CN109213936A

CN109213936A - product search method and device

Info

Publication number: CN109213936A
Application number: CN201811403386.5A
Authority: CN
Inventors: 许力斌; 姜瑞欣; 毛丹琪; 雷健雄
Original assignee: Beijing Jingdong Financial Technology Holding Co Ltd
Current assignee: JD Digital Technology Holdings Co Ltd; Jingdong Technology Holding Co Ltd
Priority date: 2018-11-22
Filing date: 2018-11-22
Publication date: 2019-01-15
Anticipated expiration: 2038-11-22
Also published as: CN109213936B

Abstract

Present disclose provides a kind of product search methods, comprising: receives the searching request that user is directed to commodity；Search result corresponding with described search request is obtained, described search result includes the relevant information of the commodity；The attribute data of attribute data and the commodity based on the user, determines optimal type of offer；Described search result is exported in association with the favor information for meeting the optimal type of offer.The disclosure additionally provides a kind of commercial articles searching device, a kind of computer equipment and a kind of computer readable storage medium.

Description

Product search method and device

Technical field

This disclosure relates to Internet technical field, more particularly, to a kind of product search method and device.

Background technique

With the fast development of Internet technology, e-commerce is rapidly growing, and various electric business platforms provide varied Line on commodity transaction channel, greatly facilitate the work and life of people.When merchandise news is shown, can usually have should The favor information of commodity, to promote purchase of the user to commodity using favor information.

However in the prior art, the putting mode of the favor information of commodity is very rough, such as institute whithin a period of time There is the favor information of commodity to belong to same type of offer, alternatively, the favor information of certain class I goods belongs to same type, does not have Consider that different user may generate different preferential preferences under the specific situation in face of different commodity.

Summary of the invention

In view of this, present disclose provides a kind of commercial articles searchings of preferential preference for more precisely, various dimensions considering user Method and apparatus.

An aspect of this disclosure provides a kind of product search method, comprising: the search for receiving user for commodity is asked It asks；Search result corresponding with described search request is obtained, described search result includes the relevant information of the commodity；It is based on The attribute data of the attribute data of the user and the commodity determines optimal type of offer；And it is described optimal excellent with meeting The favor information of favour type exports described search result in association.

In accordance with an embodiment of the present disclosure, the attribute data of the above-mentioned attribute data based on the user and the commodity, really Fixed optimal type of offer includes: that the attribute data for any type of offer, based on the user obtains the user about institute The first predilection grade of type of offer is stated, the attribute data based on the commodity obtains the commodity about the type of offer Second predilection grade；Based on first predilection grade and second predilection grade, obtains the user and the commodity close In the synthesis predilection grade of the type of offer, the synthesis predilection grade of the as described type of offer.Compare each type of offer Comprehensive predilection grade, using the comprehensive highest type of offer of predilection grade as optimal type of offer.

In accordance with an embodiment of the present disclosure, the above-mentioned attribute data based on the user obtains the user about described preferential First predilection grade of type includes: to obtain the corresponding first History Order data of the user, the first History Order data In the quantity of the order comprising meeting the favor information of the type of offer be the first quantity, wrap in the first History Order data The quantity of order containing the favor information for meeting any type of offer is the second quantity；Obtain corresponding second history of total user Order data, the quantity of the order in the second History Order data comprising meeting the favor information of the type of offer are third Quantity, the quantity of the order in the second History Order data comprising meeting the favor information of any type of offer are the 4th number Amount；Using the ratio of the first quantity and the second quantity relative to third quantity and the accounting of the ratio of the 4th quantity as the user The first predilection grade about the type of offer.

In accordance with an embodiment of the present disclosure, the above-mentioned attribute data based on the commodity obtains the commodity about described preferential Second preference-score of type includes: to determine the corresponding Price Zoning of the commodity based on the prices of the commodity；Described in acquisition The corresponding third History Order data of commodity in Price Zoning, comprising meeting the preferential class in the third History Order data The quantity of the order of the favor information of type is the 5th quantity, includes to meet any type of offer in the third History Order data The quantity of the order of favor information is the 6th quantity；Obtain the corresponding 4th History Order data of all commodity, the 4th history The quantity of order in order data comprising meeting the favor information of the type of offer is the 7th quantity, the 4th History Order The quantity of order in data comprising meeting the favor information of any type of offer is the 8th quantity；By the 5th quantity and the 6th number The ratio of amount relative to the 7th quantity and the ratio of the 8th quantity accounting as the commodity about the type of offer the Two predilection grades.

In accordance with an embodiment of the present disclosure, above-mentioned the first predilection grade and institute based on the user about the type of offer Second predilection grade of the commodity about the type of offer is stated, obtains the user and the commodity about the type of offer Comprehensive predilection grade includes: to obtain corresponding first weight of above-mentioned first predilection grade, and it is corresponding to obtain above-mentioned second predilection grade The second weight；Based on first weight and second weight to above-mentioned first predilection grade and above-mentioned second predilection grade It is weighted summation, obtains the synthesis predilection grade of the user and the commodity about the type of offer.

In accordance with an embodiment of the present disclosure, corresponding first weight of above-mentioned the first predilection grade of acquisition obtains the second preference and comments Dividing corresponding second weight includes: to determine that the corresponding objective realm of the user is other according to the attribute data of the user, different The different purchasing power of objective group's categorized representation is horizontal；The corresponding Price Zoning of the commodity is determined according to the price of the commodity；Structure Preferential preference pattern is built, the input of the preferential preference pattern includes the objective realm not about the first preference of each type of offer Scoring and second predilection grade of the Price Zoning about each type of offer, the output of the preferential preference pattern are the visitor Realm not highest comprehensive predilection grade corresponding with the Price Zoning, the parameter of the preferential preference pattern include the first power Weight parameter and the second weight parameter；Obtain the objective realm highest actual preference scoring not corresponding with the Price Zoning； It is that target optimizes the parameter of the preferential preference pattern with the highest actual preference scoring, by the first optimal power The value of weight parameter is as first weight, using the value of the second optimal weight parameter as second weight.

In accordance with an embodiment of the present disclosure, the above-mentioned attribute data according to the user determines the corresponding objective realm of the user It does not include: average income data, average order price and/or the average consumption gold based on the user in preset period of time Volume determines that the corresponding objective realm of the user is other.

In accordance with an embodiment of the present disclosure, the preferential preference pattern of above-mentioned building includes: that acquisition is other corresponding to the objective realm The attribute data of multiple sample of users obtains the attribute data for corresponding to multiple sample commodity of the Price Zoning；Based on institute State the attribute data of multiple sample of users and the attribute data of the multiple sample commodity；For any type of offer, it is based on institute The first predilection grade that the attribute data of multiple sample of users obtains the multiple sample of users about the type of offer is stated, by institute The sum of first predilection grade of multiple sample of users about the type of offer is stated as the objective realm not about the type of offer The first predilection grade；Described in attribute data based on the multiple sample commodity obtains corresponding to the multiple sample commodity Second predilection grade of the Price Zoning about the type of offer；Based on the first weight parameter and the second weight parameter to the objective group Classification about the type of offer the first predilection grade and the Price Zoning about the type of offer the second predilection grade into Row weighted sum obtains the objective realm not synthesis predilection grade about the type of offer corresponding with the Price Zoning.

In accordance with an embodiment of the present disclosure, the above-mentioned acquisition objective realm not highest reality corresponding with the Price Zoning Predilection grade includes: that any type of offer is obtained any sample of users and the Price Zoning the corresponding 5th is gone through History order data, the quantity of the order in the 5th History Order data comprising meeting the favor information of the type of offer are the Nine quantity, the quantity of the order in the 5th History Order data comprising meeting the favor information of any type of offer are the tenth Quantity；Using the ratio accounting of the 9th quantity and the tenth quantity as the sample of users and the Price Zoning about the preferential class The actual preference of type scores, and the multiple sample of users and the Price Zoning are scored about the actual preference of the type of offer The sum of do not score with the Price Zoning about the actual preference of the type of offer as the objective realm；Compare the objective realm It does not score with the Price Zoning about the actual preference of each type of offer, does not obtain the objective realm not and the Price Zoning pair The highest actual preference scoring answered.

In accordance with an embodiment of the present disclosure, corresponding first weight of above-mentioned the first predilection grade of acquisition obtains the second preference and comments Divide corresponding second weight further include: when that can not determine that the corresponding visitor's realm of the user is other, using the first preset value as institute The first weight is stated, using the second preset value as second weight.

Another aspect of the disclosure provides a kind of commercial articles searching device, including receiving module, acquisition module, decision model Block and output module.Receiving module is used to receive the searching request that user is directed to commodity.Module is obtained to search for obtaining with described Rope requests corresponding search result, and described search result includes the relevant information of the commodity.Decision-making module is used to be based on institute The attribute data of user and the attribute data of the commodity are stated, determines optimal type of offer.Output module be used for meet it is described The favor information of optimal type of offer exports described search result in association.

In accordance with an embodiment of the present disclosure, the attribute of attribute data and the commodity of the decision-making module based on the user Data determine that optimal type of offer includes: that decision-making module is used for for any type of offer, the attribute data based on the user Obtain first predilection grade of the user about the type of offer；Attribute data based on the commodity obtains the commodity The second predilection grade about the type of offer；Based on first predilection grade and second predilection grade, institute is obtained State the synthesis predilection grade of user and the commodity about the type of offer；The comprehensive highest type of offer of predilection grade is made For optimal type of offer.

In accordance with an embodiment of the present disclosure, decision-making module obtains the user about described based on the attribute data of the user First predilection grade of type of offer includes: decision-making module for obtaining the corresponding first History Order data of the user, institute The quantity for stating the order in the first History Order data comprising meeting the favor information of the type of offer is the first quantity, described The quantity of order in first History Order data comprising meeting the favor information of any type of offer is the second quantity；It obtains complete The corresponding second History Order data of body user, comprising meeting the preferential of the type of offer in the second History Order data The quantity of the order of information is third quantity, includes the preferential letter for meeting any type of offer in the second History Order data The quantity of the order of breath is the 4th quantity；By the ratio of the first quantity and the second quantity relative to third quantity and the 4th quantity The accounting of ratio is as first predilection grade.

