CN105654341A - Aviation product recommendation system and aviation product recommendation method based on cloud service - Google Patents
Aviation product recommendation system and aviation product recommendation method based on cloud service Download PDFInfo
- Publication number
- CN105654341A CN105654341A CN201511000479.XA CN201511000479A CN105654341A CN 105654341 A CN105654341 A CN 105654341A CN 201511000479 A CN201511000479 A CN 201511000479A CN 105654341 A CN105654341 A CN 105654341A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- recommendation
- product
- aeronautical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0282—Rating or review of business operators or products
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
An embodiment of the invention discloses an aviation product recommendation method based on cloud service. The aviation product recommendation method comprises the steps of receiving an information service request by a cloud server, wherein the information service request is transmitted by a client, obtaining aviation product data and user behavior data; according to the aviation product data and the user behavior data, calculating recommendation sets which correspond with different aviation products and storing the products into a memory database, wherein each recommendation set comprises a multidimensional product recommendation result of one aviation product; receiving a recommendation service request which is transmitted from the client, and calculating a user recommendation result and returning the user recommendation result to the client based on the recommendation sets in the memory database. The embodiment of the invention further discloses an aviation product recommendation system based on cloud service. The aviation product recommendation system effectively settles problems of unsuitability of product characteristics extracted in traditional electronic commerce to technical field of aviation and single user mode in traditional electronic commerce.
Description
Technical field
The present invention relates to data processing technique, relate in particular to a kind of aeronautical product commending system based on cloud serviceAnd method.
Background technology
In recent years along with the development of ecommerce and going deep into of airline's integration operation strategy, airlineNo longer be confined to the development of traditional acknowledgement of consignment business aspect, simultaneously also actively by means of electric business's platform by air ticket and wineThe product package sale such as shop, sight spot, attracts passenger, widens realm of sale. Along with airline's product categoryWith enriching constantly of content, how product is pushed to the passenger who needs it accurately, just become aviation public affairsDepartment's problem demanding prompt solution. This precision marketing method is referred to as again personalized recommendation, current main flowWay is, browses, buys and user model is set up in the behavior such as collection by all users in collecting platform, thenUsing personalized recommendation algorithm by user model and product feature, is platform user recommended products, thereby realThe target of existing precision marketing.
Build above-mentioned platform and mainly contain following two kinds of ways, one, using the Cloud Server of electric business website as basePlinth, builds electronic commerce transaction system, attracts supplier or third party seller to move into, thereby is integrated into oneLarge-scale electric business's transaction platform; Its two, using the Cloud Server of the third party service provider as basis, by cooperationMode is collected product information and the user behavior of electric business website, thereby is integrated into a large-scale service platform. RightPersonalized recommendation under civil aviaton's background, mainly there are following two problems in above-mentioned technology:
Electricity business's user model builds single: only collect user in electric business website and browse, buy and collection etc.Behavior, and build user model with this behavior. The user model building by this kind of mode, only can characterize useThe Partial Feature at family, not comprehensive to passenger's attribute description, fail to contain the endemic genus of passenger in civil aviaton fieldProperty. For example: passenger's travel time information and passenger's property of value, just can not use electric business's user modelDescribe.
The product feature of extracting is inapplicable: in electric business website, the product information of extraction mainly comprises: category,The feature such as product category and price, but the package sale situation of aeronautical product is comparatively complicated, traditional productFeature extraction can not meet existing requirement. For example: the relevant information in airline and course line, just can not useGeneralization method is extracted.
Therefore, a kind of new departure need to be proposed, taking the passenger information of cloud server and civil aviaton's product information asBasis, realizes the personalized recommendation that is applicable to civil aviaton field.
Summary of the invention
For solving the technical problem of existing existence, the embodiment of the present invention provides a kind of aviation based on cloud service to produceProduct commending system and method.
For achieving the above object, the technical scheme of the embodiment of the present invention is achieved in that
An aeronautical product recommend method based on cloud service, described method comprises:
Cloud server receives the information service request that client sends, and obtains aeronautical product data and user's rowFor data;
According to aeronautical product data and user behavior data, calculate the recommendation set of corresponding different aeronautical productsAnd being saved in memory database, the Products Show of the multiple dimensions of aeronautical product of described recommendation set-inclusion is tiedReally;
Receive the recommendation service request that client sends, the recommendation set based in described memory database, meterCalculate user's recommendation results and return to client.
Wherein, described method also comprises: client gathers aeronautical product data and user behavior data, sendsThe information service request that contains described aeronautical product data and described user behavior data is to cloud server; With/Or client sends the information service request that the unique high in the clouds of user mark is set to described cloud server; With/Or client sends the recommendation service request that is used for obtaining user's recommendation results to described cloud server; ItsIn, described information service request and recommendation service request comprise respectively the function point mark for identification request typeKnow number and the parameter relevant to self.
Wherein, described method also comprises:
Receive the information service request that the unique high in the clouds of user mark is set that client sends, described information serviceRequest comprises User IP and user cookie;
According to described information service request, judge whether user exists;
If user exists, the unique high in the clouds of corresponding inquiry user is identified and returns to described client;
If user does not exist, for user arranges the unique high in the clouds of new user mark, and by unique this userHigh in the clouds mark returns to described client;
Wherein, the unique high in the clouds of described user mark is for mark user's inter-network behavior.
Wherein, described method also comprises: user's recommendation results that described in client, cloud server returnsAnd show user.
Wherein, obtain after aeronautical product data and user behavior data, by described aeronautical product data and userBehavioral data deposits message queue in; From described message queue, obtain data and deposit respectively distributed file system inAnd memory database; Aeronautical product data in described distributed file system and user behavior data are carried outClean and processing; The passenger who prestores in distributed file system is drawn a portrait to data and deposit memory database in.
Wherein, according to aeronautical product data and user behavior data, calculate corresponding different aeronautical productRecommendation set and deposit memory database in, for: with the aeronautical product data of described distributed file system andUser behavior data is basis, uses proposed algorithm to calculate the Products Show result of different aeronautical product various dimensions,Obtain the recommendation set of different aeronautical products, deposit described recommendation set and aeronautical product data in internal storage dataStorehouse.
Wherein, described calculating user recommendation results, comprising: judge according to recommendation service request whether userLog in, if user logs in, based on the unique high in the clouds of described user mark identification passenger portrait, and by rightAeronautical product data that should described recommendation service request and active user's passenger draw a portrait data mates, rightIn the recommendation set of aeronautical product, each Products Show result is given a mark, and determines to be recommended according to marking resultProducts Show result, form described user's recommendation results by Products Show result to be recommended.
Wherein: if user logs in, the passenger of active user in described memory database is drawn a portrait to dataMiddle airline, the line of flight and consuming capacity, with airline, the target pattern in aeronautical product dataMate with product price, and from big to small each is recommended to the product in set according to the score of matching resultProduct recommendation results sorts, and is tied by the Products Show of score the highest, two or more aeronautical productsFruit obtains user's recommendation results.
