized search result;The mass data the case where Under, it is difficult to meet the real-time reproduction to the whole network user's humongous search behavior;It can not easily realize between different search users Search result is shared.
Summary of the invention
It is the replay method that scene is searched in a kind of real-time system provided by the present application that the application, which provides, to solve existing reality When personalized search reappear the problem of.In addition the application provides the reproducer that scene is searched in a kind of real-time system and a kind of reality When system in search for scene playback system.
The application provides a kind of restoring method of scene of search in real time, comprising:
It determines the search scene for needing to reappear, the initial search word string of corresponding described search scene and personalized adjustment is joined Number mutually merges, and forms reproducible search word string;
Search is executed according to the reproducible search word string, returns to the search scene of reproduction.
Preferably, to realize this method, in executing search process, following preparation is carried out:
Search is being executed, during obtaining personalized search results, is recording initial search word string, and mention for search scene The personalized adjusting parameter provided for personalized search.
Preferably, the record initial search word string and the personalization for providing personalized search for search scene and providing are adjusted Before the step of whole parameter, following step is executed:
For the generation of each searching request, uniquely corresponding searching request is identified with the searching request;Described search request mark Each data during requesting entire survival with described search are associated.
Preferably, described to execute search, during obtaining personalized search results, records initial search word string and be The personalized adjusting parameter for providing personalized search to search main body and providing, comprising:
Searching request is obtained, uniquely corresponding searching request identifies with the searching request for the generation of each searching request;
The initial search word string is extracted according to described search request, records the initial search word string;
Initial search result set is obtained according to the initial search word string, records the initial search result set;
According to the personalized adjusting parameter of acquisition, personalized search adjustment is carried out to the initial search result set, is obtained Obtain the personalized search results to search main display;
It during executing above-mentioned steps, is monitored, records the personalized situation of the query process.
Preferably, described to execute search, during obtaining personalized search results, records initial search word string and be The personalized adjusting parameter for providing personalized search to search main body and providing, record therein are based on log system, comprising:
Described the step of extracting the initial search word string according to described search request, record the initial search word string In, the first log is written in the initial search word string by front end services;
It is described that initial search result set is obtained according to the initial search word string, record the initial search result set The step of in, the front end services will the search result identification list that returned be written the first log in;
Described to be monitored, in the step of recording the personalized situation of the query process, search service searches personalization Rope parameter is written in the second log, and the random algorithm parameter during personalized search is written in the second log.
Preferably, described to execute search, during obtaining personalized search results, while recording described search knot Fruit;Described search result includes the search result identification list that search engine is returned when executing search.
Preferably, the initial search word string includes: search command and search content.
Preferably, the personalized adjusting parameter includes: personalized search parameter and random algorithm parameter.
Preferably, the personalized search parameter includes: screening search parameter and/or sorted search parameter.
Preferably, the determining search scene for needing to reappear, by the initial search word string of corresponding described search scene with The step of personalized adjusting parameter mutually merges, and forms reproducible search word string, comprising:
The personalized adjusting parameter is merged, the initial search word string is added, forms the reproducible search word string.
Preferably, described that the personalized adjusting parameter is merged into the addition initial search word string, it can be weighed described in composition The step of now searching for word string, comprising: the reproducible search word string is added in search result identification list.
Preferably, the determining search scene for needing to reappear, by the initial search word string of corresponding described search scene with The step of personalized adjusting parameter mutually merges, and forms reproducible search word string, institute was formed by searching for the log system data State reproducible search word string, comprising:
The search scene for needing to reappear is obtained, requests mark to obtain first log and described second by described search Corresponding data in log;
The data obtained in first log and second log are recombinated by preset format, obtains reappearing and search Rope word string.
Preferably, before the step of executing search according to the reproducible search word string, return to the search scene of reproduction, packet It includes: parsing the reproducible search word string by preset format.
Preferably, described to parse the reproducible search word string by preset format, parsing result includes: initial search word String, the personalized adjusting parameter provided to provide personalized search to search main body, object search identification list.
Preferably, described that search is executed according to the reproducible search word string, return to the search scene of reproduction, comprising:
The search scene of the reproduction is presented in the form of snapshot.
Preferably, user shares search scene in the following way:
The word string that re-searches for of corresponding described search scene is sent to by sharing people.
Preferably, user shares search scene in the following way: the search for the described search scene shared will be needed to ask It asks mark to be sent to and is shared people.
The application provides a kind of reduction apparatus of scene of search in real time simultaneously, comprising:
Word string unit is recombinated, the search scene reappeared for determining needs, by the initial search of corresponding described search scene Word string adjusts parameter mutually to merge with personalization, forms reproducible search word string;
Scene unit is returned, for executing search according to the reproducible search word string, returns to the search scene of reproduction.
Preferably, comprising:
Recording unit during obtaining personalized search results, records initial search word string for executing search, and The personalized adjusting parameter provided to provide personalized search to search for.
Preferably, including searching request mark unit is generated, for generating with the searching request only for each searching request One corresponding searching request mark;Described search request mark requests each data phase during entire survival with described search Association.
Preferably, the recording unit, comprising:
It generates searching request and identifies unit, for obtaining searching request, generated and the searching request for each searching request Unique corresponding searching request mark;
Initial search word string unit is extracted, for extracting the initial search word string according to described search request, records institute State initial search word string;
Result unit is obtained, for obtaining initial search result set according to the initial search word string, is recorded described first Beginning search result set;
Personalized search unit is shown, for the personalized adjustment factor according to acquisition, to the initial search result collection It closes and carries out personalized search adjustment, obtain the personalized search results to search main display;
Monitoring unit, for being monitored, recording the personalization of the query process during executing above-mentioned steps Situation.
What the application also provided a kind of scene of search in real time goes back original system, including the service mutually communicated by network implementations Device, client, database;
The server is executing search, during obtaining personalized search results, records initial search word string and is Personalized search is provided to search main body and the personalized adjusting parameter that provides is to database;
When receiving client instructions, when reappearing described search scene, server will correspond to described search in the database The initial search word string of scene mutually merges with the personalized adjusting parameter, forms reproducible search word string;
Search is executed according to the reproducible search word string, returns to the search scene of reproduction to client.
Preferably, the server is executing search, during obtaining personalized search results, records initial search word String and the personalized adjusting parameter provided to provide personalized search to search main body are to database, and record therein is based on day Aspiration system, comprising:
Client initiates searching request, and server obtains searching request, generates and the searching request for each searching request Unique corresponding searching request mark;
The initial search word string is extracted according to described search request, records the initial search word string, front end services will The first log is written in the initial value search word string;
Initial search result set is obtained according to the initial search word string, records the initial search result set, institute It states front end services the object search identification list returned is written in the first log;
According to the personalized adjusting parameter of acquisition, personalized search adjustment is carried out to the initial search result set, is obtained Obtain the personalized search results to search main display;
It during executing above-mentioned steps, is monitored, records the personalized situation of the query process, search service Personalized search parameter is written in the second log, and the second log is written into the random algorithm parameter during personalized search In.
Preferably, the server is executing search, during obtaining personalized search results, while searching described in record Hitch fruit;Described search result includes the search result identification list that search engine is returned when executing search.
Preferably, described when receiving client instructions, when reappearing described search scene, server will correspond to institute in database The initial search word string for stating search scene mutually merges with the personalized adjusting parameter, forms reproducible search word string, wraps It includes:
Server obtains redisplaying information, requests mark to obtain first log and second log by described search In corresponding data;
The data of acquisition in first log and second log are recombinated by preset format, are obtained described Reproducible search word string.
Preferably, the step of server executes search, returns to the search scene of reproduction according to the reproducible search word string Before, comprising: parse the reproducible search word string by preset format.
Preferably, described that search is executed according to the reproducible search word string, return to the search scene of reproduction to client, Include:
The search scene of the reproduction is presented in the form of snapshot.
The application provides a kind of electronic equipment of scenario reduction of search in real time, and the electronic equipment includes:
Display;
Processor;
Memory, for determining the search scene reappeared of needs, by the initial search word string of corresponding described search scene with Personalized adjusting parameter mutually merges, and forms reproducible search word string, and the personalization adjusting parameter is to provide to search main body Personalized search and the parameter provided;
Search is executed according to the reproducible search word string, returns to the search scene of reproduction.
The application also provides a kind of for searching for the storage equipment of scenario reduction in real time, which is stored with instruction, Described instruction can be loaded by processor and execute following steps:
It determines the search scene for needing to reappear, the initial search word string of corresponding described search scene and personalized adjustment is joined Number mutually merges, and forms reproducible search word string, the personalization adjusting parameter be to provide personalized search to search main body and The parameter of offer;
Search is executed according to the reproducible search word string, returns to the search scene of reproduction.
Compared with prior art, the method again that scene is searched in real-time system provided by the present application, is executing search, is obtaining During obtaining personalized search results, records initial search word string and provide personalized search to search main body and provide Personalized adjustment factor;When reappearing described search scene, by the initial search word string of corresponding described search scene and institute It states personalized adjusting parameter mutually to merge, forms reproducible search word string;Search is executed according to the reproducible search word string, it can be with Reappear search scene.
It is had an advantage that using the above method
Firstly, the changing factor that will affect search screening and sequencer procedure is recorded in real time individual search process Come, in this way, just making it possible accurately to reappear personalized search results.
Secondly as relative recording has just been carried out during personalized search, it can be in mass data using this method In the case where, by related art method, realize the real-time reproduction to the whole network user's humongous search behavior.
It again, can be by way of transmitting reproducible search word string among different users after using the technical program Share search scene, effectively promotes the search experience of user.
Detailed description of the invention
Fig. 1 is the replay method flow chart that scene is searched in the real-time system of the application first embodiment offer;
Fig. 2 is a kind of method flow diagram of record log provided by the present application;
Fig. 3 is the reproducer schematic diagram that scene is searched in the real-time system of the application second embodiment offer;
Fig. 4 is the playback system schematic diagram that scene is searched in the real-time system of the application 3rd embodiment offer;
Fig. 5 is the reproduction electronic equipment schematic diagram that scene is searched in the real-time system of the application fourth embodiment offer.
Specific embodiment
Many details are explained in the following description in order to fully understand the application.But the application can be with Much it is different from other way described herein to implement, those skilled in the art can be without prejudice to the application intension the case where Under do similar popularization, therefore the application is not limited by following public specific implementation.
Using the large-scale website of the storage of real-time online figure and inquiry system (for example, iGraph is serviced), can adjust in real time Whole user personalized information and search result;But occur to encounter problems in customer complaint or development process (for example, particular commodity exists In website search less than, sequence rearward, price is incorrect, mark is inaccurate, result is inaccurate the problems such as) or line on failure when, because Search real time individual technology produces the problem of reproduction and investigation difficulty.In addition, in many cases, different user is expected that by User search request URL shares search result, but due to searching for real time individual technology, causes the result actually shared and different It causes, to influence user experience.
This application provides the replay methods that scene is searched in a kind of real-time system;Referring to FIG. 1, being that the application first is real Apply the replay method flow chart that scene is provided in the real-time system of example offer;This method is described in detail below with reference to Fig. 1, The example used in description is to assist the understanding of the application, is not that specific application environment limits.
Step S101: determining and need the search scene reappeared, by the initial search word string of corresponding described search scene and Property adjusting parameter mutually merges, and forms reproducible search word string.
Search scene is to carry out personalized search processing based on user search request, and return to the whole process of search result General name.Under current web search technology, real time individual processing can be carried out to the searching request of user, and is generated to certain The real time individual search result of a user search request.So-called personalized search results may include for same search Word string, it is different to the displaying content of the search result of different user, and show that the sequence of content is different etc..So-called real-time individual character Change processing refers to that search service can be according to the personalized factor of different search users, and the user feedback collected at any time, in real time The search result adjustment of above-mentioned personalization is carried out to search result.
For example, being directed to same search string " one-piece dress ", search server disappears according to the difference of different search users Take situation, the search result of different series is shown on the page;And for the same search user, according to user to first The click of the search result of page browses situation, just carries out to the displaying content of search result and displaying sequence in second page personalized Adjustment (that is: real time individual adjusts).Each result of page searching that each search user obtains can be seen as a search Scene.
When user initiates searching request according to demand, server can calculate search result using personalized service, finally obtain Obtain the personalized search results;During calculating search result using personalized service, in addition to using user to initiate to search Outside the initial search word string that rope request provides, the adjustment factor of personalization results also while using for search scene is provided, it is described Adjustment factor is personalized adjusting parameter;By the above-mentioned means, can be obtained the personalization for search scene.It is corresponding, If obtaining the initial search word string and personalized adjusting parameter of certain personalized search results, i.e., the knot of reproducible this time search Therefore fruit can reappear the search scene of personalized search with this.