In accordance with an embodiment of the present disclosure, decision-making module obtains the commodity about described based on the attribute data of the commodity Second preference-score of type of offer includes: that decision-making module is used to determine the corresponding valence of the commodity based on the price of the commodity Lattice subregion；The corresponding third History Order data of commodity in the Price Zoning are obtained, in the third History Order data The quantity of order comprising meeting the favor information of the type of offer is the 5th quantity, is wrapped in the third History Order data The quantity of order containing the favor information for meeting any type of offer is the 6th quantity；Obtain corresponding 4th history of all commodity Order data, the quantity of the order in the 4th History Order data comprising meeting the favor information of the type of offer are the Seven quantity, the quantity of the order in the 4th History Order data comprising meeting the favor information of any type of offer are the 8th Quantity；Using the ratio of the 5th quantity and the 6th quantity relative to the 7th quantity and the accounting of the ratio of the 8th quantity as described the Two predilection grades.

In accordance with an embodiment of the present disclosure, decision-making module is based on first predilection grade and second predilection grade, obtains It about the synthesis predilection grade of the type of offer include: decision-making module for obtaining first partially to the user and the commodity Corresponding first weight of favorable comment point obtains corresponding second weight of the second predilection grade；Based on first weight and described Two weights are weighted summation to first predilection grade and second predilection grade, obtain the comprehensive predilection grade.

In accordance with an embodiment of the present disclosure, decision-making module obtains corresponding first weight of the first predilection grade, obtains second partially Corresponding second weight of favorable comment point includes: that decision-making module is used to determine that the user is corresponding according to the attribute data of the user Objective realm is other, and the different purchasing power of different objective group's categorized representations is horizontal；The commodity pair are determined according to the price of the commodity The Price Zoning answered；Preferential preference pattern is constructed, the input of the preferential preference pattern includes the objective realm not about each excellent Second predilection grade of the first predilection grade and the Price Zoning of favour type about each type of offer, the preferential preference mould The output of type is the objective realm not highest comprehensive predilection grade corresponding with the Price Zoning, the preferential preference pattern Parameter include the first weight parameter and the second weight parameter；Obtain the objective realm not highest corresponding with the Price Zoning Actual preference scoring；It is that target is excellent to the parameter progress of the preferential preference pattern with the highest actual preference scoring Change, using the value of the first optimal weight parameter as first weight, using the value of the second optimal weight parameter as Second weight.

In accordance with an embodiment of the present disclosure, decision-making module determines the corresponding visitor of the user according to the attribute data of the user Realm does not include: that decision-making module is used for based on average income data of the user in preset period of time, average price-purchase order Lattice and/or the average consumption amount of money determine that the corresponding objective realm of the user is other.

In accordance with an embodiment of the present disclosure, it includes: decision-making module for obtaining correspondence that decision-making module, which constructs preferential preference pattern, In the attribute data of the other multiple sample of users of the objective realm, the multiple sample commodity for corresponding to the Price Zoning are obtained Attribute data；The attribute data of attribute data and the multiple sample commodity based on the multiple sample of users；For any Type of offer, the attribute data based on the multiple sample of users obtain of the multiple sample of users about the type of offer One predilection grade, the sum of first predilection grade by the multiple sample of users about the type of offer are other as the objective realm The first predilection grade about the type of offer；Attribute data based on the multiple sample commodity obtains the multiple sample quotient Second predilection grade of the Price Zoning corresponding to product about the type of offer；Based on the first weight parameter and the second weight Parameter is to the objective realm not about the first predilection grade of the type of offer and the Price Zoning about the type of offer Second predilection grade is weighted summation, and to obtain the objective realm not corresponding with the Price Zoning about the comprehensive of the type of offer Close predilection grade.

In accordance with an embodiment of the present disclosure, it is corresponding with the Price Zoning not highest to obtain the objective realm for decision-making module Actual preference scoring includes: that decision-making module is used to obtain any sample of users and the price for any type of offer The corresponding 5th History Order data of subregion, comprising meeting the favor information of the type of offer in the 5th History Order data The quantity of order be the 9th quantity, comprising meeting the favor information of any type of offer in the 5th History Order data The quantity of order is the tenth quantity；Using the ratio accounting of the 5th quantity and the 6th quantity as the sample of users and the price Subregion scores about the actual preference of the type of offer, by the multiple sample of users and the Price Zoning about the preferential class The sum of actual preference scoring of type is not commented with the Price Zoning about the actual preference of the type of offer as the objective realm Point；Compare the objective realm not score with the Price Zoning about the actual preference of each type of offer, obtains the objective realm Highest actual preference scoring not corresponding with the Price Zoning.

In accordance with an embodiment of the present disclosure, decision-making module obtains corresponding first weight of the first predilection grade, obtains second partially Corresponding second weight of favorable comment point further include: decision-making module is also used to when that can not determine that the corresponding visitor's realm of the user is other, Using the first preset value as first weight, using the second preset value as second weight.

Another aspect of the present disclosure provides a kind of computer readable storage medium, is stored with computer executable instructions, Described instruction is when executed for realizing method as described above.

Another aspect of the present disclosure provides a kind of computer program, and the computer program, which includes that computer is executable, to be referred to It enables, described instruction is when executed for realizing method as described above.

In accordance with an embodiment of the present disclosure ,/mitigation/can at least be partially solved and inhibit/even avoid right in the prior art The rough dispensing of commodity favor information fully considers user in face of commodity when needing to export the search result of commodity to user Specific situation, the demand by user to favor information, which is placed in specific situation, carries out precise positioning, can be greatly promoted use Purchase of the family to commodity improves user's conversion ratio.

Detailed description of the invention

By referring to the drawings to the description of the embodiment of the present disclosure, the above-mentioned and other purposes of the disclosure, feature and Advantage will be apparent from, in the accompanying drawings:

Fig. 1 diagrammatically illustrates the exemplary system that can apply product search method and device according to the embodiment of the present disclosure System framework；

Fig. 2 diagrammatically illustrates the flow chart of the product search method according to the embodiment of the present disclosure；

Fig. 3 A diagrammatically illustrates the data flow diagram of the commercial articles searching process according to the embodiment of the present disclosure；

Fig. 3 B diagrammatically illustrates the schematic diagram of the IPM model according to the embodiment of the present disclosure；

Fig. 3 C diagrammatically illustrates the schematic diagram of the search result of the commodity according to the embodiment of the present disclosure；

Fig. 3 D diagrammatically illustrates the flow chart of the product search method according to another embodiment of the disclosure；

Fig. 4 diagrammatically illustrates the block diagram of the commercial articles searching device according to the embodiment of the present disclosure；And

Fig. 5 diagrammatically illustrates the block diagram of the computer equipment according to the embodiment of the present disclosure.

Specific embodiment

Hereinafter, will be described with reference to the accompanying drawings embodiment of the disclosure.However, it should be understood that these descriptions are only exemplary , and it is not intended to limit the scope of the present disclosure.In the following detailed description, to elaborate many specific thin convenient for explaining Section is to provide the comprehensive understanding to the embodiment of the present disclosure.It may be evident, however, that one or more embodiments are not having these specific thin It can also be carried out in the case where section.In addition, in the following description, descriptions of well-known structures and technologies are omitted, to avoid Unnecessarily obscure the concept of the disclosure.

Term as used herein is not intended to limit the disclosure just for the sake of description specific embodiment.It uses herein The terms "include", "comprise" etc. show the presence of the feature, step, operation and/or component, but it is not excluded that in the presence of Or add other one or more features, step, operation or component.

There are all terms (including technical and scientific term) as used herein those skilled in the art to be generally understood Meaning, unless otherwise defined.It should be noted that term used herein should be interpreted that with consistent with the context of this specification Meaning, without that should be explained with idealization or excessively mechanical mode.

It, in general should be according to this using statement as " at least one in A, B and C etc. " is similar to Field technical staff is generally understood the meaning of the statement to make an explanation (for example, " system at least one in A, B and C " Should include but is not limited to individually with A, individually with B, individually with C, with A and B, with A and C, have B and C, and/or System etc. with A, B, C).Using statement as " at least one in A, B or C etc. " is similar to, generally come Saying be generally understood the meaning of the statement according to those skilled in the art to make an explanation (for example, " having in A, B or C at least One system " should include but is not limited to individually with A, individually with B, individually with C, with A and B, have A and C, have B and C, and/or the system with A, B, C etc.).

Embodiment of the disclosure provides a kind of product search method and device.This method includes that searching request received Journey, search result acquisition process, preferential decision process and search result export process.Use is received in searching request receive process Family is directed to the searching request of commodity, the search result of corresponding commodity is got in search result acquisition process, in preferential decision mistake Journey determines optimal type of offer based on the attribute data for the user for initiating search and the attribute data for the commodity searched for, and is searching Hitch fruit output process exports search result in a manner of associated with the favor information for meeting optimal type of offer.

Fig. 1 diagrammatically illustrates the exemplary system that can apply product search method and device according to the embodiment of the present disclosure System framework 100.It should be noted that being only the example that can apply the system architecture of the embodiment of the present disclosure shown in Fig. 1, to help Those skilled in the art understand that the technology contents of the disclosure, but it is not meant to that the embodiment of the present disclosure may not be usable for other and set Standby, system, environment or scene.

As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network according to this embodiment 104 and server 105.Network 104 between terminal device 101,102,103 and server 105 to provide communication link Medium.Network 104 may include various connection types, such as wired, wireless communication link or fiber optic cables etc..

User can be used terminal device 101,102,103 and be interacted by network 104 with server 105, to receive or send out Send message etc..Various client applications can be installed, such as shopping class is applied, webpage is clear on terminal device 101,102,103 (merely illustrative) such as device of looking at application, searching class application, instant messaging tools, mailbox client, social platform softwares.

Terminal device 101,102,103 can be the various electronic equipments with display screen and supported web page browsing, packet Include but be not limited to smart phone, tablet computer, pocket computer on knee and desktop computer etc..

Server 105 can be to provide the server of various services, such as utilize terminal device 101,102,103 to user The website browsed provides the back-stage management server (merely illustrative) supported.Back-stage management server can be searched to what is received The data such as rope request analyze etc. processing, and by processing result (such as according to user's request or the search result of generation, Webpage, information or data etc.) feed back to terminal device.

It should be noted that product search method provided by the embodiment of the present disclosure can be executed by server 105.Accordingly Ground, commercial articles searching device provided by the embodiment of the present disclosure generally can be set in server 105.The embodiment of the present disclosure is mentioned The product search method of confession can also be by being different from server 105 and can be with terminal device 101,102,103 and/or server The server or server cluster of 105 communications execute.Correspondingly, commercial articles searching device provided by the embodiment of the present disclosure can also be with Be set to different from server 105 and the server that can be communicated with terminal device 101,102,103 and/or server 105 or In server cluster.