Wherein, described method also comprises: if user does not log in, based on current recommendation service request Air ChinaRecommendation set corresponding to empty product, and choose at random from fast-selling ranking list, obtaining new recommendation set, this pushes awayThe number of recommending the aeronautical product of recommending in set is not less than predetermined threshold value; According to the aviation in aeronautical product dataCompany, target pattern and product price are given a mark, and according to score from big to small in new recommendation setEach recommendation results sorts, by the Products Show of score the highest, two or more aeronautical productsResult obtains user's recommendation results; Wherein, described fast-selling ranking list is according to described aeronautical product data and userBehavioral data calculates, and includes the related data of fast-selling aeronautical product.
Wherein, described user behavior data comprises: web site name, user cookie, user session, productProduct title and user are at the behavioral data of the enterprising line operate in website;
Described aeronautical product data comprise: web site name, product category, product subdivision, name of product, boatEmpty company, target pattern and product price.
An aeronautical product commending system based on cloud service, described system comprises cloud server, described cloudEnd server comprises recommendation service engine modules and proposed algorithm module;
Described recommendation service engine modules, the information service request sending for receiving client, obtains aviationProduct data and user behavior data; And, the recommendation service request sending for receiving client, based onRecommendation set in memory database, calculates user's recommendation results and returns to client;
Described proposed algorithm module, for according to described aeronautical product data and user behavior data, calculatesThe recommendation set of corresponding different aeronautical products is also saved in described memory database, described recommendation set-inclusion oneThe Products Show result of the multiple dimensions of individual aeronautical product.
Wherein, described recommendation service engine modules, also for:
Receive the information service request that the unique high in the clouds of user mark is set that client sends, described information serviceRequest comprises User IP and user cookie;
According to described information service request, judge whether user exists;
If user exists, the unique high in the clouds of corresponding inquiry user is identified and returns to described client;
If user does not exist, for user arranges the unique high in the clouds of new user mark, and by unique this userHigh in the clouds mark returns to described client;
Wherein, the unique high in the clouds of described user mark is for mark user's inter-network behavior.
Wherein, described cloud server also comprises: real-time message processing module and data cleansing module; Wherein,
Real-time message processing module, the aeronautical product passing over for receiving described recommendation service engine modulesData and user behavior data, deposit described aeronautical product data and user behavior data in message queue; FromIn described message queue, obtain data and deposit respectively distributed file system and memory database in;
Data cleansing module, for aeronautical product data and user behavior in described distributed file systemData are cleaned and are processed; The passenger who prestores in distributed file system is drawn a portrait to data and deposit internal storage data inStorehouse.
Wherein, described proposed algorithm module, for: with the aeronautical product data of described distributed file systemWith user behavior data be basis, use proposed algorithm to calculate the Products Show knot of different aeronautical product various dimensionsReally, obtain the recommendation set of different aeronautical products, deposit described recommendation set and aeronautical product data in internal memoryDatabase.
Wherein, described proposed algorithm module, for: judge according to recommendation service request whether user logs in,If user logs in, based on the unique high in the clouds of described user mark identification passenger portrait, and by described in correspondenceThe aeronautical product data of recommendation service request and active user's passenger draw a portrait data mates, based on logging inUser's difference each Products Show result in the recommendation set of aeronautical product is given a mark, according to marking knotFruit is determined Products Show result to be recommended, forms described user recommend knot by Products Show result to be recommendedReally.
Wherein, described proposed algorithm module, for: if user logs in, will in memory database, work asFront user's passenger draws a portrait airline in data, the line of flight and consuming capacity, in aeronautical product dataAirline, target pattern and product price mate, and according to the score of matching result from big to smallRecommend the Products Show result in set to sort to each, by score the highest, two or moreThe Products Show result of aeronautical product obtains user's recommendation results.
Wherein, described proposed algorithm module, also for: if user does not log in, based on current recommendation clothesRecommendation set corresponding to aeronautical product in business request, and choose at random from fast-selling ranking list, new recommendation obtainedSet, in this recommendation set, the number of the aeronautical product of recommending is not less than predetermined threshold value; According to aeronautical product numberAccording in airline, target pattern and product price give a mark, and according to score from big to small to newRecommend each recommendation results in set to sort, by score the highest one, two or more aeronautical productProducts Show result obtain user's recommendation results; Wherein, described fast-selling ranking list is by described proposed algorithm mouldPiece calculates and deposits in memory database according to described aeronautical product data and user behavior data, includesThe related data of fast-selling aeronautical product.
Wherein, described system also comprises: client, described client comprises service request module, for adoptingCollection aeronautical product data and user behavior data, send and contain described aeronautical product data and described user behaviorThe information service request of data is to cloud server; And/or client sends the unique high in the clouds of user mark is setInformation service request give described cloud server; And/or client sends and is used for obtaining user's recommendation resultsRecommendation service request give described cloud server; Wherein, described information service request and recommendation service requestComprise respectively function point identification number and the parameter relevant to self for identification request type; And, forReceive user's recommendation results that described recommendation service engine modules returns and show user.
Wherein, described service request module is deployed in the client of the electricity business of airline website.
Wherein, described user behavior data comprises: web site name, user cookie, user session, productProduct title and user are at the behavioral data of the enterprising line operate in website; Described aeronautical product data comprise: website nameTitle, product category, product subdivision, name of product, airline, target pattern and product price.
Aeronautical product commending system and the method for the embodiment of the present invention based on cloud service, for user arranges uniqueHigh in the clouds mark, captures aeronautical product data and user behavior data, can not affect the electronics business of airlineUnder the prerequisite of the normal operation in business website, obtain data, effectively solved conditional electronic commercial affairs and extracted product featureThe problem in inapplicable aeronautical technology field, and can all sidedly website user and passenger's portrait be merged,Utilize data passenger's attribute to be carried out to the description of various dimensions, and carry out product in conjunction with the feature of aeronautical product and push awayRecommend, effectively solved the single problem of conditional electronic business users model.
Brief description of the drawings
In accompanying drawing (it is not necessarily drawn in proportion), similar Reference numeral can be at different viewsThe parts that middle description is similar. The similar Reference numeral with different letter suffix can represent the difference of similar partsExample. With example, unrestriced mode shows each embodiment discussed herein to accompanying drawing substantially.
Fig. 1 is the flow chart of the aeronautical product recommend method of the embodiment of the present invention based on cloud service;
Fig. 2 is the specific implementation flow chart of the aeronautical product recommend method of the embodiment of the present invention based on cloud service;
Fig. 3 is the composition knot of the aeronautical product commending system cloud server of the embodiment of the present invention based on cloud serviceStructure schematic diagram;
Fig. 4 is the composition structural representation of the aeronautical product commending system of the embodiment of the present invention based on cloud service;
Fig. 5 is the execution schematic flow sheet of embodiment of the present invention service request module;
Fig. 6 is the execution schematic flow sheet of embodiment of the present invention recommendation service engine modules;
Fig. 7 is the execution schematic flow sheet of embodiment of the present invention real-time message processing module;
Fig. 8 is the execution schematic flow sheet of embodiment of the present invention data cleansing module;
Fig. 9 is the execution schematic flow sheet of embodiment of the present invention proposed algorithm module.