From the above it can be seen that for the method for realizing reproduction search scene, it is necessary first in terms of obtaining following two Information, it may be assumed which search scene what determining needs reappeared is;And obtain the initial search word string of corresponding described search scene With personalized adjusting parameter.After obtaining information above, it is likely to when needing to reappear search scene, acquisition can be used in reappearing All information of described search scene.
In order to obtain above- mentioned information step, it is obviously desirable to just be prepared accordingly when executing search;Specifically, just It is to issue searching request in user, during server provides personalized search results, records initial search word string, and to search Rope scene provides the personalized adjusting parameter of real time individual search and offer.
Above-mentioned two aspect --- it executes the preparation carried out when search and obtains related letter when reappearing search scene Breath --- it combines, can finally realize described in this step: determining the search scene for needing to reappear, by corresponding described search The initial search word string of scene mutually merges with personalized adjusting parameter, reproducible search word string is formed, to re-search for scene.
It is just illustrated respectively around above-mentioned several aspects below
As previously mentioned, in order to realize the replay method for searching for scene in real-time system provided in this embodiment, it is necessary first to It just needs accordingly to be prepared when executing search, i.e., during obtaining personalized search results, records initial search word String, and the personalized adjusting parameter provided to provide personalized search to search main body.Above-mentioned steps are although belong to preparation step It suddenly, is to realize therefore the premise for reappearing search scene is below first described in detail the step.
As described above, needing the initial search word of some personalized search results of physical record in executing search process String and corresponding personalized adjusting parameter, the premise inevitable requirement personalized search results can by accurate identification, for this purpose, Firstly the need of generating for each searching request, uniquely corresponding searching request is identified with the searching request, it is clear that the searching request Mark needs in record initial search word string and for search scene provides personalized search and the personalized adjusting parameter that provides It is just obtained before step, hereafter, described search request mark requests each data phase during entire survival with described search Association.Specifically, by generating unique searching request mark RN (Request Number), as search access each time Unique identification, and as the unique parameters for identifying search access, and described search request mark and described search request is whole Each data during a survival are associated.After having above-mentioned searching request mark, the personalization it can both be searched with corresponding record The initial search word string of rope and personalized adjusting parameter.
For a further understanding of the initial search word string of record personalized search and the process of personalized adjusting parameter, below The specific implementation of personalized search is illustrated.
So-called real time individual search result is to search for obtain by real time individual, specifically, being exactly by user The user characteristics and typing search system of each dimension are converted into the click behavior of every page, and adjustment user is a in real time with this Property information and search result;Such as: preference is browsed in the click of search result first page according to user, corrects second page in real time It is presented to the search result of user, enhances the personalized of second page and shows.Certainly, when user issues searching request for the first time, Personalized search processing has been carried out according to the search main body relevant parameter obtained by the historical data of the user.
The prior art provides the tool for realizing real time individual search, for example, the iGraph for recording user characteristics Service.The system that iGraph is to provide real-time online figure storage and inquiry, it has a characteristic that efficient index of the picture and inquiry Mechanism;Super large figure (1,000,000,000 nodes, 10,000,000,000 relationships) are supported to index online;Flexible query language support;Pass through various plug-in unit branch Hold service logic customization;Millisecond rank inquiry response is provided, 10W magnitude qps handles up;Online diagram data in real time is supported to update;With The self-service service access system in family.For search service, during search service, user can in the click row of every page Be converted into each dimension user characteristics typing iGraph (these user characteristics search when be specifically reflected in real time individual In parameter).When user enters lower one page, so that it may obtain searching for further personalization by adjusting personalized adjusting parameter in real time Rope service.When realizing personalized search using above-mentioned iGraph service, the real-time individual character that search scene must be extracted is reappeared Change adjusting parameter, can be realized using the journal file for obtaining iGraph service;It is discussed in greater detail and is serviced using iGraph below The concrete scheme of personalized adjusting parameter is obtained when recording user characteristics.
As previously mentioned, executing search, during obtaining personalized search results, need to record initial search word string and The personalized adjusting parameter used to provide real time individual search to search main body.Record initial search is introduced individually below The method of word string and personalized adjusting parameter.
The initial search word string includes: search command and search content.
Described search order is the affairs clicking search button and completing;Described search content is determined search key Justice can be certain class special key words, be also possible to the keyword with qualifications;Such as: input " the even clothing in search box Skirt ";Or " long-sleeved dress ";Described search content can be also obtained by kind of Class Options, such as: " skirt is selected in type classification Son ", reselection " one-piece dress ".
The personalization adjusting parameter, refers to the parameter provided to provide personalized search to search main body;It include: Property search parameter and random algorithm parameter;The former according to search for personalized factor and the adjusting parameter that is arranged, the latter It is then to search for see same search result and the search result closer to ranking score is beaten every time in order to avoid user The scattered parameter used that sorts.
The personalized search parameter includes: screening search parameter and/or sorted search parameter.
The screening search parameter is limited the details of selected search content;Such as: price range, brand, material, spy Color etc.;The sorted search parameter is the restriction to described search content display mode;Such as: price height, sales volume height, Evaluation is high low.
During the search of above-mentioned realization real time individual, the entire chain for needing to record personalized search process is passed by Journey, for reappearing personalized search scene.It is i.e. described to execute search, during obtaining personalized search results, on the one hand It needs according to initial search word string and the personalized adjusting parameter provided to search main body, complete and searches for that scene is relevant searches Rope;On the other hand, in order to realize to search scene reproduction, it is also necessary to the real-time individual character of initial search word string and the secondary search Change adjusting parameter to be recorded.For this angle, above-mentioned search process can be summarized as follows:
Searching request is obtained, uniquely corresponding searching request identifies RN with the searching request for the generation of each searching request;
The initial search word string is extracted according to described search request, records the initial search word string;
Initial search result set is obtained according to the initial search word string, records the initial search result set;
According to the personalized adjusting parameter of acquisition, personalized search adjustment is carried out to the initial search result set, is obtained Obtain the personalized search results to search main display;
It during executing above-mentioned steps, is monitored, records the personalized situation of the query process.
In conjunction with the above process, the present embodiment provides a kind of schemes that this search process is recorded based on log system;It please join Fig. 2 is examined, is a kind of process of method for carrying out real time individual search and being recorded using log system provided by the present application Figure.It in following steps, illustrates to monitor simultaneously in conjunction with each step, the method for the personalized situation of record queries process.