Alternatively, product search method provided by the embodiment of the present disclosure can also be executed by terminal device 101,102,103. Correspondingly, commercial articles searching device provided by the embodiment of the present disclosure generally can be set in terminal device 101,102,103.This Product search method provided by open embodiment can also be by being different from terminal device 101,102,103 and can set with terminal The server or server cluster communicated for 101,102,103 and/or server 105 executes.Correspondingly, embodiment of the present disclosure institute The commercial articles searching device of offer also can be set in be different from terminal device 101,102,103 and can with terminal device 101, 102,103 and/or server 105 communicate server or server cluster in.

It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical, according to realization need It wants, can have any number of terminal device, network and server.

Fig. 2 diagrammatically illustrates the flow chart of the product search method according to the embodiment of the present disclosure.

As shown in Fig. 2, this method is included in operation S201, the searching request that user is directed to commodity is received.

In this operation, the searching request that user is directed to commodity is received, can be and receive the search behavior for corresponding to user Searching request when such as inputting merchandise related information in search column as user, receives the searching request that user is directed to the commodity, It can be the searching request for receiving other behaviors for corresponding to user, such as need to open up when in user's one webpage of opening, the webpage Show the relevant information of one or more commodity, receives the searching request, etc. that user is directed to the one or more commodity at this time.? That is needing to can be seen as under the scene of the relevant information output of commodity in any request received about commodity Receive the searching request that active user is directed to corresponding commodity.

Then, in operation S202, search result corresponding with described search request is obtained, described search result includes institute State the relevant information of commodity.

In this operation, each commodity are identified jointly by Business Information and product information, and different businessmans is gone out The identical product sold can regard different commodity, and the different products that same businessman is sold can also regard different quotient Product, the identical product that only same businessman is sold can just regard same commodity.When in described search result include it is multiple When the relevant information of commodity, multiple commodity refer to mutually different multiple commodity.The relevant information of commodity may include: commodity The corresponding product information of corresponding Business Information, commodity, price of commodity etc. understand the information of the commodity, Ke Yigen for user According to needing to select, herein with no restrictions.

In operation S203, the attribute data of attribute data and the commodity based on the user determines optimal preferential class Type.

In this operation, the user refers to the corresponding user of searching request in operation S201, and the attribute data of user is anti- The characteristic attribute of the user itself is reflected, the characteristic attribute of user itself will affect user to the preference of type of offer, commodity Attribute data reflects the characteristic attribute of commodity, and when user faces a commodity, the characteristic attribute of commodity also will affect user To the preference of type of offer, based on these two aspects because usually determining that when user faces commodity user may most preference Type of offer.

In operation S204, described search result is exported in association with the favor information for meeting the optimal type of offer.

In this operation, determining that user and may meet after the optimal type of offer of most preference when in face of a commodity The favor information of the optimal type of offer exports search result in association, so that being finally presented to the commercial articles searching result of user The favor information is included, more accurately to position user for the demand of favor information, promotes purchase of the user to commodity It buys.

It should be noted that product search method shown in Fig. 2 can be implemented in server side, it can also be in client-side Implement, when this method is when server side is implemented, receives the user that client is submitted in operation S201 and asked for the search of commodity Ask, obtain corresponding with described search request search result in operation S202, operation S203 consideration user's oneself factor with Commodity factor determines optimal type of offer, defeated in association in the favor information for operating S204 with meeting the optimal type of offer Described search is as a result, show corresponding user for the output of described search result by client out.When this method is in client-side When implementation, the searching request that user's input is directed to commodity is received in operation S201, obtains in operation S202 and is requested with described search Corresponding search result considers that user's oneself factor and commodity factor determine optimal type of offer in operation S203, and is grasping Make S204 and exports described search result in association with the favor information for meeting the optimal type of offer.

As it can be seen that method shown in Fig. 2 when needing to export the search result of commodity to user, fully considers user in face of quotient The specific situation of product both will receive user's unique characteristics attribute according to preferential preference of the particular user when facing specific commodity It influences, also will receive this rule of the influence of the characteristic attribute of commodity, the attribute data of attribute data and commodity based on user Determine the optimal type of offer of user's most possible preference under this specific situation, the search result of commodity is optimal preferential with this The corresponding favor information of type exports in association, and the demand by user to favor information, which is placed in specific situation, precisely determine Position can be greatly promoted purchase of the user to commodity, improve user's conversion ratio.

In one embodiment of the present disclosure, attribute data and the commodity of the aforesaid operations S203 based on the user Attribute data determines that optimal type of offer includes: described in attribute data for any type of offer, based on the user obtains First predilection grade of the user about the type of offer, the attribute data based on the commodity obtain the commodity about described Second predilection grade of type of offer, be based on first predilection grade and second predilection grade, obtain the user and Synthesis predilection grade of the commodity about the type of offer, i.e., the synthesis predilection grade of the described type of offer.Compare each excellent The synthesis predilection grade of favour type, using the comprehensive highest type of offer of predilection grade as optimal type of offer.

Citing is illustrated, when receiving searching request of the user A for commodity B, it is assumed that there are three types of type of offer altogether: Type of offer 1, type of offer 2 and type of offer 3.For type of offer 1, attribute data based on user A obtain user A about First predilection grade of type of offer 1, first predilection grade indicate do not considering that commodity are which only considers user A itself spy In the case where levying attribute, be equivalent to user A under the scene of any commodity, user A to the preference of type of offer 1, with And the attribute data based on commodity B obtains second predilection grade of the commodity B about type of offer 1, which indicates In the case where not considering user is the characteristic attribute who only considers commodity, total user is equivalent under the scene of commodity B, Then total user can obtain the preference of type of offer 1 based on above-mentioned first predilection grade and the second predilection grade Synthesis predilection grade of the user A and commodity B about type of offer 1 is obtained, which indicates in the spy for both considering user A Attribute is levied it is further contemplated that being equivalent to user A under the scene of commodity B, user A is to preferential in the case where the characteristic attribute of commodity B The preference of Class1.Similarly, the synthesis predilection grade and type of offer 3 of type of offer 2 can be obtained according to above-mentioned logic Synthesis predilection grade, it is noted that these synthesis predilection grades be all be limited in user A in face of commodity B scene under calculate , the synthesis predilection grade highest of which type of offer indicates that user A is to the preferential class under the scene in user A in face of commodity B The preference highest of type, as optimal type of offer.

It can be seen that on the one hand attribute data that the present embodiment passes through user considers user itself for the inclined of type of offer It is good, the degree that is adapted to of product features and type of offer is on the other hand considered by the attribute data of commodity, is obtained in conjunction with two aspects The synthesis predilection grade of each type of offer under concrete scene, can be accurately by comparing the synthesis predilection grade of each type of offer Position the optimal type of offer under concrete scene.

Specifically, in embodiment of the disclosure, above-mentioned attribute data based on the user obtain the user about First predilection grade of the type of offer may include: to obtain the corresponding first History Order data of the user, and described the In one History Order data comprising meet the type of offer favor information order quantity be the first quantity, described first The quantity of order in History Order data comprising meeting the favor information of any type of offer is the second quantity；Obtain all use The corresponding second History Order data in family, comprising meeting the favor information of the type of offer in the second History Order data The quantity of order be third quantity, comprising meeting the favor information of any type of offer in the second History Order data The quantity of order is the 4th quantity；Ratio by the ratio of the first quantity and the second quantity relative to third quantity and the 4th quantity Accounting as first predilection grade.

The example continued to use above is illustrated, and above the attribute data based on user A obtains user A about preferential class The process of first predilection grade of type 1, which may is that, obtains the corresponding first History Order data of user A, first History Order Data contain the relevant data of the corresponding valid order of all history buying behaviors of user A, and counting user A corresponding first is gone through The quantity of order in history order data comprising meeting the favor information of type of offer 1 is the first quantity x1, and counting user A is corresponding The first History Order data in comprising meeting the excellent of any type of offer (type of offer 1, type of offer 2 or type of offer 3) The quantity of the order of favour information is the second quantity x2, calculates the ratio x1/x2 of the first quantity x1 and the second quantity x2, the ratio table Show that user A in history buying behavior, uses the ratio of all orders of the order and any type of offer of use of type of offer 1 Example.And the corresponding second History Order data of total user are obtained, total user is selected biggish use as needed Family sample set, such as can be all or part of registration user on electric business platform, which contains The relevant data of the corresponding valid order of each all history buying behaviors of user, corresponding second History Order of statistics total user The quantity of order in data comprising meeting the favor information of type of offer 1 is third quantity x3, and statistics total user is corresponding Comprising meeting the preferential of any type of offer (type of offer 1, type of offer 2 or type of offer 3) in second History Order data The quantity of the order of information is the 4th quantity x4, calculates the ratio x3/x4 of third quantity x3 and the 4th quantity x4, which indicates Total user is in history buying behavior, the ratio of the order using type of offer 1 and all orders using any type of offer Example.Accounting of the ratio calculated x1/x2 relative to ratio x3/x4, using the accounting as user A about type of offer 1 first partially Favorable comment point, first predilection grade indicate the phase in the case where not considering commodity is which only considers user's A unique characteristics attribute It under the scene of any commodity, on the basis of preference of the total user to type of offer 1, is calculated when in user A User A to the preference of type of offer 1.Similarly, user A can be calculated to the preference journey of other type of offer Degree, details are not described herein.

In the present embodiment, user's portrait is sketched the contours according to a large amount of user attribute data, the category of user in above-described embodiment Property data be user History Order data, the History Order data of user can accurately reflect user buying behavior practise It is used, and then therefrom can reflect out user itself for the preference of various type of offer.

Specifically, in embodiment of the disclosure, above-mentioned attribute data based on the commodity obtain the commodity about Second preference-score of the type of offer may include: to determine the commodity corresponding price point based on the prices of the commodity Area；The corresponding third History Order data of commodity in the Price Zoning are obtained, include in the third History Order data The quantity for meeting the order of the favor information of the type of offer is the 5th quantity, includes symbol in the third History Order data The quantity for closing the order of the favor information of any type of offer is the 6th quantity；Obtain corresponding 4th History Order of all commodity Data, the quantity of the order in the 4th History Order data comprising meeting the favor information of the type of offer are the 7th number It measures, the quantity of the order in the 4th History Order data comprising meeting the favor information of any type of offer is the 8th number Amount；Using the ratio of the 5th quantity and the 6th quantity relative to the 7th quantity and the accounting of the ratio of the 8th quantity as described second Predilection grade.