Detailed description of the invention
Embodiment mono-
The aeronautical product recommend method of the embodiment of the present invention based on cloud service, as shown in Figure 1, described method bagDraw together:
Step 101: cloud server receives the information service request that client sends, and obtains aeronautical product numberAccording to and user behavior data;
Wherein, described user behavior data comprises: web site name, user cookie, user session, productProduct title and user are at the behavioral data of the enterprising line operate in website; Described aeronautical product data comprise: website nameTitle, product category, product subdivision, name of product, airline, target pattern and product price.
Step 102: according to aeronautical product data and user behavior data, calculate corresponding different aeronautical productRecommendation set and be saved in described memory database, multiple dimensions of aeronautical product of described recommendation set-inclusionThe Products Show result of degree;
Step 103: receive the recommendation service request that client sends, based on pushing away in described memory databaseRecommend set, calculate user's recommendation results and be pushed to client.
Wherein, obtain after aeronautical product data and user behavior data, by described aeronautical product data and userBehavioral data deposits message queue in; From described message queue, obtain data and deposit respectively distributed file system inAnd memory database.
Here also comprise: to aeronautical product data and user behavior data in described distributed file systemClean and process; The passenger who prestores in distributed file system is drawn a portrait to data and deposit memory database in.
Wherein, in step 102, with aeronautical product data and the user behavior number of described distributed file systemAccording to being basis, use proposed algorithm to calculate the recommendation results of different aeronautical product various dimensions, obtain different aviationsThe recommendation set of product, deposits described recommendation set and aeronautical product data in memory database.
Particularly, in step 103, described by the aeronautical product data of described correspondence recommendation service request with work asFront user's passenger draws a portrait data mates, the recommendation set based in described memory database and fast-selling rowRow list, calculating active user's recommendation results, comprising: according to described recommendation service request, judge that user isNo logging in; If user does not log in, active user's passenger draws a portrait data and does not exist, directly from described heatIn pin ranking list, choose at random the recommendation results of aeronautical product as active user; If user logs in,The aeronautical product data of described correspondence recommendation service request and active user's passenger is drawn a portrait to data to carry outJoin, from memory database, each recommends to extract in set the recommendation results of different aeronautical products, obtains currentUser's recommendation results.
Embodiment bis-
The specific implementation process of the aeronautical product recommend method to based on cloud service is done detailed theory by the present embodimentBright.
As shown in Figure 2, described method comprises:
Step 201: the information service that client arranges the unique high in the clouds of user mark to cloud server transmission is askedAsk;
Wherein, the unique high in the clouds of user mark is for mark user's inter-network behavior.
Step 202: cloud server receives described information service request, arranges the unique high in the clouds of user and identifies alsoReturn to described client;
Here, first cloud server verifies the unique high in the clouds of the user mark that whether has active user, ifExist, the unique high in the clouds of active user's user mark returns to client, if there is no, is this useFamily arranges the unique high in the clouds of new user mark, and above-mentioned judgement state and the unique high in the clouds of user mark are returned toClient.
Step 203: client gathers aeronautical product data and user behavior data, and sends out to cloud serverCarry information service request;
Here, client sends the information clothes that include described aeronautical product data and described user behavior dataBusiness request is to cloud server.
Here, client gathers after aeronautical product data, also carries out label for different web sites and different productConversion. The user behavior data gathering relates to the related data of the behaviors such as browsing, buying of user and collection.
Step 204: cloud server receives described information service request, to described user behavior data and boatEmpty product data are carried out data transaction, and deposit described aeronautical product data and user behavior data in message teamRow;
Step 205: obtain data and deposit respectively distributed file system and interior poke in from described message queueAccording to storehouse;
Step 206: the aeronautical product data in described distributed file system and user behavior data are carried outClean and processing, the passenger who prestores is drawn a portrait to data deposit memory database in distributed file system;
Wherein, passenger portrait be a kind of that generate as basis taking the historical log of passenger, can characterize passenger spyHave and the label system of dynamic attribute, comprise ID, airline, the line of flight and consuming capacity etc.Label.
Here, passenger to draw a portrait data and user behavior data be all with key-value form in memory databaseStorage, its key is the unique high in the clouds of user mark, value is corresponding data content. User behavior dataFor calculated recommendation results set, passenger draws a portrait data for process follow-up and recommendation results sets matchIn.
Here, data cleansing is data and the abnormal data etc. that filtering can not normally be resolved, and data processing comprisesThe data of cleaning are processed to processing.
Step 207: according to aeronautical product data and user behavior data, calculate corresponding different aeronautical productRecommendation set and fast-selling ranking list, and be saved in described memory database, one of described recommendation set-inclusionThe Products Show result of the multiple dimensions of aeronautical product;
Wherein, described fast-selling ranking list includes the related data of fast-selling aeronautical product.
Step 208: client is asked according to user, sends recommendation service request to cloud server;
Step 209: cloud server receives the recommendation service request that client sends, based on described interior pokeAccording to the recommendation set in storehouse and fast-selling ranking list, calculate user's recommendation results and return to client;
Here identify to identify passenger's portrait based on the unique high in the clouds of user, by described correspondence recommendation service request,Aeronautical product data and active user's passenger draw a portrait data and mate, user based on logging in different comeEach Products Show result in the recommendation set of aeronautical product is given a mark, determine and recommend according to marking resultThe Products Show result of which dimension of some or several aeronautical products, and form user's recommendation results with this.
In practical application, recommend set to comprise that multiple aeronautical product, each aeronautical product have again three attributes:Airline, target pattern and product price, by three attribute (aviations of these three attributes and passenger's portraitCompany, the line of flight and consuming capacity) carry out cosine method of average coupling, can obtain the score between 0 to 1,According to score from height to low arrangement Products Show result. Therefore, can think that matching result is exactly Products Show knotReally, also can think that matching result adds that aeronautical product data form Products Show result.
Step 210: user's recommendation results that described in client, cloud server returns also shows user.
Embodiment tri-
As shown in Figure 3, the aeronautical product commending system of the embodiment of the present invention based on cloud service, comprises high in the clouds clothesBusiness device, this cloud server comprises recommendation service engine modules and proposed algorithm module, recommendation service engine mouldPiece, for carrying out the method shown in embodiment mono-Fig. 1, repeats no more.