Step S201: obtaining searching request, and for the generation of each searching request, uniquely corresponding search is asked with the searching request Seek mark RN.
This method relies on log system to the reproduction that real time individual is searched for, and the influence factor of personalized search is stored in day Aspiration system, inquiry log system, that is, reproducible search process, but should be noted that the related data in primary search may root It is respectively stored in different logs according to different parameter property;The distributed non-relational database of log system utilization (such as HBase database) storage search data;Unique searching request mark is for being associated with all dependency numbers once searched for According to more parts of daily record datas can be associated with according to " external key " RN of log system.
Specifically, user initiates a searching request, such as searches for " one-piece dress ";Front end services can be produced to when previous search Raw one uniquely corresponding searching request identifies RN with the searching request;This parameter is used to connect entirely from sending searching request The unique identification (for electric business, can be a commercial articles searching and strike a bargain to final) completed to final search process, meeting exists It is transmitted in entire search link, and as the foundation for reappearing search scene;In the present example it is assumed that the RN is to be specially searchID。
Step S202: the initial search word string is extracted according to described search request, records the initial search word string;By It is recorded in the present embodiment using log system, in this step, the initial search word string is written the front end services First log.
After search service provider receives search command, the key words content and search main body that include according to original word string Related data, which calculate, obtains search result;Meanwhile it parsing initial search word string and being stored in log system by preset format.
For example, server receives search access request s.m.taobao.com/search? key=Lian Yiqun &id= Xxx&searchId=%C1%AC%D2%C2%C8%B9;This section of character string is parsed, keyword key and search main body are obtained id;It by the stored corresponding characteristic of search main body id, calculates and obtains search result, while by keyword key's The corresponding search main body id of value is stored in database, and the value of searchId is stored in database according to id;Wherein searchId is exactly unique Searching request identify RN, encoded to obtain according to Session data, Cookie data and current time by browser and anti- It feeds server (it is possible only to enumerate a kind of realization here);Each value searched for according to searchId under same search main body id It distinguishes, is then associated with according to the value of searchId with primary search.
Step S203: initial search result set is obtained according to the initial search word string, records the initial ranging knot Fruit set.In this step, the search result identification list returned is written in the first log the front end services.
Search result is obtained by calculation in search service;In order to meet the correctness of reproducible results, searching for result will be obtained Log system is recorded in hitch fruit identification list.
For example, server receives search access request s.m.taobao.com/search? key=Lian Yiqun &id= Xxx&searchId=%C1%AC%D2%C2%C8%B9;The highest preceding n commodity of the score being calculated ResultId, for example, the commodity of front 500;The value of search result mark resultId is stored in database, according to searchId It is associated with other data of this search;Return to the search result list formed with described search result mark resuldId.
Storage includes search result list (being made of resultId) in first log, and search engine source excludes non-search Index is held up the problem of source causes and (is not handled in search log system if the result of promotion item).Wherein, described search The results list is as initial search result set.
Step S204: according to the personalized adjusting parameter of acquisition, personalization is carried out to the initial search result set and is searched Rope adjustment, obtains the personalized search results to search main display.
According to the operation (browsing, collection etc.) to search result, these operations are recorded by browser cookie, and accordingly Real-time adjusting parameter is obtained, and in triggering lookup or page turning by real-time adjusting parameter feedback to server;Server is providing When search result, the personalized search results of the current presentation page are calculated according to the real-time adjusting parameter, in general, these individual characteies Change search result to select to obtain in the initial search result set.
User can be written real-time system in the click behavior of every page and generate real-time adjusting parameter accordingly.For example, false If current page is second page, only it is only when the user characteristics that the correspondence second page stored in log system is written in time inquiry Accurate characteristic information.Can not be using real-time adjusting parameter real-time with the third page adjusting parameter of first page the reason of, is as follows: the The click behavior of one page is updated to after igraph database, in this way, before the search result that second page is provided, it can be according to The click behavior of first page, generates new real-time adjusting parameter, and the search result of the second page is according to the new real-time tune Whole gain of parameter, therefore, if directlying adopt the real-time adjusting parameter of first page, it will lead to data biased forwards;Likewise, It using the in due course adjusting parameter of third page, then can update because of the click behavior of second page to igraph, cause data to retrodeviating It moves.
Step S205: during executing above-mentioned steps, being monitored, and the personalized shape of the query process is recorded Condition.
Specifically, search service personalized adjusting parameter will be written in the second log in real time, and by personalized search Random algorithm parameter in the process is also written in the second log simultaneously.
During entire personalized search, according to the information that personalized search is fed back, search result is adjusted, and remember in real time Record personalized adjusting parameter (including personalized search parameter and random algorithm parameter).
For example, user selects one of them to check (as clicked image link) in search result, browser page Cookie will record this commodity id;Second page commodity (one-piece dress) list is checked in page turning, and server can repeat step S201, but The commodity id checked can be added in search word string at this time;Server is by being checked commodity corresponding characteristic (price, money Formula, material etc.) carry out analytical calculation, personalized search parameter is adjusted in real time, in conjunction with random algorithm parameter after, acquisition The personalized search results of second page;When recording the result, while recording personalized search parameter and random algorithm parameter.
In the detailed process of above-mentioned record log method, user initiates a searching request, such as searches for " one-piece dress ", front end Servicing wsearch can be to when previous search generates a unique identification RN parameter;This parameter is used to connect entirely from sending To search is finally completed, (for the search in shopping website, completion search can be client and issues the finger that places an order searching request Enable) unique identification, can entire link transmit;The searching request of user can pass through front end services and search service provider, Front end services will record the searching request URL of user, be written in the first log;Meanwhile front end services will record current page return Search result list, and be written the first log in;Search service provider can look into during executing entire inquiry operation Online data is ask, obtains the real time individual information of user, while personalized search parameter being written in the second log;Meanwhile Search service provider can record the random algorithm parameter of algorithm during executing algorithm groupware, and be recorded in the second log In.
Above step realizes a complete search link repeatedly, and is stored in log system.To which this searching structure need to be reappeared When, so that it may all related datas by searching for this search in log system are realized.
By the description of this step, the design of log system needs efficiently and rapidly to realize data storage and data search; So commodity data can store in the log of front end in actual design;Individuation data is stored in the day of search service provider In aspiration system.