Still the example continued to use above is illustrated, and above the attribute data based on commodity B obtains commodity B about excellent The process of second preference-score of favour Class1, which may is that, divides multiple Price Zonings, all commodity according to the price of all commodity It is selected biggish commodity original set as needed, such as can be all or part of commodity on electric business platform.Base The corresponding Price Zoning of commodity B is determined in the price of commodity B, such as 100~200 yuan.Obtain the Price Zoning (100~200 yuan) In the corresponding third History Order data of commodity, which contains that total user is corresponding to fall into the valence The relevant data of all valid orders in lattice subregion (100~200 yuan) count in the third History Order data comprising meeting The quantity of the order of the favor information of type of offer 1 is the 5th quantity x5, is counted in the third History Order data comprising meeting The quantity of the order of the favor information of any type of offer (type of offer 1, type of offer 2 or type of offer 3) is the 6th quantity X6, calculate the 5th quantity x5 and the 6th quantity x6 ratio x5/x6, the ratio indicate total user in history buying behavior, The order for falling into the commodity of Price Zoning (100~200 yuan) is bought using type of offer 1 and is fallen using the purchase of any type of offer Enter the ratio of all orders of the commodity of Price Zoning (100~200 yuan).And obtain corresponding 4th history of all commodity Order data, the 4th History Order data contain the relevant data of the corresponding all valid orders of total user, and statistics should The quantity of order in 4th History Order data comprising meeting the favor information of type of offer 1 is the 7th quantity x7, and statistics should Comprising meeting the preferential of any type of offer (type of offer 1, type of offer 2 or type of offer 3) in 4th History Order data The quantity of the order of information is the 8th quantity x8, calculates the ratio x7/x8 of the 7th quantity x7 and the 8th quantity x8, which indicates Total user is in history buying behavior, the ratio of the order using type of offer 1 and all orders using any type of offer Example.Using the ratio of the 5th quantity and the 6th quantity relative to the 7th quantity and the accounting of the ratio of the 8th quantity as described second Predilection grade.Accounting of the ratio calculated x5/x6 relative to ratio x7/x8, using the accounting as commodity B about type of offer 1 Second predilection grade, second predilection grade indicate in the case where not considering user is the characteristic attribute who only considers commodity, On the basis of preference when facing all commodity by total user to type of offer, the total user being calculated is in face of commodity To the preference of type of offer 1 when B.Similarly, to other type of offer when can calculate total user in face of commodity B Preference, details are not described herein.

In the present embodiment, the attribute data of commodity includes pricing information and History Order data, since commodity price can Which kind of greatly influence the selection such as whether use favor information, use favor information when user buys commodity, therefore according to commodity Price commodity are divided in different Price Zonings, influence of the different Price Zonings to user preferential preference is different.Quotient The corresponding History Order data of Price Zoning where product can accurately reflect that commodity in each Price Zoning are bought by user Situation, and then therefrom can reflect out total user when in face of corresponding commodity for the preference of various type of offer.

It is specifically, in embodiment of the disclosure, above-mentioned to be based on first predilection grade and second predilection grade, It includes: to obtain the first predilection grade to correspond to that the user and the commodity, which are obtained, about the synthesis predilection grade of the type of offer The first weight, obtain corresponding second weight of the second predilection grade；Based on first weight and second weight to institute It states the first predilection grade and second predilection grade is weighted summation, obtain the comprehensive predilection grade.

Continue to use example above, obtain user A about type of offer 1 the first predilection grade and commodity B about excellent After second predilection grade of favour Class1, first predilection grade corresponding first weight of the user A about type of offer 1 is obtained A obtains second predilection grade corresponding second weight b of the commodity B about type of offer 1, since first predilection grade indicates Under scene of the user A in face of any commodity, user A is subjective factor to the preference of type of offer 1, then corresponding the One weight a characterization is influence degree brought by subjective factor, and the second predilection grade indicates to face commodity B in total user Scene under, total user is objective factor caused by commodity price to the preference of type of offer 1, then corresponding Two weight b characterization is influence degree brought by objective factor, therefore inclined to first based on the first weight a and the second weight b Favorable comment point and the second predilection grade are weighted summation and obtain the synthesis predilection grade of type of offer 1.

According to the present embodiment, when user A faces commodity B, subjective factor and objective factor are comprehensively considered to user preferential The synthesis predilection grade of each type of offer is calculated in the influence of preference, meets the agenda rule that user buys commodity, energy Enough accurately reflect comprehensive preferential preference.

It should be noted that influence degree brought by subjective factor may be different, right for the user of different purchasing power In the commodity of different prices, influence degree brought by objective factor may be different, therefore the feelings different in user's difference commodity Under scape, the first accessed weight and the second weight may be different, and need to take in as the case may be.

In one embodiment of the present disclosure, different purchases can be obtained by way of building model, training pattern parameter Buy corresponding second weight of commodity of corresponding first weight of power user and different Price Zonings.Specifically, for total user, Total user is selected biggish user's sample set as needed, such as can be all or part on electric business platform Register user, according to the purchasing power level of user by each user be divided to multiple objective realms not in.For all commodity, Quan Tishang Product are selected biggish commodity original set as needed, such as can be all or part of commodity on electric business platform, Each commodity are divided in multiple Price Zonings according to the price of commodity.And any Price Zoning other for any objective realm, structure Preferential preference pattern is built, the input of the preferential preference pattern includes the objective realm not about the first preference of each type of offer Scoring and second predilection grade of the Price Zoning about each type of offer, the output of the preferential preference pattern are the visitor Realm not highest comprehensive predilection grade corresponding with the Price Zoning, the parameter of the preferential preference pattern include the first power Weight parameter and the second weight parameter.The objective realm highest actual preference scoring not corresponding with the Price Zoning is obtained, It is that target optimizes the parameter of the preferential preference pattern with the highest actual preference scoring, optimization is obtained most The value of the first excellent weight parameter is as not corresponding first weight of visitor's realm, the second optimal power that optimization is obtained The value of weight parameter is as corresponding second weight of the Price Zoning.

Wherein, as an optional embodiment, the above-mentioned preferential preference pattern of building includes: to obtain to correspond to the objective group The attribute data of multiple sample of users of classification obtains the attribute data for corresponding to multiple sample commodity of the Price Zoning； The attribute data of attribute data and the multiple sample commodity based on the multiple sample of users；For any type of offer, Attribute data based on the multiple sample of users obtains the multiple sample of users and comments about the first preference of the type of offer Point, the sum of first predilection grade by the multiple sample of users about the type of offer is not excellent about this as the objective realm First predilection grade of favour type；Attribute data based on the multiple sample commodity obtains corresponding to the multiple sample commodity Second predilection grade of the Price Zoning about the type of offer；Based on the first weight parameter and the second weight parameter to institute State second preference of the objective realm not about the first predilection grade of the type of offer and the Price Zoning about the type of offer Scoring, which is weighted summation and obtains the objective realm not synthesis preference about the type of offer corresponding with the Price Zoning, to be commented Point.

As an optional embodiment, the above-mentioned acquisition objective realm not highest reality corresponding with the Price Zoning Border predilection grade includes: to obtain any sample of users and the Price Zoning the corresponding 5th for any type of offer The quantity of History Order data, the order in the 5th History Order data comprising meeting the favor information of the type of offer is 9th quantity, the quantity of the order in the 5th History Order data comprising meeting the favor information of any type of offer are the Ten quantity；The ratio accounting of 9th quantity and the tenth quantity is preferential about this as the sample of users and the Price Zoning The actual preference of type scores, and the multiple sample of users and the Price Zoning are commented about the actual preference of the type of offer / and as the actual preference scoring that the objective realm is other and the Price Zoning is about the type of offer.Compare the objective group Classification and the Price Zoning are scored about the actual preference of each type of offer, do not obtain the objective realm not and the Price Zoning Corresponding highest actual preference scoring.

Then for above user A face commodity B scene, obtain about type of offer 1 the first predilection grade and After second predilection grade, determines that the corresponding objective realm of user A is other according to the attribute data of user A, determined according to the price of commodity B The corresponding Price Zoning of commodity B.The corresponding objective realm not corresponding first of user A is obtained based on model training process above Weight and corresponding second weight of the corresponding Price Zoning of commodity B, based on acquired the first weight and the second weight to One predilection grade and the second predilection grade are weighted the corresponding comprehensive predilection grade of the available type of offer 1 of summation.

It can be seen that the limitation in the above process by model learning, objective function can be quickly obtained and meet specific field The first weight and the second weight of scape demand, wherein the first weight is associated with the purchasing power of user, the second weight and commodity Price is associated.

In a specific embodiment, the above-mentioned attribute data according to user A determines that the corresponding objective realm of user A is not wrapped It includes: based on average income data, average order price and/or the average consumption amount of money of the user A in preset period of time, determining The corresponding objective realm of user A is other.

In another embodiment of the disclosure, if active user be new user, corresponding to attribute data very little, When can not know the purchasing power level of the user according to the big data of its history buying behavior, i.e., it can not be instructed based on above-mentioned model It, can be using preset numerical value as the first weight and the second weight, specifically when experienced mode obtains the first weight and the second weight Ground, corresponding first weight of above-mentioned the first predilection grade of acquisition obtain corresponding second weight of the second predilection grade further include: when When can not determine that the corresponding visitor's realm of the user is other, using the first preset value as first weight, the second preset value is made For second weight.The present embodiment is in view of no history buying behavior data or the less user of history buying behavior data Scene when carrying out preferential decision, avoids the occurrence of exception.

Below with reference to Fig. 3 A~Fig. 3 D, method shown in Fig. 2 is described further in conjunction with specific embodiments.

Fig. 3 A diagrammatically illustrates the data flow diagram of the commercial articles searching process according to the embodiment of the present disclosure.

As shown in Figure 3A, the searching request that user u is directed to commodity B is received, the search result of commodity B, the search knot are obtained Include the relevant information of commodity B in fruit, then the face user u is determined based on the attribute data of user u and the attribute data of commodity B Optimal type of offer when to commodity B, and the preferential letter of the optimal type of offer will be met when exporting the search result of commodity B Breath.