In addition,, cloud server also comprises: real-time message processing module, data cleansing module andProposed algorithm module. Particularly,
Recommendation service engine modules, the information service request sending for receiving client, obtains aeronautical productData and user behavior data, and described aeronautical product data and user behavior data are passed to real-time messagesProcessing module; And the recommendation service request sending for receiving client, based on described memory databaseIn recommendation set and fast-selling ranking list, calculate user's recommendation results and also return to client.
Proposed algorithm module, for according to aeronautical product data and user behavior data, calculates corresponding differentThe recommendation set of aeronautical product and include the fast-selling ranking list of at present the most fast-selling aeronautical product relevant information,And being saved in described memory database, the product of the multiple dimensions of aeronautical product of described recommendation set-inclusion pushes awayRecommend result.
Real-time message processing module, for depositing described aeronautical product data and user behavior data in message teamRow obtain data and deposit respectively distributed file system and memory database in from described message queue.
Data cleansing module, for aeronautical product data and user behavior in described distributed file systemData are cleaned and are processed, and the passenger who prestores in distributed file system is drawn a portrait to data deposit interior poke inAccording to storehouse, during for described proposed algorithm module and the calculating of recommendation service engine modules.
Here, recommendation service engine modules, also for receiving the unique high in the clouds of user mark that client sendsInformation service request, described information service request comprises User IP and user cookie; According to described information clothesBusiness request, judges whether user exists; If user exists, will inquire about the unique high in the clouds of corresponding user markAnd return to described client; If user does not exist, for user arranges the unique high in the clouds of new user mark,And unique this user high in the clouds mark is returned to described client.
Embodiment tetra-
The aeronautical product commending system of the embodiment of the present invention based on cloud service, system architecture diagram as shown in Figure 4,Comprise client and cloud server, specifically with lower module: service request module, recommendation service engine modules,Real-time message processing module, data cleansing module and proposed algorithm module, wherein, service request module is disposedIn the client of the electricity business of airline website, recommendation service engine modules, real-time message processing module, dataCleaning module and proposed algorithm module are deployed in cloud server, and multiple service request modules can while and high in the cloudsService end is carried out alternately.
Wherein, each module general function is described as follows:
Service request module, for sending the information service request that the unique high in the clouds mark of user is set to recommendationService-Engine module; And, for gathering aeronautical product data and user behavior data, and send correspondingInformation service request is to recommendation service engine modules; And, for sending recommendation service request to recommendation serviceEngine modules, receives and shows returning results of described recommendation service engine modules. Here service request mould,Piece gathers aeronautical product data, for different web sites and different product, it is carried out to label conversion, and by aviationProduct data pass to recommendation service engine modules; And service request module gathers user and browses, buysWith relevant user behavior datas such as collections, user behavior data is passed to recommendation service engine modules.
Recommendation service engine modules, for the information service request sending based on service request module, establishes in real timePut the unique high in the clouds of user mark; And, for the information service request sending according to service request module, obtainGet aeronautical product data and user behavior data, and data are passed to real-time message queue module; And,Be used for responding recommendation service request, calculate in real time user's recommendation results and return to service request module.
Real-time message processing module, the aeronautical product data of transmitting for receiving described recommendation service engine modulesAnd user behavior data, deposit data in message queue, and obtain data from message queue, deposit in respectivelyDistributed file system and memory database.
Data cleansing module, for aeronautical product data and user behavior data in distributed file systemClean and process, and the passenger who prestores in distributed file system is drawn a portrait to data deposit memory database in.
Proposed algorithm module, for the aeronautical product data taking distributed file system and user behavior data asBasis, uses proposed algorithm to calculate the various dimensions Products Show result based on aeronautical product, obtains each aviationThe recommendation set of product, and deposit recommendation set and aeronautical product data in memory database.
Wherein, proposed algorithm module is the Products Show result that corresponding each aeronautical product obtains multiple dimensions,Recommendation service engine modules is user based on the logging in product to each dimension in the recommendation set of different aeronautical productsProduct recommendation results is given a mark, and determines which dimension of recommending some or several aeronautical products according to marking resultThe Products Show result of degree, and form user's recommendation results with this and be pushed to again user.
In practical application, recommend set to comprise that multiple aeronautical product, each aeronautical product have again three attributes:Airline, target pattern and product price, by three attribute (aviations of these three attributes and passenger's portraitCompany, the line of flight and consuming capacity) carry out cosine method of average coupling, can obtain the score between 0 to 1,According to score from height to low arrangement Products Show result. Therefore, can think that matching result is exactly Products Show knotReally, also can think that matching result adds that aeronautical product data form Products Show result.
Particularly, as shown in Figure 5, the execution flow process of service request module comprises:
Step 501: receive user's request;
Step 502: package request information;
Concrete, receive user's request, according to user's request type, the page elements on electric business website is enteredLine parameter conversion, and encapsulate different HTTP solicited messages, every HTTP solicited message comprise askThe contents such as the identification number of type and the desired different parameters of each request type, sending to cloud server pleaseAsk.
Step 503: send request to cloud server;
Here, ask to comprise two types of information service request and recommendation service requests, information service request alsoThe information service request being divided into for the unique high in the clouds of user mark is set please with the information service of uploading related dataAsk. Each request comprises function point identification number and parameter, and wherein, function point identification number can be according to actual needArrange, for identifying the type of current request.
For example, the information service request that the unique high in the clouds mark of user is set comprises: function point identification number is 1,Parameter is User IP and user cookie etc.
For example, the information service request of uploading aeronautical product data comprises, function point identification number is 2, parameterFor the title that browses web sites, product category, product subdivision, name of product, airline, route information and valencyLattice etc.
For example, the information service request of uploading user behavior comprises: function point identification number is 3, and parameter is clearLook at web site name, user cookie, user session, name of product and user behavior and (as the row such as see, buyFor) etc.
For example, obtain current aeronautical product and " seen and seen " that the recommendation service request on recommendation hurdle comprises function point markKnowledge number is 11, and parameter is the title that browses web sites, product category, product subdivision, name of product and IDDeng.
Step 504: receive the reply of cloud server;
Here, if the request sending is the information service request that the unique high in the clouds of user mark is set, receive,To return information comprise the unique high in the clouds of user mark, obtain the unique high in the clouds of user by resolving this return informationMark, the setting of the unique high in the clouds of completing user mark.
Here,, if the request sending is recommendation service request, the return information receiving comprises user to be pushed awayRecommend result, obtain user's recommendation results by resolving this return information.
Step 505: show that user's recommendation results is to user.
The execution flow process of recommendation service engine modules as shown in Figure 6, comprises:
Analysis service request, the request type identification number sending according to service request module is resolved;
Jump to respectively information service request branch or recommendation service request divides according to request type identification number; Here, if request type identification number mistake is returned to error message (not shown in Fig. 6);
For information service request branch, first judge request type and parameter, if request type or parameterMistake, returns to error message, if request type and parameter are correct, enters respectively product information placeReason branch, user behavior process branch and user is mated branch;
Process branch in product information, the aeronautical product data of obtaining are carried out to format conversion, and write serviceDevice daily record;
Process branch at user behavior, the user behavior data obtaining is carried out to format conversion, and write serviceDevice daily record;
Mate branch user, judge whether the unique high in the clouds of user mark exists (i.e. judgement in serverWhether be same people), if existed, think same person (user), if there is no, be this useFamily arranges the unique high in the clouds of new user mark, and above-mentioned judgement state and the unique high in the clouds of user mark are returned.