From the foregoing, it will be observed that reappearing search scene is based on log system.Under the prior art, logged result collection is daily morning It produces offline, for the feedback of user, such response time is too long.It can also currently go inside journal file to carry out Inquiry, such efficiency are relatively low (finding corresponding log recording from 2 parts of journal files, then spliced manually).This reality In the technical solution that example offer is provided, log is subscribed to by TT, as long as there is log generation, so that it may it obtains at any time, and And every a log is recombinated by calculating service Galaxy in real time, it is parsed into the request string that can be used for back query. In the actual implementation process, for bulk delay probably at 1 minute or so, time loss is mainly in the generation for waiting log.Institute It states TT and refers to TT log center, be the Log Administration System of the acquisition for integrating log, transmission, storage, monitoring, pass through Log center, user can be convenient daily record data real-time storage to cloud.
The initial search word string by corresponding described search scene mutually merges with the personalized adjusting parameter, group At reproducible search word string, the personalization adjusting parameter is generally comprised: adjustment name field and corresponding adjustment value list.
It is described when reappearing described search scene, will corresponding described search scene initial search word string and the personalization The step of adjusting parameter mutually merges, and forms reproducible search word string, comprising:
The personalized adjusting parameter is merged, the initial search word string is added, forms the reproducible search word string.
Specifically, the message that original system subscribes to log center is gone back, log can be pushed to subscriber by TT, obtain described the One log and second log;2 parts of logs are carried out join connection by RN by system, recombinate out complete reproducible request word String, reduction request string format: s.m.taobao.com/search.json? q=Lian Yiqun &mock=algorithm factor: xxx; Nid:1,2,3,4,5,6,7;And obtain customized information: xxx&page=1.The page=1 indicates the reproducible search Word string corresponds to the first page of this search.
It is described that the personalized adjusting parameter is merged into the addition initial search word string, form the reproducible search word The step of string, comprising: the reproducible search word string is added in object search identification list (nid:1,2,3,4,5,6,7).
Step S102: search is executed according to the reproducible search word string, returns to the search scene of reproduction.
The search result details finally returned according to search service sp, so that it may the search scene of user at that time out is reappeared, The diagnosis investigation of problem on line.
Before the step of executing search according to the reproducible search word string, return to the search scene of reproduction, comprising: by pre- If format parses the reproducible search word string.
For example, reproducible search substring s.m.taobao.com/search.json? q=Lian Yiqun &mock=algorithm because Son: xxx;Nid:1,2,3,4,5,6,7;Customized information: xxx&page=1;Real time individual is transmitted by Json character string Search for situation;Searching request q is " one-piece dress ".
By this parameter of mock as identifying, needs the field of mock to parse therein these, work as search service Provider can parse content therein when getting this mock field.
It is such as resolved to algorithm factor, then search service provider would not visit again the model of algorithm factor, and it is direct It is calculated using this value that mock field provides;The principle of customized information is same.Mock be for it is certain cannot The interface directly used or service create virtual interface or service to simulate a kind of method of its function.
Before the step of executing search according to the reproducible search word string, return to the search scene of reproduction, by default lattice Formula parses the reproducible search word string, comprising:
Inquire the page turning search result identification list in the reproducible search word string, and to search main display The described search result of the corresponding page turning search result identification list is removed in current page.The search result mark of page turning Know and records needing this time reappeared search front page turning to show in list as a result, these results are needing that reappears originally to search Rope results page there can be no.It, can be to current after search service provider gets page turning search result identification list The search result filtering of page, removes shown search result, guarantees that shown search result does not repeat.It is needing to reappear Search scene in, these object searches mark listed in the search result list of page turning for re-searching for word string, generation Table need the search result of duplicate removal.
It is described that search is executed according to the reproducible search word string, return to the search scene of reproduction, comprising:
The search scene of the reproduction is saved in the form of snapshot.
By the description above to the replay method for searching for scene in a kind of real-time system of the application, solves real-time individual character Change the problem of search reappears;About this method, also some implementation detail problems are needed to pay attention to.
The performance main points of real time service, the real-time search behavior that the service of reproduction provides the whole network user are reappeared, and delay need to be in the second Grade hereinafter, for this purpose, can take some measures in the specific implementation.For example, being stored in storage using HBASE table, roekey It is distributed using hash;Field can compress storage, it is necessary to compression storing daily record data as far as possible, with a unicode between field Character segmentation removes unnecessary character;The data that amount associated by the RN next day volunteers are stored in same a line of HBASE table Different lines on, avoid being stored in two tables and do join operation again, concurrently read and write simultaneously influence to performance to reduce Deng.
After restoring method using above-mentioned real-time search scene, it can be realized using following manner to some specific implementation Search for the sharing of scene:
First way, be will correspond to described search scene re-search for word string be sent to by share people;Due to can Reappearing search word string is obtained after being merged by initial search word string with personalized adjusting parameter, and the reproducible search word is passed through String is obtained with all information of described search scene, and realizes the reduction to described search scene;
The second way is that the searching request mark for the described search scene that will need to share is sent to by sharing people;It is logical Crossing described search request mark can equally reappear in conjunction with the record of log system and identify associated search with the searching request Scene.
According to a kind of restoring method of scene of search in real time, the application provides a kind of reduction apparatus of scene of search in real time, Referring to FIG. 3, being the reduction apparatus schematic diagram for the real-time search scene that the application second embodiment provides.The description as described in device Fairly simple, details please refer to embodiment of the method.
A kind of reduction apparatus of real-time search scene, comprising:
Word string unit 302 is recombinated, the search scene reappeared for determining needs searches corresponding the original of described search scene Rope word string mutually merges with the personalized adjusting parameter, forms reproducible search word string.
Scene unit 303 is returned, for executing search according to the reproducible search word string, returns to the search field of reproduction Scape.
Optionally, including recording unit 301 during obtaining personalized search results, records former for executing search Begin search word string, and the personalized adjusting parameter provided to provide personalized search to search main body.
Optionally, including searching request mark unit is generated, for generating with the searching request only for each searching request One corresponding searching request mark;Described search request mark requests each data phase during entire survival with described search Association.
Optionally, the recording unit 301, comprising:
It generates searching request and identifies unit, for obtaining searching request, generated and the searching request for each searching request Unique corresponding searching request mark;Generation searching request mark unit can with as mentioned earlier, as a separate unit, It can also be used as a subelement of recording unit 301, both embodiments are all feasible;
Initial search word string unit is extracted, for extracting the initial search word string according to described search request, records institute State initial search word string;
Result unit is obtained, for obtaining initial search result set according to the initial search word string, is recorded described first Beginning search result set;
Personalized search unit is shown, for the personalized adjustment factor according to acquisition, to the initial search result collection It closes and carries out personalized search adjustment, obtain the personalized search results to search main display;
Monitoring unit, for being monitored, recording the personalization of the query process during executing above-mentioned steps Situation.
Optionally, the recording unit 301, for executing search, during obtaining personalized search results, record Initial search word string and the personalized adjusting parameter provided to provide personalized search to search main body;Simultaneously for recording institute State search result;Described search result includes the search result identification list that search engine is returned when executing search.