Default in the present embodiment there are three types of type of offer altogether: free of interest is preferential, lands vertically preferential and completely subtracts preferential.Getting quotient It is preferential for free of interest after the search result of product B, calculate the first user u preferential about free of interest predilection grade mx_prefer_u, Calculate the price range y second predilection grade mx_prefer preferential about the free of interest where commodity B_y；It is preferential for landing vertically, User u is calculated to land vertically about this first preferential predilection grade zj_prefer_u, calculate commodity B where price range y about The second preferential predilection grade zj_prefer that lands vertically_y；It is preferential for completely subtracting, it calculates user u and completely subtracts preferential first about this Predilection grade mj_prefer_u, calculate commodity B where price range y completely subtract the second preferential predilection grade mj_ about this prefer_y。

Illustrate the detailed process for calculating the first predilection grade and the second predilection grade, other preferential classes by taking free of interest is preferential as an example Type similarly, repeats no more.

The user u first predilection grade mx_prefer preferential about the free of interest is calculated by following formula_u:

Wherein, mx_ords_uThe quantity of the preferential order of free of interest, coupon-ords are used for user u_uIt is used for user u excellent The quantity of the order of favour, mx-prefer_u' it is that the order that user u uses free of interest preferential uses preferential all orders with user u Specific gravity, referred to as user u free of interest order specific gravity, mx_ords is quantity of the total user using the preferential order of free of interest, Coupon-ords is the quantity that total user uses preferential order, and mx_prefer ' is that total user is preferential using free of interest Order and total user use the specific gravity of preferential all orders, referred to as total user free of interest order specific gravity, then user u is about exempting from Cease the first preferential predilection grade mx_prefer_uUser u free of interest order specific gravity is defined as relative to total user free of interest order ratio The promotion of weight.

It can be seen that relying on electric business platform mass users consumer behavior data, abnormal data is excluded by algorithm process.It is logical Calculating total user is crossed to the service condition of three kinds of different type of offer (land vertically preferential, completely subtract that preferential, free of interest is preferential), can be obtained The preferential preference profile of whole user out can similarly calculate the preferential preference profile of individual of each user.It is personal by calculating Preference can identify user to the first predilection grade of different type of offer, the first preference is commented relative to whole promotion situation It is point higher that represent user itself higher to the preference of the type of offer.

Pass through the second predilection grade mx_ preferential about the free of interest of the price range y where following formula calculating commodity B prefer_y:

Wherein, mx_ords_y, for the quantity in price range y using the preferential order of free of interest, coupon_ords_yFor in valence Lattice section y uses the quantity of preferential order, mx_prefer_u' to use interest-free preferential order in price range y and in valence Lattice section y uses the specific gravity of preferential all orders, the interest-free order specific gravity of referred to as price range y, and mx_prefer ' is entirety User uses the specific gravity of preferential all orders using the preferential order of free of interest and total user, also corresponds to full price and exempts from for section Order specific gravity is ceased, then the second price range y where commodity B preferential about free of interest predilection grade mx_prefer_yIt is defined as The interest-free order specific gravity of price range y is the promotion of section free of interest order specific gravity relative to full price.

Based on electric business platform mass data, can extract in different price ranges each type of offer (land vertically it is preferential, completely subtract It is preferential, free of interest it is preferential) service condition can measure price factor pair by dividing the comparison of price range Yu full price range The influence degree of user preferential selection.

After corresponding first predilection grade of each type of offer and the second predilection grade is calculated, need to obtain corresponding The first weight and the second weight, to combine the first predilection grade and the second predilection grade to calculate the synthesis preference of each type of offer Scoring.

The present embodiment obtains the first weight and the second weight in the following manner:

Objective heap sort is carried out first, and according to IPM model, (wherein I represents income, P represents order average unit price, M representative disappears Take the amount of money) total user is divided into purchasing power level by other to weak 6 objective realms by force.

Fig. 3 B diagrammatically illustrates the schematic diagram of the IPM model according to the embodiment of the present disclosure.

As shown in Figure 3B, three-dimensional cartesian coordinate system is constituted by I axis, P axis and M axis, as I indicates the income of user's every month, P Indicate the average unit price for the order bought user's every month, M indicates the overall consumption amount of money of user's every month, according to user's every month Income, every month purchase order average unit price and the overall consumption amount of money of every month judge the purchasing power water of the user Flat, when corresponding I, P and M are higher, the purchasing power level of user is higher, when corresponding I, P and M are lower, user's Purchasing power level is lower, and specific different objective realm another characteristics are as shown in table 1:

Table 1

Wherein, the height judgment criteria of each index can be set as needed, and is such as set as the 25% of the index, works as height Be determined as height when the 25% of the index, when lower than the index 25% be determined as it is low.

After passerby's heap sort, it is known that different user group is selected in the selection of coupon type by user The influence of (subjectivity) and two aspect factor of price (objective) has differences, for high purchasing power crowd, when buying commodity The preferential preference showed should mainly be influenced by subjective factor, and it is excellent to influence its for low purchasing power crowd, when buying commodity The leading factor of favour selection should be price factor.

Each user is finally calculated in different price by the different weight of design subjective and objective factor for different objective groups The preferential preference of lattice, it is as follows to construct preferential preference pattern:

Wherein, k indicates that objective realm is other, and as other in shared 6 kinds of objective realms in table 1, r indicates type of offer, if r=1 is free of interest It is preferential, r=2 be land vertically it is preferential, r=3 be completely subtract it is preferential, i indicate current objective realm not in i-th of user, current visitor's realm N user, x are shared in not_r[i] indicates first predilection grade of i-th of user about type of offer r, y_r[j] is indicated in price Second predilection grade of the commodity about type of offer r in subregion j, a_k×x_r[i]+b_k×y_r[j] indicates that current objective realm is other In first predilection grade of i-th of user about type of offer r, pass through the first weight parameter a_kWith the second weight parameter b_kIt is right Current visitor's realm not in i-th of user about type of offer r the first predilection grade and Price Zoning j about type of offer r The second predilection grade be weighted summation, be calculated current objective realm not in i-th of user for Price Zoning j about The synthesis predilection grade of type of offer r, max (a_k×x_r[i]+b_k×y_r[j]) be i-th of user for Price Zoning j about each The maximum value of the synthesis predilection grade of type of offer,Indicating will current visitor Realm not in each user sum for Price Zoning j about the maximum value of the synthesis predilection grade of each type of offer, it is defeated It is out the maximum value of the corresponding comprehensive predilection grade of each type of offer, i.e., current objective realm is not for corresponding to Price Zoning j The comprehensive predilection grade of maximum.

Above-mentioned preferential preference pattern is optimized, so that meeting following constraint condition:

Wherein, z_r[i, j] is z₁[i, j], z₂[i, j] and z₃Maximum value in [i, j], z₁[i, j] is that current objective realm is other In i-th of user score for Price Zoning j about the preferential actual preference of free of interest, for example, actual preference scoring can be with I-th of user of quantity and this of order equal to i-th of user using the preferential purchase of free of interest in Price Zoning j uses preferential Buy the ratio of the quantity of the order in Price Zoning j.z₂[i, j] be current objective realm not in i-th of user for valence Lattice subregion j is about the actual preference scoring for landing vertically preferential, z₃[i, j] be current objective realm not in i-th of user for price Subregion j is about preferential actual preference scoring is completely subtracted, with z₁Similarly details are not described herein for [i, j] calculating process, Indicate by current objective realm not in each user score and carry out about the maximum actual preference of type of offer r for Price Zoning j Summation obtains current objective realm and does not score for Price Zoning j about the maximum actual preference of type of offer r.On it is recognised that Stating optimization process is the maximum actual preference scoring conduct by current objective realm not for Price Zoning j about type of offer r Optimization aim enables the calculated current objective realm of preferential preference pattern institute maximum comprehensive preference not corresponding for Price Zoning j Scoring is constantly close to maximum actual preference scoring, the first weight parameter a that final optimization pass is obtained_kWith the second weight parameter b_kMake The first weight and the second weight when the commodity in Price Zoning j are not faced for current objective realm.In more simplified process, valence Lattice subregion j may correspond to the complete section of price, and such first weight and the second weight be not only associated with objective realm.

Therefore, when to obtain the first weight and the second weight of hereinbefore user u for the commodity B in Price Zoning y, First determine user u where objective realm it is other, by above-mentioned model optimization process obtain visitor's realm not for Price Zoning y when First weight a_kWith the second weight b_k。

Then user u highest comprehensive predilection grade corresponding for Price Zoning y is calculated by the following formula:

prefer_uy=max { zj_prefer_uy, mj_prefer_uy, mx_prefer_uy}=max { a_k×zj_prefer_u+b_k ×zj_prefer_y, a_k×mj_prefer_u+b_k×mj_prefer_y, a_k×mx_prefer_u+b_k×mx_prefer_y}

Wherein, zj_prefer_uyFor user u for Price Zoning y about the synthesis predilection grade for landing vertically preferential, mj_ prefer_uyFor user u for Price Zoning y about completely subtracting preferential synthesis predilection grade, mx_prefer_uyFor user u for Synthesis predilection grade Price Zoning y preferential about free of interest.Highest comprehensive predilection grade prefer_uyCorresponding type of offer is Identified optimal type of offer when for user u for commodity B in Price Zoning y, by the search result of commodity B with meeting this The favor information of optimal type of offer exports displaying together.

Fig. 3 C diagrammatically illustrates the schematic diagram of the search result of the commodity according to the embodiment of the present disclosure.

As shown in Figure 3 C, identified optimal preferential when user u is for commodity B in Price Zoning y in the figure on the left side Type is to land vertically preferential, then includes that " vertical subtract 8 yuan " printed words meet the preferential information that lands vertically in the search result of commodity B；? In intermediate figure, when user u is for commodity B in Price Zoning y identified optimal type of offer be completely subtract it is preferential, then in quotient It include that " full 300 subtract 10 " completely subtracts preferential information meeting for printed words in the search result of product B；In figure on the right, user u for Identified optimal type of offer is that free of interest is preferential when commodity B in Price Zoning y, then includes in the search result of commodity B " enjoying 12 phases, 6 foldings of breath expense " printed words meet the preferential information of free of interest.