For recommendation service request branch, first judge request type and parameter, if request type or parameterMistake, returns to error message, if request type and parameter are correct, enters various recommendations hurdle branch.In every kind of recommendation hurdle branch, judge according to required parameter whether user logs in, if user does not log in,Recommendation set based on aeronautical product in current request, and choosing at random from the fast-selling ranking list of memory databaseGet Products Show result; If user logs in,, according to recommending hurdle type, from memory database, obtainCorresponding recommendation results collection (being recommendation set mentioned above); Obtained Products Show result is carried out to intelligenceCan sort, obtain user's recommendation results, user's recommendation results is returned to service request module. Recommending clothesWhen request in business, can comprise the parameter of recommending hurdle type, for example, recommends hurdle type can comprise three kinds: seen andSee, seen final purchase and guessed that you like. Accordingly, recommend the product of set also to comprise these three kinds recommendation classesType.
Here, no matter the Products Show result of choosing at random or the specific products obtaining from internal storage data are recommendedAs a result, all to carry out intelligent sequencing to recommendation results. Particularly, draw a portrait according to passenger in memory databaseThe labels such as airline, the line of flight and consuming capacity, with airline, the target in aeronautical product dataCourse line and product price are mated, and according to the score of matching result from big to small to the Products Show obtainingResult sorts, and is formed by the Products Show result of score the highest, two or more aeronautical productsUser's recommendation results is also returned; For the Products Show result of choosing at random, according in aeronautical product dataAirline, target pattern and product price are given a mark, and according to score from big to small to the product of obtainingRecommendation results sorts, by the Products Show result of score the highest, two or more aeronautical productsForm user's recommendation results and return.
For example, when a login user at single product page browsing certain a commodity, this single product page seen andSee and recommend 5 aeronautical products, be called basis with front 20 of the recommendation set of having seen and seen of this aeronautical product,Now recommend set element number be greater than recommended products number threshold value 10 and do not need fast-selling ranking list to mendFill, mate to recommend aeronautical product data and passenger in set to draw a portrait data, matching result is carried outMarking, and sorts to recommended products from big to small according to weighting score, by the highest 5 of weighting scoreThe Products Show result of aeronautical product forms user's recommendation results and returns.
In fact, recommend set to comprise that multiple aeronautical product, each aeronautical product have again three attributes (boatEmpty company, target pattern and product price), by three attribute (aviation public affairs of these three attributes and passenger's portraitDepartment, the line of flight and consuming capacity) carry out cosine method of average coupling, the result of coupling is between 0 to 1Score, according to the score of matching result from height to low arrangement Products Show result. Therefore, can think and mate knotFruit is exactly Products Show result, also can think that matching result adds that aeronautical product data form Products Show knotReally.
Again for example, when one not login user at single product page browsing certain a commodity, the seeing of this single product pageBuy and recommend 5 products, with seeing of this aeronautical product buy and all recommend 6 aviations of set to produceProduct are basis, now recommend set element number to be less than to recommend aeronautical product number threshold value 10 and need to be from aviationThe fast-selling ranking list of product is obtained 4 and is supplemented, and draws a portrait data because login user does not have passenger, straightConnect according to the airline of aviation product data in the recommendation set after supplementing, target boat letter and product priceGive a mark, from big to small Products Show result is sorted according to score, gathered by the recommendation after supplementingIn come the aeronautical product of first 5 Products Show result form user's recommendation results returning.
As shown in Figure 7, the execution flow process of real-time message processing module comprises:
Receive log information, receive and include aeronautical product data and use from the transmission of recommendation service engine modulesThe daily record of family behavioral data;
Deposit daily record in Distributed Message Queue;
Streaming Message Processing is obtained daily record from Distributed Message Queue, according to the format analysis day definingWill.
The daily record of parsing is write respectively in distributed file system and memory database.
The execution flow process of data cleansing module as shown in Figure 8, comprises the steps:
Start off-line operation every day, from distributed file system, extract data.
Data cleansing, data and abnormal data etc. that filtering can not normally be resolved.
Data processing, processes processing to the data of cleaning.
Propelling data, deposits data after treatment processing in distributed file system in, by distributed field systemThe passenger who prestores in system draws a portrait data and process data after treatment and deposit memory database in. For example, can be byUser behavior data and passenger draw a portrait data correlation, obtain the corresponding relation of website user and passenger's portrait and depositEnter memory database.
As shown in Figure 9, proposed algorithm module execution flow process comprises:
Start off-line operation every day, extract aeronautical product data and user behavior number in distributed file systemAccording to;
Calculate the Products Show result of each aeronautical product various dimensions according to the user behavior data cleaning,To the recommendation set of each aeronautical product;
Calculate fast-selling ranking list according to the user behavior data cleaning;
Deposit recommendation results, fast-selling ranking list and aeronautical product data in memory database.
Here, can adopt based on number of clicks and click the collaborative filtering of time weight method, calculate withThe Products Show result that aeronautical product is relevant.
For example, can be respectively with website all over products, each product series products and segmentation classification product, according to purchasingBuy number of times and calculate fast-selling ranking list.
Embodiment of the present invention advantage is: for user arranges unique high in the clouds mark, capture aeronautical product dataAnd user behavior data, can under the prerequisite that does not affect the normal operation of airline's e-commerce website, obtainData, have effectively solved conditional electronic commercial affairs and have extracted the problem in the inapplicable aeronautical technology of product feature field,And can all sidedly website user and passenger's portrait be merged, utilize data to carry out many to passenger's attributeThe description of dimension, and carry out Products Show in conjunction with the feature of aeronautical product, effectively solve conditional electronicThe problem that business users model is single.
Those skilled in the art should understand, embodiments of the invention can be provided as method, system or meterCalculation machine program product. Therefore, the present invention can adopt hardware implementation example, implement software example or in conjunction with software andThe form of the embodiment of hardware aspect. And the present invention can adopt one or more and wherein include calculatingThe computer-usable storage medium of machine usable program code (includes but not limited to magnetic disc store and optical storageDevice etc.) form of the upper computer program of implementing.
The present invention is that reference is according to the method for the embodiment of the present invention, equipment (system) and computer programFlow chart and/or block diagram describe. Should understand can be by computer program instructions realization flow figure and/or sideFlow process in each flow process in block diagram and/or square frame and flow chart and/or block diagram and/or the knot of square frameClose. Can provide these computer program instructions to all-purpose computer, special-purpose computer, Embedded Processor orThe processor of other programmable data processing device is to produce a machine, and making can by computer or otherThe instruction that the processor of programming data treatment facility is carried out produces for realizing in flow process or multiple of flow chartThe device of the function of specifying in square frame of flow process and/or block diagram or multiple square frame.