Optionally, the initial search word string includes: search command and search content.
Optionally, the personalized adjusting parameter includes: personalized search parameter and random algorithm parameter.
Optionally, the personalized search parameter includes: screening search parameter and/or sorted search parameter.
Optionally, described original search is added for merging the personalized adjusting parameter in the recombination word string unit 302 Rope word string forms the reproducible search word string.
Optionally, the reproducible search word string is added in search result identification list by the recombination word string unit 302.
Optionally, the recombination word string unit 302, when the search scene for determining that needs reappear, by corresponding described search field The initial search word string of scape mutually merges with personalized adjusting parameter, during forming reproducible search word string, by searching for institute It states log system data and forms the reproducible search word string, comprising:
Redisplaying information unit is obtained, for obtaining the search scene for needing to reappear, requests mark to obtain by described search Corresponding data in first log and the second log;
Form word string unit, data for will be obtained in first log and second log by preset format into Row recombination obtains reappearing search word string.
Optionally, comprising: parsing word string unit returns to weight for executing search according to the reproducible search word string Before the step of existing search scene, the reproducible search word string is parsed by preset format.
Optionally, described to parse the reproducible search word string by preset format, parsing result includes: initial search word String, the personalized adjusting parameter provided to provide personalized search to search main body, search result identification list.
Optionally, when the search scene for returning to the return reproduction of scene unit 303, the search scene of the reproduction is with fast According to form present.
Referring to FIG. 4, being the reduction system schematic for the real-time search scene that the application 3rd embodiment provides.System is real It applies example and is similar to embodiment of the method, therefore the description of system embodiment is simpler, specifically please refers to embodiment of the method.
A kind of playback system that scene is searched in real-time system, including mutually communicated by network implementations server 402, Client, database:
The server 402 is executing search, during obtaining personalized search results, record initial search word string and To provide personalized search to search main body and the personalized adjusting parameter that provides is to the database 403;
It is instructed when receiving the client 401, when reappearing described search scene, the server 402 will be in database 403 The initial search word string of corresponding described search scene mutually merges with the adjustment factor, forms reproducible search word string;
Search is executed according to the reproducible search word string, returns to the search scene of reproduction to the client 401.
Optionally, the server 402 is executing search, during obtaining personalized search results, records original search Rope word string and to provide personalized search to search main body and the personalized adjusting parameter that provides is to the database 403, wherein Record be based on log system, comprising:
Client initiates searching request, and server 402 obtains searching request, asks for the generation of each searching request with the search Seek unique corresponding searching request mark;
The initial search word string is extracted according to described search request, records the initial search word string, front end services will The first log is written in the initial value search word string;
Initial search result set is obtained according to the initial search word string, records the initial search result set, institute It states front end services the object search identification list returned is written in the first log;
According to the personalized adjusting parameter of acquisition, personalized search adjustment is carried out to the initial search result set, is obtained Obtain the personalized search results to search main display;
It during executing above-mentioned steps, is monitored, records the personalized situation of the query process, search service Personalized search parameter is written in the second log, and the second log is written into the random algorithm parameter during personalized search In.
Optionally, the server 402 is executing search, during obtaining personalized search results, while recording institute State search result;Described search result includes the search result identification list that search engine is returned when executing search.
Optionally, described when receiving the instruction of client 401, when reappearing described search scene, the server 402 will be described The initial search word string that described search scene is corresponded in database 403 mutually merges with the adjustment factor, forms reproducible Word string is searched for, the adjusting parameter is merged, the initial search word string is added, forms the reproducible search word string.
Optionally, described to instruct when receiving client 401, when reappearing described search scene, the server 402 is by data The initial search word string that described search scene is corresponded in library 403 mutually merges with the personalized adjusting parameter, and composition can weigh Now search for word string, comprising:
The server 402 obtains redisplaying information, requests mark to obtain first log and described by described search Corresponding data in second log;
The data of acquisition in first log and second log are recombinated by preset format, are reappeared Search for word string.
Optionally, the server 402 executes search according to the reproducible search word string, returns to the search scene of reproduction The step of before, comprising: parse the reproducible search word string by preset format.
Optionally, described that search is executed according to the reproducible search word string, return to the search scene of reproduction to client, Include:
The search scene of the reproduction is presented in the form of snapshot.
Referring to FIG. 5, being the reduction electronic equipment schematic diagram for the real-time search scene that the application fourth embodiment provides.Electricity Sub- apparatus embodiments are similar to embodiment of the method, therefore the description of electronic equipment embodiment is simpler, specifically please refer to method implementation Example.
The electronic equipment that search scene is reappeared in a kind of real-time system, the electronic equipment include:
Display 501;
Processor 502;
Memory 503, the search scene reappeared for determining needs, by the initial search word string of corresponding described search scene Mutually merge with personalized adjusting parameter, form reproducible search word string, the personalization adjusting parameter is to mention to search main body The parameter provided for personalized search;
Search is executed according to the reproducible search word string, returns to the search scene of reproduction.
The 5th embodiment of the application provides a kind of for searching for the storage equipment of scenario reduction in real time, storage equipment storage There is instruction, described instruction can be loaded by processor and execute following steps:
It determines the search scene for needing to reappear, the initial search word string of corresponding described search scene and personalized adjustment is joined Number mutually merges, and forms reproducible search word string, the personalization adjusting parameter be to provide personalized search to search main body and The parameter of offer;
Search is executed according to the reproducible search word string, returns to the search scene of reproduction.
Although the application is disclosed as above with preferred embodiment, it is not for limiting the application, any this field skill Art personnel are not departing from spirit and scope, can make possible variation and modification, therefore the guarantor of the application Shield range should be subject to the range that the claim of this application defined.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium Example.
1, computer-readable medium can be by any side including permanent and non-permanent, removable and non-removable media Method or technology realize that information stores.Information can be computer readable instructions, data structure, the module of program or other numbers According to.The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory techniques, CD-ROM are read-only Memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or Other magnetic storage devices or any other non-transmission medium, can be used for storage can be accessed by a computing device information.According to Herein defines, and computer-readable medium does not include non-temporary computer readable media (transitory media), such as modulates Data-signal and carrier wave.
2, it will be understood by those skilled in the art that embodiments herein can provide as the production of method, system or computer program Product.Therefore, complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects can be used in the application Form.It can be used moreover, the application can be used in the computer that one or more wherein includes computer usable program code The computer program product implemented on storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) Form.