User is new user in special circumstances a kind of, corresponding to historical data without or it is less, can not be by upper The method for stating model training obtains its corresponding first weight and the second weight, then in addition setting the 7th objective realm is other, for this Objective realm is other, and presetting the first weight is 0, and the second weight is 1, which can be adjusted according to the actual situation, herein not It is limited.Optimal preferential class of new user when in face of commodity can be determined using preset first weight and the second weight Type carries out the output and displaying of corresponding favor information.

Fig. 3 D diagrammatically illustrates the flow chart of the product search method according to another embodiment of the disclosure.

As shown in Figure 3D, this method includes operation S301~S308, specific as follows:

In operation S301, user login information is obtained, obtains merchandise news.

In operation S302, judge whether user is new user, is to execute operation S303, otherwise executes operation S304.

In operation S303, determines that user belongs to objective realm other 7, obtain preset first weight a₇With the second weight b₇。

In operation S304, identifies the other k of objective realm belonging to user, it is k pairs other which is obtained based on preferential preference pattern The the first weight a answered_kWith the second weight b_k。

In operation S305, first predilection grade of the user about each type of offer is calculated.

In second predilection grade of the Price Zoning where operation S306, calculating commodity about each type of offer.

In operation S307, the synthesis for being merged to obtain each type of offer to the first predilection grade and the second predilection grade is inclined Favorable comment point, is ranked up the synthesis predilection grade of each type of offer, using the comprehensive highest type of offer of predilection grade as most Excellent type of offer.

In operation S308, the search result of the favor information for meeting optimal type of offer and commodity is opened up in front end Show.

This programme constructs preferential preference mould by collecting user's history consumer record and preferential service condition related data Type portrays user preference, and operation system shows user's current commodity by transferring model result, in the auspicious page of quotient with advertisement word form The preferential form of preference the most realizes " thousand people, thousand face " (same commodity different user shows that favor information is different) and " people is more Face " (same user's difference commodity show that favor information is different).Wherein, electric business mass data is relied on, it can be in Beijing with calling and obtaining user All consumer records and discount coupon service condition in eastern store, while will be seen that various types of on different items The service condition of discount coupon, above data are the data that real trade generates, have the characteristics that it is authentic and valid, based on above several According to, can purchasing power, preferential preference to user measure.

Since (History Order data, the history of commodity of such as user is ordered for the attribute data and the attribute datas of commodity of user Forms data) it is middle in the presence of a large amount of the case where cancelling the order, the discount coupon service condition of this part order finally can be remembered by system Record, it is therefore desirable to reject this influence of partial invalidity order to result first.And due to meeting in the consumption order of electric business platform There is super Large Order Number (businessman's test) or extra small volume order (second kills), while will appear the order volume in certain customers' short time Considerably beyond integral level (brush is single), this part order can produce serious influence to result, therefore extract disappearing for user When taking record, eliminating the order amount of money is more than whole 95% horizontal and order of the order amount of money lower than whole 5%, while also being excluded Order volume is more than the user of 95% or more integral level, to guarantee the validity of above-mentioned calculating process.

The disclosure is by big data and financial cloud computing technology, by the analysis to user's history consumer behavior, respectively from Two dimensions of product (commodity) and user, in conjunction with objective group character, the discount coupon differentiation algorithm of designing user rank portrays user In the preferential preference of different items.By the algorithm, differentiation price can be implemented to user: 1) realized in same part quotient Show the profit point (type of offer) of its most preference on product in the information page of commodity to different user；2) it realizes and exists to same user The profit point (type of offer) of its most preference is shown on the information page of different commodity.The disclosure mainly solves the problems, such as 3: from It is seen in this control, by preferential preference difference, optimizes marketing resource dispensing, avoid marketing resource mistake from throwing, to reach section The purpose of province's expense；User experience can be optimized from the point of view of user, provide its most desirable preferential class in user's shopping Type promotes user's conversion ratio, realizes " thousand people, thousand face " and " people's multi-panel "；User combines with two dimensions of commodity, solves The problem of new visitor can not carry out preferential preference difference because that can not obtain historical consumption data.

Fig. 4 diagrammatically illustrates the block diagram of the commercial articles searching device according to the embodiment of the present disclosure.

As shown in figure 4, commercial articles searching device 400 includes receiving module 410, obtains module 420, decision-making module 430 and defeated Module 440 out.

Receiving module 410 is used to receive the searching request that user is directed to commodity.

Module 420 is obtained for obtaining search result corresponding with described search request, described search result includes institute State the relevant information of commodity.

Decision-making module 430 is used for the attribute data of attribute data and the commodity based on the user, determines optimal excellent Favour type.

Output module 440 is used to export described search knot in association with the favor information for meeting the optimal type of offer Fruit.

In accordance with an embodiment of the present disclosure, the attribute of attribute data and the commodity of the decision-making module 430 based on the user Data determine that optimal type of offer includes: that decision-making module 430 is used for for any type of offer, the attribute based on the user Data obtain first predilection grade of the user about the type of offer；Described in attribute data based on the commodity obtains Second predilection grade of the commodity about the type of offer；Based on first predilection grade and second predilection grade, obtain Synthesis predilection grade to the user and the commodity about the type of offer；It will the comprehensive highest preferential class of predilection grade Type is as optimal type of offer.

In accordance with an embodiment of the present disclosure, decision-making module 430 based on the attribute data of the user obtain the user about First predilection grade of the type of offer includes: decision-making module 430 for obtaining corresponding first History Order of the user Data, the quantity of the order in the first History Order data comprising meeting the favor information of the type of offer are the first number It measures, the quantity of the order in the first History Order data comprising meeting the favor information of any type of offer is the second number Amount；Obtain the corresponding second History Order data of total user, it is described preferential comprising meeting in the second History Order data The quantity of the order of the favor information of type is third quantity, includes to meet any preferential class in the second History Order data The quantity of the order of the favor information of type is the 4th quantity；By the ratio of the first quantity and the second quantity relative to third quantity with The accounting of the ratio of 4th quantity is as first predilection grade.

In accordance with an embodiment of the present disclosure, decision-making module 430 based on the attribute data of the commodity obtain the commodity about Second preference-score of the type of offer includes: that decision-making module 430 is used to determine the commodity based on the price of the commodity Corresponding Price Zoning；The corresponding third History Order data of commodity in the Price Zoning are obtained, the third history is ordered The quantity of order in forms data comprising meeting the favor information of the type of offer is the 5th quantity, the third History Order The quantity of order in data comprising meeting the favor information of any type of offer is the 6th quantity；It is corresponding to obtain all commodity 4th History Order data, comprising meeting the order of the favor information of the type of offer in the 4th History Order data Quantity is the 7th quantity, includes the number for meeting the order of favor information of any type of offer in the 4th History Order data Amount is the 8th quantity；The ratio of 5th quantity and the 6th quantity is made relative to the 7th quantity and the accounting of the ratio of the 8th quantity For second predilection grade.

In accordance with an embodiment of the present disclosure, decision-making module 430 is based on first predilection grade and second predilection grade, It includes: decision-making module 430 for obtaining that the user and the commodity, which are obtained, about the synthesis predilection grade of the type of offer Corresponding first weight of first predilection grade obtains corresponding second weight of the second predilection grade；Based on first weight and Second weight is weighted summation to first predilection grade and second predilection grade, obtains the comprehensive preference Scoring.

In accordance with an embodiment of the present disclosure, decision-making module 430 obtains corresponding first weight of the first predilection grade, obtains second Corresponding second weight of predilection grade includes: that decision-making module 430 is used to determine the user according to the attribute data of the user Corresponding visitor's realm is other, and the different purchasing power of different objective group's categorized representations is horizontal；According to the determination of the price of the commodity The corresponding Price Zoning of commodity；Preferential preference pattern is constructed, the input of the preferential preference pattern includes that the objective realm is not closed It is described preferential in the second predilection grade about each type of offer of the first predilection grade and the Price Zoning of each type of offer The output of preference pattern is the objective realm not highest comprehensive predilection grade corresponding with the Price Zoning, described preferential inclined The parameter of good model includes the first weight parameter and the second weight parameter；It is not corresponding with the Price Zoning to obtain the objective realm Highest actual preference scoring；With the highest actual preference scoring be target to the parameter of the preferential preference pattern into Row optimization, using the value of the first optimal weight parameter as first weight, by the value of the second optimal weight parameter As second weight.

In accordance with an embodiment of the present disclosure, decision-making module 430 determines that the user is corresponding according to the attribute data of the user Objective realm do not include: that decision-making module 430 is used for based on average income data of the user in preset period of time, average Order price and/or the average consumption amount of money determine that the corresponding objective realm of the user is other.

In accordance with an embodiment of the present disclosure, it includes: decision-making module 430 for obtaining that decision-making module 430, which constructs preferential preference pattern, The attribute data for corresponding to the other multiple sample of users of the objective realm is taken, the multiple samples for corresponding to the Price Zoning are obtained The attribute data of commodity；The attribute data of attribute data and the multiple sample commodity based on the multiple sample of users；It is right In any type of offer, the attribute data based on the multiple sample of users obtains the multiple sample of users about the preferential class First predilection grade of type, the sum of first predilection grade by the multiple sample of users about the type of offer are used as the visitor Realm is not about the first predilection grade of the type of offer；Attribute data based on the multiple sample commodity obtains the multiple Second predilection grade of the Price Zoning corresponding to sample commodity about the type of offer；Based on the first weight parameter and Two weight parameters are not preferential about this about the first predilection grade of the type of offer and the Price Zoning to the objective realm Second predilection grade of type is weighted summation, and to obtain the objective realm not corresponding with the Price Zoning about the preferential class The synthesis predilection grade of type.

In accordance with an embodiment of the present disclosure, decision-making module 430 obtain the objective realm not it is corresponding with the Price Zoning most High actual preference scoring includes: that decision-making module 430 is used to obtain any sample of users and institute for any type of offer The corresponding 5th History Order data of Price Zoning are stated, include to meet the excellent of the type of offer in the 5th History Order data The quantity of the order of favour information is the 9th quantity, includes to meet the preferential of any type of offer in the 5th History Order data The quantity of the order of information is the tenth quantity；Using the ratio accounting of the 5th quantity and the 6th quantity as the sample of users and institute The actual preference that Price Zoning is stated about the type of offer scores, by the multiple sample of users and the Price Zoning about this The sum of actual preference scoring of type of offer as the objective realm not with reality of the Price Zoning about the type of offer Predilection grade；Compare the objective realm not score with the Price Zoning about the actual preference of each type of offer, obtain described The highest actual preference scoring not corresponding with the Price Zoning of objective realm.