These computer program instructions also can be stored in can vectoring computer or other programmable data processing deviceIn computer-readable memory with ad hoc fashion work, make to be stored in this computer-readable memoryInstruction produces the manufacture that comprises command device, and this command device is realized at flow process of flow chart or multiple streamThe function of specifying in square frame of journey and/or block diagram or multiple square frame.
These computer program instructions also can be loaded in computer or other programmable data processing device, makeMust on computer or other programmable devices, carry out sequence of operations step to produce computer implemented placeReason, thus the instruction of carrying out on computer or other programmable devices is provided for realizing one of flow chartThe step of the function of specifying in square frame of flow process or multiple flow process and/or block diagram or multiple square frame.
The above, be only preferred embodiment of the present invention, is not intended to limit protection model of the present inventionEnclose.
Claims (20)
1. the aeronautical product recommend method based on cloud service, is characterized in that, described method comprises:
Cloud server receives the information service request that client sends, and obtains aeronautical product data and user's rowFor data;
According to aeronautical product data and user behavior data, calculate the recommendation set of corresponding different aeronautical productsAnd being saved in memory database, the Products Show of the multiple dimensions of aeronautical product of described recommendation set-inclusion is tiedReally;
Receive the recommendation service request that client sends, the recommendation set based in described memory database, meterCalculate user's recommendation results and return to client.
2. method according to claim 1, is characterized in that, described method also comprises:
Client gathers aeronautical product data and user behavior data, send contain described aeronautical product data andThe information service request of described user behavior data is to cloud server;
And/or client transmission arranges the information service request of the unique high in the clouds of user mark and serves to described high in the cloudsDevice;
And/or the recommendation service request that client sends for obtaining user's recommendation results is served to described high in the cloudsDevice;
Wherein, described information service request and recommendation service request comprise respectively the merit for identification request typeCan point identification number and the parameter relevant to self.
3. method according to claim 1 and 2, is characterized in that, described method also comprises:
Receive the information service request that the unique high in the clouds of user mark is set that client sends, described information serviceRequest comprises User IP and user cookie;
According to described information service request, judge whether user exists;
If user exists, the unique high in the clouds of corresponding inquiry user is identified and returns to described client;
If user does not exist, for user arranges the unique high in the clouds of new user mark, and by unique this userHigh in the clouds mark returns to described client;
Wherein, the unique high in the clouds of described user mark is for mark user's inter-network behavior.
4. method according to claim 1 and 2, is characterized in that, described method also comprises: clientTermination is received user's recommendation results that described cloud server returns and is showed user.
5. method according to claim 1, is characterized in that:
Obtain after aeronautical product data and user behavior data, by described aeronautical product data and user behavior numberAccording to depositing message queue in;
From described message queue, obtain data and deposit respectively distributed file system and memory database in;
Aeronautical product data in described distributed file system and user behavior data are cleaned and addedWork;
The passenger who prestores in distributed file system is drawn a portrait to data and deposit memory database in.
6. method according to claim 5, is characterized in that, according to aeronautical product data and userBehavioral data, calculates the recommendation set of corresponding different aeronautical products and deposits memory database in, for:
Taking the aeronautical product data of described distributed file system and user behavior data as basis, use and recommendAlgorithm calculates the Products Show result of different aeronautical product various dimensions, obtains the recommendation set of different aeronautical products,Deposit described recommendation set and aeronautical product data in memory database.
7. according to the method described in claim 1 or 6, it is characterized in that, described calculating user recommendation results,Comprise:
Judge according to recommendation service request whether user logs in, if user logs in, based on described useUnique high in the clouds, family mark identification passenger portrait, and by the aeronautical product data of described correspondence recommendation service request withActive user's passenger draws a portrait data mates, to each Products Show knot in the recommendation set of aeronautical productFruit is given a mark, and determines Products Show result to be recommended, by Products Show to be recommended according to marking resultResult forms described user's recommendation results.
8. method according to claim 7, is characterized in that:
If user logs in, the passenger of active user in described memory database is drawn a portrait to aviation in dataCompany, the line of flight and consuming capacity, with airline, target pattern and the product in aeronautical product dataPrice is mated, and from big to small each is recommended to the Products Show in set according to the score of matching resultResult sorts, and is obtained by the Products Show result of score the highest, two or more aeronautical productsUser's recommendation results.
9. method according to claim 6, is characterized in that, described method also comprises:
If user does not log in, based on recommendation set corresponding to aeronautical product in current recommendation service request,And choose at random from fast-selling ranking list, obtain new recommendation set, the aeronautical product of recommending in this recommendation setNumber be not less than predetermined threshold value; According to the airline in aeronautical product data, target pattern and product valencyLattice are given a mark, and from big to small each recommendation results in new recommendation set are sorted according to score,Products Show result by score the highest, two or more aeronautical products obtains user's recommendation results;
Wherein, described fast-selling ranking list calculates according to described aeronautical product data and user behavior data,Include the related data of fast-selling aeronautical product.
10. method according to claim 1, is characterized in that:
Described user behavior data comprises: web site name, user cookie, user session, name of productWith the behavioral data of user at the enterprising line operate in website;
Described aeronautical product data comprise: web site name, product category, product subdivision, name of product, boatEmpty company, target pattern and product price.
11. 1 kinds of aeronautical product commending systems based on cloud service, is characterized in that, described system comprises cloudEnd server, described cloud server comprises recommendation service engine modules and proposed algorithm module;
Described recommendation service engine modules, the information service request sending for receiving client, obtains aviationProduct data and user behavior data; And, the recommendation service request sending for receiving client, based onRecommendation set in memory database, calculates user's recommendation results and returns to client;
Described proposed algorithm module, for according to described aeronautical product data and user behavior data, calculatesThe recommendation set of corresponding different aeronautical products is also saved in described memory database, described recommendation set-inclusion oneThe Products Show result of the multiple dimensions of individual aeronautical product.
12. systems according to claim 11, is characterized in that, described recommendation service engine modules,Also for:
Receive the information service request that the unique high in the clouds of user mark is set that client sends, described information serviceRequest comprises User IP and user cookie;
According to described information service request, judge whether user exists;
If user exists, the unique high in the clouds of corresponding inquiry user is identified and returns to described client;
If user does not exist, for user arranges the unique high in the clouds of new user mark, and by unique this userHigh in the clouds mark returns to described client;
Wherein, the unique high in the clouds of described user mark is for mark user's inter-network behavior.