Claims (29)

1. a kind of restoring method of the scene of search in real time characterized by comprising
The search scene for needing to reappear is determined, by the initial search word string of corresponding described search scene and personalized adjusting parameter phase Merge, forms reproducible search word string;
Search is executed according to the reproducible search word string, returns to the search scene of reproduction.
2. the restoring method of real-time search scene according to claim 1, which is characterized in that realize this method, holding In row search process, following preparation is carried out:
It is executing search, during obtaining personalized search results, is recording initial search word string, and providing for search scene a The personalized adjusting parameter of propertyization search and offer.
3. the restoring method of real-time search scene according to claim 2, which is characterized in that the record initial search word String and for search scene personalized search is provided and provide personalized adjusting parameter the step of before, execute following step:
For the generation of each searching request, uniquely corresponding searching request is identified with the searching request;Described search request mark and institute The each data stated during searching request entirely survives are associated.
4. the restoring method of real-time search scene according to claim 2, which is characterized in that it is described to execute search, it obtains During obtaining personalized search results, records initial search word string and provide personalized search to search main body and provide Personalized adjusting parameter, comprising:
Searching request is obtained, uniquely corresponding searching request identifies with the searching request for the generation of each searching request:
The initial search word string is extracted according to described search request, records the initial search word string;
Initial search result set is obtained according to the initial search word string, records the initial search result set;
According to the personalized adjusting parameter of acquisition, personalized search adjustment is carried out to the initial search result set, obtain to Search for the personalized search results of main display;
It during executing above-mentioned steps, is monitored, records the personalized situation of the query process.
5. the restoring method of real-time search scene according to claim 4, which is characterized in that it is described to execute search, it obtains During obtaining personalized search results, records initial search word string and provide personalized search to search main body and provide Personalized adjusting parameter, record therein are based on log system, comprising:
It is preceding in described the step of extracting the initial search word string according to described search request, record the initial search word string The first log is written in the initial search word string by end service;
It is described that initial search result set is obtained according to the initial search word string, record the step of the initial search result set In rapid, the search result identification list returned is written in the first log the front end services;
Described to be monitored, in the step of recording the personalized situation of the query process, search service joins personalized search In number the second log of write-in, and will be in random algorithm parameter the second log of write-in during personalized search.
6. the restoring method of real-time search scene according to claim 2, which is characterized in that it is described to execute search, it obtains During obtaining personalized search results, while recording described search result;Described search result includes searching for when executing search The search result identification list that engine is returned.
7. the restoring method of real-time search scene according to claim 4, which is characterized in that the initial search word string packet It includes: search command and search content.
8. the restoring method of real-time search scene according to claim 1, which is characterized in that the personalization adjusting parameter It include: personalized search parameter and random algorithm parameter.
9. the restoring method of real-time search scene according to claim 8, which is characterized in that the personalized search parameter It include: screening search parameter and/or sorted search parameter.
10. the restoring method of real-time search scene according to claim 1, which is characterized in that the determination needs to reappear Search scene, the initial search word string of corresponding described search scene is mutually merged with personalized adjusting parameter, composition is reproducible The step of searching for word string, comprising:
The personalized adjusting parameter is merged, the initial search word string is added, forms the reproducible search word string.
11. the restoring method of real-time search scene according to claim 10, which is characterized in that described by the personalization Adjusting parameter merges the step of initial search word string is added, forms the reproducible search word string, comprising: by search result The reproducible search word string is added in identification list.
12. the restoring method of real-time search scene according to claim 5, which is characterized in that the determination needs to reappear Search scene, the initial search word string of corresponding described search scene is mutually merged with personalized adjusting parameter, composition is reproducible The step of searching for word string forms the reproducible search word string by searching for the log system data, comprising:
The search scene for needing to reappear is obtained, requests mark to obtain first log and second log by described search In corresponding data;
The data obtained in first log and second log are recombinated by preset format, obtain reappearing search word String.
13. the restoring method of real-time search scene according to claim 1, which is characterized in that reproducible searched according to described Before the step of rope word string executes search, returns to the search scene of reproduction, comprising: parse the reproducible search by preset format Word string.
14. the restoring method of real-time search scene according to claim 13, which is characterized in that described to press preset format solution Analyse the reproducible search word string, parsing result includes: initial search word string, mentions to provide personalized search to search main body The personalized adjusting parameter of confession, object search identification list.
15. the restoring method of real-time search scene according to claim 1, which is characterized in that described to be weighed according to Now search word string executes search, returns to the search scene of reproduction, comprising:
The search scene of the reproduction is presented in the form of snapshot.
16. the restoring method of real-time search scene according to claim 1, which is characterized in that user is in the following way Share search scene:
The word string that re-searches for of corresponding described search scene is sent to by sharing people.
17. the restoring method of real-time search scene according to claim 3, which is characterized in that user is in the following way Share search scene: the searching request mark for needing the described search scene shared being sent to and is shared people.
18. a kind of reduction apparatus of the scene of search in real time characterized by comprising
Word string unit is recombinated, the search scene reappeared for determining needs, by the initial search word string of corresponding described search scene It adjusts parameter mutually to merge with personalization, forms reproducible search word string;
Scene unit is returned, for executing search according to the reproducible search word string, returns to the search scene of reproduction.
19. the reduction apparatus of real-time search scene according to claim 18 characterized by comprising
Recording unit during obtaining personalized search results, records initial search word string, He Weixiang for executing search Search for provides personalized search and the personalized adjusting parameter that provides.
20. the reduction apparatus of real-time search scene according to claim 18, which is characterized in that including generating searching request Unit is identified, uniquely corresponding searching request identifies with the searching request for generating for each searching request;Described search is asked Each data during asking mark to request entire survival with described search are associated.
21. the reduction apparatus of real-time search scene according to claim 19, which is characterized in that the recording unit, packet It includes:
It generates searching request and identifies unit, for obtaining searching request, generated for each searching request unique with the searching request Corresponding searching request mark;
Initial search word string unit is extracted, for extracting the initial search word string according to described search request, records the original Begin search word string;
Result unit is obtained, for obtaining initial search result set according to the initial search word string, records and described initially searches Rope results set;
Show personalized search unit, for the personalized adjustment factor according to acquisition, to the initial search result set into The adjustment of row personalized search, obtains the personalized search results to search main display;
Monitoring unit, for being monitored, recording the personalized shape of the query process during executing above-mentioned steps Condition.
22. a kind of scene of search in real time goes back original system, which is characterized in that including the service mutually communicated by network implementations Device, client, database;
The server is executing search, during obtaining personalized search results, records initial search word string and for searching Rope main body provides personalized search and the personalized adjusting parameter that provides is to database;
When receiving client instructions, when reappearing described search scene, server will correspond to described search scene in the database The initial search word string mutually merge with the personalized adjusting parameter, form reproducible search word string;
Search is executed according to the reproducible search word string, returns to the search scene of reproduction to client.
23. real-time search scene according to claim 22 goes back original system, which is characterized in that the server is executing Search during obtaining personalized search results, records initial search word string and to provide personalized search to search main body And the personalized adjusting parameter provided is to database, it is therein to record based on log system, comprising:
Client initiates searching request, and server obtains searching request, generates for each searching request unique with the searching request Corresponding searching request mark;
The initial search word string is extracted according to described search request, records the initial search word string, front end services will be described Initial value searches for word string and the first log is written;
Initial search result set is obtained according to the initial search word string, the initial search result set is recorded, before described The object search identification list returned is written in the first log for end service;
According to the personalized adjusting parameter of acquisition, personalized search adjustment is carried out to the initial search result set, obtain to Search for the personalized search results of main display;
It during executing above-mentioned steps, is monitored, records the personalized situation of the query process, search service will be a Property search parameter be written in the second log, and random algorithm parameter during personalized search is written in the second log.
24. real-time search scene according to claim 22 goes back original system, which is characterized in that the server is executing Search during obtaining personalized search results, while recording described search result;Described search result includes executing search When the search result identification list that is returned of search engine.
25. real-time search scene according to claim 22 goes back original system, which is characterized in that described when reception client Instruction, reappear described search scene when, server by correspond in database described search scene the initial search word string and The personalization adjusting parameter mutually merges, and forms reproducible search word string, comprising:
Server obtains redisplaying information, requests mark to obtain in first log and second log by described search Corresponding data;
The data of acquisition in first log and second log are recombinated by preset format, obtain described to weigh Now search for word string.
26. in real-time system according to claim 22 search for scene playback system, which is characterized in that server according to Before the step of reproducible search word string executes search, returns to the search scene of reproduction, comprising: parse institute by preset format State reproducible search word string.
27. real-time search scene according to claim 22 goes back original system, which is characterized in that described to be weighed according to Now search word string executes search, returns to the search scene of reproduction to client, comprising:
The search scene of the reproduction is presented in the form of snapshot.
28. a kind of electronic equipment of the scenario reduction of search in real time, which is characterized in that the electronic equipment includes:
Display;
Processor;
Memory, the search scene reappeared for determining needs, by the initial search word string and individual character of corresponding described search scene Change adjusting parameter mutually to merge, form reproducible search word string, the personalization adjusting parameter is to provide individual character to search main body The parameter changing search and providing;
Search is executed according to the reproducible search word string, returns to the search scene of reproduction.
29. a kind of for searching for the storage equipment of scenario reduction in real time, which is characterized in that be stored with instruction, described instruction can It is loaded by processor and executes following steps:
The search scene for needing to reappear is determined, by the initial search word string of corresponding described search scene and personalized adjusting parameter phase Merge, form reproducible search word string, the personalization adjusting parameter is provided to provide personalized search to search main body Parameter;
Search is executed according to the reproducible search word string, returns to the search scene of reproduction.
CN201710747385.1A 2017-08-25 2017-08-25 Method, device and system for restoring real-time search scene Active CN110020101B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710747385.1A CN110020101B (en) 2017-08-25 2017-08-25 Method, device and system for restoring real-time search scene