In accordance with an embodiment of the present disclosure, decision-making module 430 obtains corresponding first weight of the first predilection grade, obtains second Corresponding second weight of predilection grade further include: decision-making module 430, which is also used to work as, can not determine the corresponding objective realm of the user When other, using the first preset value as first weight, using the second preset value as second weight.

It should be noted that commercial articles searching device 400 shown in Fig. 4 can be configured at server side, can also be configured at Client-side.And, it should be noted that the embodiment of each module/unit/subelement etc., solution in device section Example Certainly the technical issues of, the function of realization and the technical effect that reaches respectively with corresponding step each in method section Example Embodiment, the technical issues of solving, the function of realization and the technical effect that reaches it is same or like, it is no longer superfluous herein It states.

It is module according to an embodiment of the present disclosure, submodule, unit, any number of or in which any more in subelement A at least partly function can be realized in a module.It is single according to the module of the embodiment of the present disclosure, submodule, unit, son Any one or more in member can be split into multiple modules to realize.According to the module of the embodiment of the present disclosure, submodule, Any one or more in unit, subelement can at least be implemented partly as hardware circuit, such as field programmable gate Array (FPGA), programmable logic array (PLA), system on chip, the system on substrate, the system in encapsulation, dedicated integrated electricity Road (ASIC), or can be by the hardware or firmware for any other rational method for integrate or encapsulate to circuit come real Show, or with any one in three kinds of software, hardware and firmware implementations or with wherein any several appropriately combined next reality It is existing.Alternatively, can be at least by part according to one or more of the module of the embodiment of the present disclosure, submodule, unit, subelement Ground is embodied as computer program module, when the computer program module is run, can execute corresponding function.

For example, receiving module 410, obtain module 420, any number of in decision-making module 430 and output module 440 can It is realized in a module with merging or any one module therein can be split into multiple modules.Alternatively, these moulds At least partly function of one or more modules in block can be combined at least partly function of other modules, and at one It is realized in module.In accordance with an embodiment of the present disclosure, receiving module 410, acquisition module 420, decision-making module 430 and output module At least one of 440 can at least be implemented partly as hardware circuit, such as field programmable gate array (FPGA), can compile Journey logic array (PLA), system on chip, the system on substrate, the system in encapsulation, specific integrated circuit (ASIC), or can be with It is realized by carrying out the hardware such as any other rational method that is integrated or encapsulating or firmware to circuit, or with software, hardware And it any one in three kinds of implementations of firmware or several appropriately combined is realized with wherein any.Alternatively, receiving module 410, obtaining at least one of module 420, decision-making module 430 and output module 440 can at least be implemented partly as counting Calculation machine program module can execute corresponding function when the computer program module is run.

Fig. 5 is diagrammatically illustrated according to the computer equipment for being adapted for carrying out method as described above of the embodiment of the present disclosure Block diagram.Computer equipment shown in Fig. 5 is only an example, should not function to the embodiment of the present disclosure and use scope bring Any restrictions.

As shown in figure 5, include processor 501 according to the computer equipment 500 of the embodiment of the present disclosure, it can be according to storage It is loaded into random access storage device (RAM) 503 in the program in read-only memory (ROM) 502 or from storage section 508 Program and execute various movements appropriate and processing.Processor 501 for example may include general purpose microprocessor (such as CPU), refer to Enable set processor and/or related chip group and/or special microprocessor (for example, specific integrated circuit (ASIC)), etc..Processing Device 501 can also include the onboard storage device for caching purposes.Processor 501 may include for executing according to disclosure reality Apply single treatment unit either multiple processing units of the different movements of the method flow of example.

In RAM 503, it is stored with computer equipment 500 and operates required various programs and data.Processor 501, ROM 502 and RAM 503 is connected with each other by bus 504.Processor 501 is by executing the journey in ROM 502 and/or RAM 503 Sequence executes the various operations of the method flow according to the embodiment of the present disclosure.It is being removed it is noted that described program also can store In one or more memories other than ROM 502 and RAM 503.Processor 501 can also be stored in described one by executing Program in a or multiple memories executes the various operations of the method flow according to the embodiment of the present disclosure.

In accordance with an embodiment of the present disclosure, computer equipment 500 can also include input/output (I/O) interface 505, input/ Output (I/O) interface 505 is also connected to bus 504.Computer equipment 500 can also be including being connected to the following of I/O interface 505 It is one or more in component: the importation 506 including keyboard, mouse etc.；Including such as cathode-ray tube (CRT), liquid crystal The output par, c 507 of display (LCD) etc. and loudspeaker etc.；Storage section 508 including hard disk etc.；And including such as The communications portion 509 of the network interface card of LAN card, modem etc..Communications portion 509 is held via the network of such as internet Row communication process.Driver 510 is also connected to I/O interface 505 as needed.Detachable media 511, such as disk, CD, magnetic CD, semiconductor memory etc. are mounted on as needed on driver 510, in order to from the computer program read thereon It is mounted into storage section 508 as needed.

In accordance with an embodiment of the present disclosure, computer software journey may be implemented as according to the method flow of the embodiment of the present disclosure Sequence.For example, embodiment of the disclosure includes a kind of computer program product comprising carry meter on a computer-readable medium Calculation machine program, the computer program include the program code for method shown in execution flow chart.In such embodiments, The computer program can be downloaded and installed from network by communications portion 509, and/or be pacified from detachable media 511 Dress.When the computer program is executed by processor 501, the above-mentioned function of limiting in the system of the embodiment of the present disclosure is executed.Root According to embodiment of the disclosure, system as described above, unit, module, unit etc. can by computer program module come It realizes.

The disclosure additionally provides a kind of computer readable storage medium, which can be above-mentioned reality It applies included in equipment/device/system described in example；Be also possible to individualism, and without be incorporated the equipment/device/ In system.Above-mentioned computer readable storage medium carries one or more program, when said one or multiple program quilts When execution, the method according to the embodiment of the present disclosure is realized.

In accordance with an embodiment of the present disclosure, computer readable storage medium can be non-volatile computer-readable storage medium Matter, such as can include but is not limited to: portable computer diskette, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM or flash memory), portable compact disc read-only memory (CD-ROM), light Memory device, magnetic memory device or above-mentioned any appropriate combination.In the disclosure, computer readable storage medium can With to be any include or the tangible medium of storage program, the program can be commanded execution system, device or device use or Person is in connection.For example, in accordance with an embodiment of the present disclosure, computer-readable medium may include above-described ROM One or more memories other than 502 and/or RAM 503 and/or ROM 502 and RAM 503.

Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the disclosure, method and computer journey The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation A part of one module, program segment or code of table, a part of above-mentioned module, program segment or code include one or more Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants It is noted that the combination of each box in block diagram or flow chart and the box in block diagram or flow chart, can use and execute rule The dedicated hardware based systems of fixed functions or operations is realized, or can use the group of specialized hardware and computer instruction It closes to realize.

It will be understood by those skilled in the art that the feature recorded in each embodiment and/or claim of the disclosure can To carry out multiple combinations and/or combination, even if such combination or combination are not expressly recited in the disclosure.Particularly, exist In the case where not departing from disclosure spirit or teaching, the feature recorded in each embodiment and/or claim of the disclosure can To carry out multiple combinations and/or combination.All these combinations and/or combination each fall within the scope of the present disclosure.

Embodiment of the disclosure is described above.But the purpose that these embodiments are merely to illustrate that, and It is not intended to limit the scope of the present disclosure.Although respectively describing each embodiment above, but it is not intended that each reality Use cannot be advantageously combined by applying the measure in example.The scope of the present disclosure is defined by the appended claims and the equivalents thereof.It does not take off From the scope of the present disclosure, those skilled in the art can make a variety of alternatives and modifications, these alternatives and modifications should all fall in this Within scope of disclosure.

Claims

1. a kind of product search method, comprising:

Receive the searching request that user is directed to commodity；

Search result corresponding with described search request is obtained, described search result includes the relevant information of the commodity；

The attribute data of attribute data and the commodity based on the user, determines optimal type of offer；

Described search result is exported in association with the favor information for meeting the optimal type of offer.

2. according to the method described in claim 1, wherein, the attribute of the attribute data based on the user and the commodity Data determine that optimal type of offer includes:

For any type of offer, the attribute data based on the user obtains first of the user about the type of offer Predilection grade；

Attribute data based on the commodity obtains second predilection grade of the commodity about the type of offer；

Based on first predilection grade and second predilection grade, the user and the commodity are obtained about described preferential The synthesis predilection grade of type；

By the comprehensive highest type of offer of predilection grade as optimal type of offer.

3. according to the method described in claim 2, wherein, the attribute data based on the user obtain the user about First predilection grade of the type of offer includes:

Obtain the corresponding first History Order data of the user, it is described preferential comprising meeting in the first History Order data The quantity of the order of the favor information of type is the first quantity, includes to meet any preferential class in the first History Order data The quantity of the order of the favor information of type is the second quantity；

Obtain the corresponding second History Order data of total user, it is described preferential comprising meeting in the second History Order data The quantity of the order of the favor information of type is third quantity, includes to meet any preferential class in the second History Order data The quantity of the order of the favor information of type is the 4th quantity；

Using the ratio of the first quantity and the second quantity relative to third quantity and the accounting of the ratio of the 4th quantity as described the One predilection grade.

4. according to the method described in claim 2, wherein, the attribute data based on the commodity obtain the commodity about Second preference-score of the type of offer includes:

The corresponding Price Zoning of the commodity is determined based on the price of the commodity；

The corresponding third History Order data of commodity in the Price Zoning are obtained, include in the third History Order data The quantity for meeting the order of the favor information of the type of offer is the 5th quantity, includes symbol in the third History Order data The quantity for closing the order of the favor information of any type of offer is the 6th quantity；

Obtain the corresponding 4th History Order data of all commodity, it is described preferential comprising meeting in the 4th History Order data The quantity of the order of the favor information of type is the 7th quantity, includes to meet any preferential class in the 4th History Order data The quantity of the order of the favor information of type is the 8th quantity；

Using the ratio of the 5th quantity and the 6th quantity relative to the 7th quantity and the accounting of the ratio of the 8th quantity as described the Two predilection grades.