13. systems according to claim 11, is characterized in that, described cloud server also comprises:Real-time message processing module and data cleansing module; Wherein,
Real-time message processing module, the aeronautical product passing over for receiving described recommendation service engine modulesData and user behavior data, deposit described aeronautical product data and user behavior data in message queue; FromIn described message queue, obtain data and deposit respectively distributed file system and memory database in;
Data cleansing module, for aeronautical product data and user behavior in described distributed file systemData are cleaned and are processed; The passenger who prestores in distributed file system is drawn a portrait to data and deposit internal storage data inStorehouse.
14. systems according to claim 13, is characterized in that,
Described proposed algorithm module, for: with aeronautical product data and the user of described distributed file systemBehavioral data is basis, uses proposed algorithm to calculate the Products Show result of different aeronautical product various dimensions,To the recommendation set of different aeronautical products, deposit described recommendation set and aeronautical product data in memory database.
15. systems according to claim 14, is characterized in that,
Described proposed algorithm module, for: judge according to recommendation service request whether user logs in, ifUser logs in, based on the unique high in the clouds of described user mark identification passenger portrait, and by described correspondence recommendationThe aeronautical product data of service request and active user's passenger draw a portrait data mates, based on the use logging inFamily difference is given a mark to each Products Show result in the recommendation set of aeronautical product, true according to marking resultFixed Products Show result to be recommended, forms described user's recommendation results by Products Show result to be recommended.
16. methods according to claim 15, is characterized in that:
Described proposed algorithm module, for: if user logs in, by active user in memory databasePassenger draw a portrait airline in data, the line of flight and consuming capacity, with the aviation in aeronautical product dataCompany, target pattern and product price are mated, and according to the score of matching result from big to small to eachRecommend the Products Show result in set to sort, produced by score the highest, two or more aviationThe Products Show result of product obtains user's recommendation results.
17. methods according to claim 14, is characterized in that,
Described proposed algorithm module, also for: if user does not log in, based on current recommendation service requestThe recommendation set that middle aeronautical product is corresponding, and choose at random from fast-selling ranking list, obtain new recommendation set,In this recommendation set, the number of the aeronautical product of recommending is not less than predetermined threshold value; According in aeronautical product dataAirline, target pattern and product price are given a mark, and according to score from big to small to new recommendation collectionIn closing, each recommendation results sorts, by the product of score the highest, two or more aeronautical productsRecommendation results obtains user's recommendation results;
Wherein, described fast-selling ranking list by described proposed algorithm module according to described aeronautical product data and userBehavioral data calculates and deposits in memory database, includes the related data of fast-selling aeronautical product.
18. systems according to claim 11, is characterized in that, described system also comprises: client,Described client comprises service request module, for gathering aeronautical product data and user behavior data, sendsThe information service request that contains described aeronautical product data and described user behavior data is to cloud server; With/Or client sends the information service request that the unique high in the clouds of user mark is set to described cloud server; With/Or client sends the recommendation service request that is used for obtaining user's recommendation results to described cloud server; ItsIn, described information service request and recommendation service request comprise respectively the function point mark for identification request typeKnow number and the parameter relevant to self;
And, for receiving user's recommendation results that described recommendation service engine modules returns and showing user.
19. systems according to claim 19, is characterized in that, described service request module is deployed inThe client of the electricity business of airline website.
20. methods according to claim 11, is characterized in that:
Described user behavior data comprises: web site name, user cookie, user session, name of productWith the behavioral data of user at the enterprising line operate in website;
Described aeronautical product data comprise: web site name, product category, product subdivision, name of product, boatEmpty company, target pattern and product price.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511000479.XA CN105654341A (en) | 2015-12-28 | 2015-12-28 | Aviation product recommendation system and aviation product recommendation method based on cloud service |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511000479.XA CN105654341A (en) | 2015-12-28 | 2015-12-28 | Aviation product recommendation system and aviation product recommendation method based on cloud service |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105654341A true CN105654341A (en) | 2016-06-08 |
Family
ID=56476905
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511000479.XA Pending CN105654341A (en) | 2015-12-28 | 2015-12-28 | Aviation product recommendation system and aviation product recommendation method based on cloud service |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105654341A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107240005A (en) * | 2017-06-13 | 2017-10-10 | 携程旅游网络技术（上海）有限公司 | The commending system and method for air ticket addition product |
| CN107332910A (en) * | 2017-07-03 | 2017-11-07 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
| CN107369045A (en) * | 2017-07-20 | 2017-11-21 | 无锡零零七科技有限公司 | Advertisement placement method and device |
| CN107786659A (en) * | 2017-10-27 | 2018-03-09 | 中航信移动科技有限公司 | Service methods of exhibiting and device |
| CN108197980A (en) * | 2017-12-27 | 2018-06-22 | 中国科学院上海高等研究院 | Illustration generation method/system, storage medium and the terminal of personalized shopper |
| CN109300021A (en) * | 2018-11-29 | 2019-02-01 | 爱保科技（横琴）有限公司 | Insure recommended method and device |
| CN109493200A (en) * | 2019-01-24 | 2019-03-19 | 深圳市活力天汇科技股份有限公司 | A kind of recommended method of air ticket trip commodity |
| WO2019061614A1 (en) * | 2017-09-29 | 2019-04-04 | 平安科技（深圳）有限公司 | Loan product matching method, apparatus and computer-readable storage medium |
| CN110110236A (en) * | 2019-05-16 | 2019-08-09 | 苏州达家迎信息技术有限公司 | A kind of information-pushing method, device, equipment and storage medium |
| CN110134869A (en) * | 2019-05-16 | 2019-08-16 | 苏州达家迎信息技术有限公司 | A kind of information-pushing method, device, equipment and storage medium |
| CN110555781A (en) * | 2019-08-30 | 2019-12-10 | 苏州熙烁数字科技有限公司 | working method of multi-angle associated digital service community based on target form |
| CN111368190A (en) * | 2020-02-28 | 2020-07-03 | 北京百度网讯科技有限公司 | Information recommendation method and device |
| CN112884399A (en) * | 2021-01-28 | 2021-06-01 | 重庆允成互联网科技有限公司 | Industrial product online marketing method, server and system based on storeroom front-end processor |