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710747385.1A CN110020101B (en) 2017-08-25 2017-08-25 Method, device and system for restoring real-time search scene

Publications (2)

Publication Number Publication Date
CN110020101A true CN110020101A (en) 2019-07-16
CN110020101B CN110020101B (en) 2023-09-12

Family

ID=67186150

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710747385.1A Active CN110020101B (en) 2017-08-25 2017-08-25 Method, device and system for restoring real-time search scene

Country Status (1)

Country Link
CN (1) CN110020101B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114445197A (en) * 2022-02-10 2022-05-06 北京沃东天骏信息技术有限公司 Commodity list page reproduction method and device, storage medium and electronic equipment

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005267053A (en) * 2004-03-17 2005-09-29 Nec Corp Speech search device, speech search server, speech search method, and speech search program
US20060218146A1 (en) * 2005-03-28 2006-09-28 Elan Bitan Interactive user-controlled relevance ranking of retrieved information in an information search system
CN102521251A (en) * 2011-11-16 2012-06-27 北京搜狗科技发展有限公司 Method for directly realizing personalized search, device for realizing method, and search server
US20140289323A1 (en) * 2011-10-14 2014-09-25 Cyber Ai Entertainment Inc. Knowledge-information-processing server system having image recognition system
CN104699725A (en) * 2013-12-10 2015-06-10 阿里巴巴集团控股有限公司 Data searching processing method and system
CN105808605A (en) * 2014-12-31 2016-07-27 北京奇虎科技有限公司 Search log combination method and system
CN106446132A (en) * 2016-09-19 2017-02-22 百度在线网络技术(北京)有限公司 Search processing method and device
US20170193056A1 (en) * 2014-06-27 2017-07-06 Sap Se Fuzzy substring search

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005267053A (en) * 2004-03-17 2005-09-29 Nec Corp Speech search device, speech search server, speech search method, and speech search program
US20060218146A1 (en) * 2005-03-28 2006-09-28 Elan Bitan Interactive user-controlled relevance ranking of retrieved information in an information search system
US20140289323A1 (en) * 2011-10-14 2014-09-25 Cyber Ai Entertainment Inc. Knowledge-information-processing server system having image recognition system
CN102521251A (en) * 2011-11-16 2012-06-27 北京搜狗科技发展有限公司 Method for directly realizing personalized search, device for realizing method, and search server
CN104699725A (en) * 2013-12-10 2015-06-10 阿里巴巴集团控股有限公司 Data searching processing method and system
US20170193056A1 (en) * 2014-06-27 2017-07-06 Sap Se Fuzzy substring search
CN105808605A (en) * 2014-12-31 2016-07-27 北京奇虎科技有限公司 Search log combination method and system
CN106446132A (en) * 2016-09-19 2017-02-22 百度在线网络技术(北京)有限公司 Search processing method and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
张晓飞;余建桥;: "基于用户兴趣模型构建与个性化搜索算法研究", 电脑知识与技术, no. 18 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114445197A (en) * 2022-02-10 2022-05-06 北京沃东天骏信息技术有限公司 Commodity list page reproduction method and device, storage medium and electronic equipment

Also Published As

Publication number Publication date
CN110020101B (en) 2023-09-12

Similar Documents

Publication Publication Date Title
US11349947B1 (en) Proxying hypertext transfer protocol (HTTP) requests for microservices
US10789626B2 (en) Deep-linking system, method and computer program product for online advertisement and e-commerce
US11803548B1 (en) Automated generation of metrics from log data
US20210075667A1 (en) Generating actionable alert messages for resolving incidents in an information technology environment
US11829330B2 (en) Log data extraction from data chunks of an isolated execution environment
US11232125B1 (en) Conversion of cloud computing platform data for ingestion by data intake and query system
US11775501B2 (en) Trace and span sampling and analysis for instrumented software
US11288319B1 (en) Generating trending natural language request recommendations
US11886430B1 (en) Intent-based natural language processing system
JP5721818B2 (en) Use of model information group in search
US11657582B1 (en) Precise plane detection and placement of virtual objects in an augmented reality environment
US20190238635A1 (en) Method and system for synchronizing webpages within a web browser
JP2012510128A (en) Image search apparatus and method
US11663172B2 (en) Cascading payload replication
US11755531B1 (en) System and method for storage of data utilizing a persistent queue
US10839431B1 (en) Systems, methods and computer program products for cross-marketing related products and services based on machine learning algorithms involving field identifier level adjacencies
US20220191105A1 (en) Generating a three-dimensional cityscape including a cluster of nodes
US11875275B1 (en) Custom time series models in computer analytics systems
US20230123539A1 (en) Stitching event data using identity mappings
JP2011526026A (en) Method and apparatus for providing data statistics
US9734529B1 (en) Method, system, and computer readable medium for selection of catalog items for inclusion on a network page
CN110020101A (en) The restoring method of search scene, device and system in real time
JP3525885B2 (en) Multidirectional search service method and recording medium recording the program
US11113301B1 (en) Generating metadata for events based on parsed location information of data chunks of an isolated execution environment
US10839430B1 (en) Systems, methods and computer program products for populating field identifiers from telephonic or electronic automated conversation, generating or modifying elements of telephonic or electronic automated conversation based on values from field identifiers

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
TA01 Transfer of patent application right

Effective date of registration: 20230630

Address after: Room 554, floor 5, building 3, No. 969, Wenyi West Road, Wuchang Street, Yuhang District, Hangzhou City, Zhejiang Province

Applicant after: TAOBAO (CHINA) SOFTWARE CO.,LTD.

Address before: Box 847, four, Grand Cayman capital, Cayman Islands, UK

Applicant before: ALIBABA GROUP HOLDING Ltd.

TA01 Transfer of patent application right
GR01 Patent grant
GR01 Patent grant