5. described to be commented based on first predilection grade and second preference according to the method described in claim 2, wherein Point, the synthesis predilection grade for obtaining the user and the commodity about the type of offer includes:

Corresponding first weight of the first predilection grade is obtained, corresponding second weight of the second predilection grade is obtained；

First predilection grade and second predilection grade are added based on first weight and second weight Power summation, obtains the comprehensive predilection grade.

6. according to the method described in claim 5, wherein, corresponding first weight of the first predilection grade of the acquisition obtains the Corresponding second weight of two predilection grades includes:

Determine that the corresponding objective realm of the user is other according to the attribute data of the user, different objective group's categorized representations is different Purchasing power is horizontal；

The corresponding Price Zoning of the commodity is determined according to the price of the commodity；

Construct preferential preference pattern, the input of the preferential preference pattern includes the of the objective realm not about each type of offer The second predilection grade of one predilection grade and the Price Zoning about each type of offer, the output of the preferential preference pattern are Visitor's realm not highest comprehensive predilection grade corresponding with the Price Zoning, the parameter of the preferential preference pattern include First weight parameter and the second weight parameter；

Obtain the objective realm highest actual preference scoring not corresponding with the Price Zoning；

It is that target optimizes the parameter of the preferential preference pattern with the highest actual preference scoring, by optimal the The value of one weight parameter is as first weight, using the value of the second optimal weight parameter as second weight.

7. according to the method described in claim 6, wherein, the attribute data according to the user determines that the user is corresponding Objective realm do not include:

Average income data, average order price and/or the average consumption amount of money based on the user in preset period of time, Determine that the corresponding objective realm of the user is other.

8. according to the method described in claim 6, wherein, the preferential preference pattern of building includes:

The attribute data for corresponding to the other multiple sample of users of the objective realm is obtained, obtains and corresponds to the more of the Price Zoning The attribute data of a sample commodity；

The attribute data of attribute data and the multiple sample commodity based on the multiple sample of users；

For any type of offer, the attribute data based on the multiple sample of users obtains the multiple sample of users about this First predilection grade of type of offer, the conduct of the sum of first predilection grade by the multiple sample of users about the type of offer Visitor's realm is not about the first predilection grade of the type of offer；

Attribute data based on the multiple sample commodity obtains the Price Zoning corresponding to the multiple sample commodity and closes In the second predilection grade of the type of offer；

Based on the first weight parameter and the second weight parameter to the objective realm not about the first predilection grade of the type of offer About the second predilection grade of the type of offer being weighted summation with the Price Zoning, to obtain the objective realm not and described The corresponding synthesis predilection grade about the type of offer of Price Zoning.

9. according to the method described in claim 8, wherein, it is described obtain the objective realm not it is corresponding with the Price Zoning most High actual preference scores

For any type of offer, any sample of users and the corresponding 5th History Order number of the Price Zoning are obtained According to, the quantity of the order in the 5th History Order data comprising meeting the favor information of the type of offer is the 9th quantity, The quantity of order in the 5th History Order data comprising meeting the favor information of any type of offer is the tenth quantity；

Using the ratio accounting of the 9th quantity and the tenth quantity as the sample of users and the Price Zoning about the preferential class The actual preference of type scores, and the multiple sample of users and the Price Zoning are scored about the actual preference of the type of offer The sum of do not score with the Price Zoning about the actual preference of the type of offer as the objective realm；

Compare the objective realm not score with the Price Zoning about the actual preference of each type of offer, obtains the objective realm Highest actual preference scoring not corresponding with the Price Zoning.

10. according to the method described in claim 6, wherein, corresponding first weight of the first predilection grade of the acquisition obtains the Corresponding second weight of two predilection grades further include:

It is pre- by second using the first preset value as first weight when that can not determine that the corresponding visitor's realm of the user is other If value is used as second weight.

11. a kind of commercial articles searching device, comprising:

Receiving module, the searching request for being directed to commodity for receiving user；

Module is obtained, for obtaining search result corresponding with described search request, described search result includes the commodity Relevant information；

Decision-making module determines optimal type of offer for the attribute data of attribute data and the commodity based on the user；

Output module, for exporting described search result in association with the favor information for meeting the optimal type of offer.

12. device according to claim 11, wherein attribute data of the decision-making module based on the user and described The attribute data of commodity determines that optimal type of offer includes:

The decision-making module, for for any type of offer, the attribute data based on the user obtain the user about First predilection grade of the type of offer；Attribute data based on the commodity obtains the commodity about the type of offer The second predilection grade；Based on first predilection grade and second predilection grade, the user and the commodity are obtained Synthesis predilection grade about the type of offer；By the comprehensive highest type of offer of predilection grade as optimal type of offer.

13. device according to claim 12, wherein the decision-making module obtains institute based on the attribute data of the user The first predilection grade that user is stated about the type of offer includes:

The decision-making module, for obtaining the corresponding first History Order data of the user, the first History Order data In comprising the quantity that meets the order of the favor information of the type of offer be the first quantity, in the first History Order data The quantity of order comprising meeting the favor information of any type of offer is the second quantity；Total user corresponding second is obtained to go through The quantity of history order data, the order in the second History Order data comprising meeting the favor information of the type of offer is Third quantity, the quantity of the order in the second History Order data comprising meeting the favor information of any type of offer are the Four quantity；Using the ratio of the first quantity and the second quantity relative to third quantity and the accounting of the ratio of the 4th quantity as described in First predilection grade.

14. device according to claim 12, wherein the decision-making module obtains institute based on the attribute data of the commodity The second preference-score that commodity are stated about the type of offer includes:

The decision-making module, for determining the corresponding Price Zoning of the commodity based on the price of the commodity；Obtain the valence The corresponding third History Order data of commodity in lattice subregion, comprising meeting the preferential class in the third History Order data The quantity of the order of the favor information of type is the 5th quantity, includes to meet any type of offer in the third History Order data Favor information order quantity be the 6th quantity；The corresponding 4th History Order data of all commodity of acquisition, the described 4th Quantity in History Order data comprising meeting the order of the favor information of the type of offer is the 7th quantity, and the described 4th goes through The quantity of order in history order data comprising meeting the favor information of any type of offer is the 8th quantity；By the 5th quantity with The ratio of 6th quantity is relative to the 7th quantity and the accounting of the ratio of the 8th quantity as second predilection grade.

15. device according to claim 12, wherein the decision-making module is based on first predilection grade and described the Two predilection grades, the synthesis predilection grade for obtaining the user and the commodity about the type of offer include:

The decision-making module obtains the second predilection grade corresponding for obtaining corresponding first weight of the first predilection grade Two weights；First predilection grade and second predilection grade are carried out based on first weight and second weight Weighted sum obtains the comprehensive predilection grade.

16. device according to claim 15, wherein the decision-making module obtains corresponding first power of the first predilection grade Weight, obtaining corresponding second weight of the second predilection grade includes:

The decision-making module, it is different for determining that the corresponding objective realm of the user is other according to the attribute data of the user The different purchasing power of objective group's categorized representation is horizontal；The corresponding Price Zoning of the commodity is determined according to the price of the commodity；Structure Preferential preference pattern is built, the input of the preferential preference pattern includes the objective realm not about the first preference of each type of offer Scoring and second predilection grade of the Price Zoning about each type of offer, the output of the preferential preference pattern are the visitor Realm not highest comprehensive predilection grade corresponding with the Price Zoning, the parameter of the preferential preference pattern include the first power Weight parameter and the second weight parameter；Obtain the objective realm highest actual preference scoring not corresponding with the Price Zoning； It is that target optimizes the parameter of the preferential preference pattern with the highest actual preference scoring, by the first optimal power The value of weight parameter is as first weight, using the value of the second optimal weight parameter as second weight.

17. device according to claim 16, wherein the decision-making module determines institute according to the attribute data of the user Stating the corresponding objective realm of user does not include:

The decision-making module, for average income data, the average order price based on the user in preset period of time And/or the average consumption amount of money, determine that the corresponding objective realm of the user is other.

18. device according to claim 16, wherein the decision-making module constructs preferential preference pattern and includes:

The decision-making module is obtained and is corresponded to for obtaining the attribute data for corresponding to the other multiple sample of users of the objective realm In the attribute data of multiple sample commodity of the Price Zoning；Attribute data based on the multiple sample of users and described more The attribute data of a sample commodity；For any type of offer, described in the attribute data acquisition based on the multiple sample of users First predilection grade of multiple sample of users about the type of offer, by the multiple sample of users about the type of offer The sum of one predilection grade is as the objective realm not about the first predilection grade of the type of offer；Based on the multiple sample quotient The attribute data of product obtains second preference of the Price Zoning corresponding to the multiple sample commodity about the type of offer Scoring；Based on the first weight parameter and the second weight parameter to the objective realm not about the first predilection grade of the type of offer About the second predilection grade of the type of offer being weighted summation with the Price Zoning, to obtain the objective realm not and described The corresponding synthesis predilection grade about the type of offer of Price Zoning.

19. device according to claim 18, wherein the decision-making module obtains the objective realm and do not divide with the price The corresponding highest actual preference in area, which scores, includes:

The decision-making module, for obtaining any sample of users and the Price Zoning being corresponding for any type of offer The 5th History Order data, comprising meeting the order of the favor information of the type of offer in the 5th History Order data Quantity is the 9th quantity, includes the number for meeting the order of favor information of any type of offer in the 5th History Order data Amount is the tenth quantity；Using the ratio accounting of the 9th quantity and the tenth quantity as the sample of users and the Price Zoning about The actual preference of the type of offer scores, the reality by the multiple sample of users and the Price Zoning about the type of offer The sum of predilection grade does not score with the Price Zoning about the actual preference of the type of offer as the objective realm；Compare institute It states objective realm not score with the Price Zoning about the actual preference of each type of offer, does not obtain the objective realm not and the valence The corresponding highest actual preference scoring of lattice subregion.

20. device according to claim 16, wherein the decision-making module obtains corresponding first power of the first predilection grade Weight obtains corresponding second weight of the second predilection grade further include:

The decision-making module is also used to when that can not determine that the corresponding visitor's realm of the user is other, using the first preset value as institute The first weight is stated, using the second preset value as second weight.

21. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor Calculation machine program, the processor realize such as commercial articles searching side of any of claims 1-10 when executing described program Method.

22. a kind of computer readable storage medium, is stored thereon with executable instruction, which makes to handle when being executed by processor Device executes such as product search method of any of claims 1-10.