| CN113032587A (en) * | 2019-12-25 | 2021-06-25 | 北京达佳互联信息技术有限公司 | Multimedia information recommendation method, system, device, terminal and server |
| CN113360808A (en) * | 2021-06-30 | 2021-09-07 | 中国农业银行股份有限公司 | Information recommendation method, device and equipment |
| CN113850695A (en) * | 2021-09-07 | 2021-12-28 | 海南太美航空股份有限公司 | Big data-based aviation information management platform and method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104090894A (en) * | 2013-12-18 | 2014-10-08 | 深圳市腾讯计算机系统有限公司 | Method for online parallel computing of recommended information, device for online parallel computing of recommended information, and server for online parallel computing of recommended information |
| CN104166713A (en) * | 2014-08-14 | 2014-11-26 | 百度在线网络技术（北京）有限公司 | Network service recommending method and device |
| CN104835057A (en) * | 2015-04-02 | 2015-08-12 | 百度在线网络技术（北京）有限公司 | Method and device for obtaining consumption feature information of network user |
| CN105117418A (en) * | 2015-07-30 | 2015-12-02 | 百度在线网络技术（北京）有限公司 | Search based service information management system and method |
-
2015
- 2015-12-28 CN CN201511000479.XA patent/CN105654341A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104090894A (en) * | 2013-12-18 | 2014-10-08 | 深圳市腾讯计算机系统有限公司 | Method for online parallel computing of recommended information, device for online parallel computing of recommended information, and server for online parallel computing of recommended information |
| CN104166713A (en) * | 2014-08-14 | 2014-11-26 | 百度在线网络技术（北京）有限公司 | Network service recommending method and device |
| CN104835057A (en) * | 2015-04-02 | 2015-08-12 | 百度在线网络技术（北京）有限公司 | Method and device for obtaining consumption feature information of network user |
| CN105117418A (en) * | 2015-07-30 | 2015-12-02 | 百度在线网络技术（北京）有限公司 | Search based service information management system and method |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107240005A (en) * | 2017-06-13 | 2017-10-10 | 携程旅游网络技术（上海）有限公司 | The commending system and method for air ticket addition product |
| CN107332910B (en) * | 2017-07-03 | 2020-07-31 | 北京京东尚科信息技术有限公司 | Information pushing method and device |
| CN107332910A (en) * | 2017-07-03 | 2017-11-07 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
| CN107369045A (en) * | 2017-07-20 | 2017-11-21 | 无锡零零七科技有限公司 | Advertisement placement method and device |
| WO2019061614A1 (en) * | 2017-09-29 | 2019-04-04 | 平安科技（深圳）有限公司 | Loan product matching method, apparatus and computer-readable storage medium |
| CN107786659A (en) * | 2017-10-27 | 2018-03-09 | 中航信移动科技有限公司 | Service methods of exhibiting and device |
| CN108197980A (en) * | 2017-12-27 | 2018-06-22 | 中国科学院上海高等研究院 | Illustration generation method/system, storage medium and the terminal of personalized shopper |
| CN108197980B (en) * | 2017-12-27 | 2022-02-01 | 中国科学院上海高等研究院 | Method/system for generating portrait of personalized shopper, storage medium and terminal |
| CN109300021A (en) * | 2018-11-29 | 2019-02-01 | 爱保科技（横琴）有限公司 | Insure recommended method and device |
| CN109493200A (en) * | 2019-01-24 | 2019-03-19 | 深圳市活力天汇科技股份有限公司 | A kind of recommended method of air ticket trip commodity |
| CN109493200B (en) * | 2019-01-24 | 2021-08-10 | 深圳市活力天汇科技股份有限公司 | Method for recommending airline ticket travel commodities |
| CN110134869B (en) * | 2019-05-16 | 2022-03-18 | 苏州达家迎信息技术有限公司 | Information pushing method, device, equipment and storage medium |
| CN110134869A (en) * | 2019-05-16 | 2019-08-16 | 苏州达家迎信息技术有限公司 | A kind of information-pushing method, device, equipment and storage medium |
| CN110110236B (en) * | 2019-05-16 | 2021-10-26 | 苏州达家迎信息技术有限公司 | Information pushing method, device, equipment and storage medium |
| CN110110236A (en) * | 2019-05-16 | 2019-08-09 | 苏州达家迎信息技术有限公司 | A kind of information-pushing method, device, equipment and storage medium |
| CN110555781A (en) * | 2019-08-30 | 2019-12-10 | 苏州熙烁数字科技有限公司 | working method of multi-angle associated digital service community based on target form |
| CN113032587A (en) * | 2019-12-25 | 2021-06-25 | 北京达佳互联信息技术有限公司 | Multimedia information recommendation method, system, device, terminal and server |
| CN113032587B (en) * | 2019-12-25 | 2023-07-28 | 北京达佳互联信息技术有限公司 | Multimedia information recommendation method, system, device, terminal and server |
| CN111368190A (en) * | 2020-02-28 | 2020-07-03 | 北京百度网讯科技有限公司 | Information recommendation method and device |
| CN111368190B (en) * | 2020-02-28 | 2023-08-15 | 北京百度网讯科技有限公司 | Information recommendation method and device |
| CN112884399A (en) * | 2021-01-28 | 2021-06-01 | 重庆允成互联网科技有限公司 | Industrial product online marketing method, server and system based on storeroom front-end processor |
| CN112884399B (en) * | 2021-01-28 | 2023-07-25 | 重庆允丰科技有限公司 | Online marketing method, server and system for industrial products based on warehouse front-end processor |
| CN113360808A (en) * | 2021-06-30 | 2021-09-07 | 中国农业银行股份有限公司 | Information recommendation method, device and equipment |
| CN113850695A (en) * | 2021-09-07 | 2021-12-28 | 海南太美航空股份有限公司 | Big data-based aviation information management platform and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105654341A (en) | Aviation product recommendation system and aviation product recommendation method based on cloud service | |
| CN110222272B (en) | Potential customer mining and recommending method | |
| JP5596152B2 (en) | Information matching method and system on electronic commerce website | |
| US9569499B2 (en) | Method and apparatus for recommending content on the internet by evaluating users having similar preference tendencies | |
| CN102254028A (en) | Personalized commodity recommending method and system which integrate attributes and structural similarity | |
| CN108805598B (en) | Similarity information determination method, server and computer-readable storage medium | |
| CN106355442A (en) | Online precise advertising method and system based on big data driving | |
| CN105069654A (en) | User identification based website real-time/non-real-time marketing investment method and system | |
| CN105069168A (en) | Search word recommendation method and apparatus | |
| KR20200088131A (en) | Item recommendation method and apparatus using graph database | |
| CN105447186A (en) | Big data platform based user behavior analysis system | |
| JP6679451B2 (en) | Selection device, selection method, and selection program | |
| CN102713889A (en) | Correlating web page visits and conversions with external references | |
| CN105678317B (en) | Information processing method and server | |
| CN103412930A (en) | Method for identifying attributes of internet users | |
| CN111949887A (en) | Item recommendation method and device and computer-readable storage medium | |
| CN105046630A (en) | image tag add system | |
| KR20190097879A (en) | System for providing marketing platform and method and computer program for social network based marketing using the same | |
| CN102117467A (en) | Question and answer mode online shopping guide method | |
| JP7231317B2 (en) | Estimation device, estimation method and estimation program | |
| CN112632405A (en) | Recommendation method, device, equipment and storage medium | |
| CN111784428A (en) | Information pushing method and device, electronic commerce system and storage medium | |
| CN105022773A (en) | Image processing system including image priority | |
| CN109242656A (en) | A kind of commodity supplying system based on cloud computing sales data and user's evaluation | |
| CN105183739A (en) | Image Processing Server |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160608 |
|
| WD01 | Invention patent application deemed withdrawn after publication |