CN107967637A

CN107967637A - The recommendation method, apparatus and electronic equipment of a kind of merchandise items model

Info

Publication number: CN107967637A
Application number: CN201610916477.3A
Authority: CN
Inventors: 刘晓倩
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2016-10-20
Filing date: 2016-10-20
Publication date: 2018-04-27
Anticipated expiration: 2036-10-20
Also published as: CN107967637B

Abstract

This application discloses the recommendation method, apparatus and electronic equipment of a kind of merchandise items model, a kind of choosing method, device and the electronic equipment of merchandise items model, a kind of selecting system of merchandise items model.Wherein, the recommendation method of the merchandise items model includes：Receive the model recommendation request for particular commodity object of the first subscription client transmission；Comment information of the user to its actual purchase model has been bought according to the particular commodity object, the model recommendation information of the particular commodity object provided to first user is provided；To model recommendation information described in the first subscription client loopback.The recommendation method of the merchandise items model provided using the application, user's offer model recommendation information can be newly bought from trend to the comment information of its actual purchase model according to user has been bought, recommend efficiency so as to reach and improve marque and recommend the effect of precision.

Description

The recommendation method, apparatus and electronic equipment of a kind of merchandise items model

Technical field

This application involves technical field of data processing, and in particular to a kind of recommendation method of merchandise items model；Correspond to The above method, the application are related to a kind of recommendation apparatus and electronic equipment of merchandise items model, a kind of merchandise items type at the same time Number choosing method, device and electronic equipment, a kind of selecting system of merchandise items model, and another merchandise items type Number choosing method and device.

Background technology

Buyer is when buying nonstandard class commodity (such as clothes, cap, shoes) on electric business platform, due to that can not try on Wear, therefore, model is (such as：Size model or color model etc.) select to be a more doubt link.In order to avoid buying type Therefore number inappropriate commodity, part buyer may abandon online shopping mode, so as to be converted to a certain extent to commodity Rate has an impact.

At present, buyer mainly determines the commodity to be bought when buying commodity on electric business platform using following two modes Model：

1) by way of being putd question in marque question and answer area, the marque to be bought is determined, for example, in mobile phone Taobao Dotey's details page in be provided with " asking an everybody " module, for similar " women's dress " this kind of nonstandard class commodity, the overwhelming majority is bought Family is all inquiring what size problem the height and weight of oneself should buy, and then, is returned by the people for buying the commodity Answer.

2) by way of checking the keyword for the related model extracted from information on commodity comment, the business to be bought is determined Product model, for example, in dotey's details page of mobile phone Taobao, by the related sizing information given by the buyer of purchased item Keyword is (such as：Size is bigger than normal, the less than normal or keyword such as suitable) extracted from information on commodity comment, it is placed on dotey's evaluation Before referred to for following buyer；When a user will buy the commodity, the information such as these keywords are referred to, make size Selection.

However, above two method is required to buyer by checking that existing information voluntarily selects marque, electric business platform It can not recommend the marque being applicable in buyer automatically.In conclusion exist can not Recommendations model automatically for the prior art Problem.

The content of the invention

The application provides a kind of recommendation method of merchandise items model, with solve can not automatic Recommendations under the prior art The problem of model.The application also provides a kind of recommendation apparatus and electronic equipment of merchandise items model, a kind of merchandise items type Number choosing method, device and a kind of selecting system of merchandise items model of electronic equipment, and another merchandise items type Number choosing method and device.

The application provides a kind of recommendation method of merchandise items model, including：

Receive the model recommendation request for particular commodity object of the first subscription client transmission；

Comment information of the user to its actual purchase model has been bought according to the particular commodity object, has been determined to described The model recommendation information for the particular commodity object that first user provides；

To model recommendation information described in the first subscription client loopback.

Optionally, the purchase user for having bought user and having included there is model associations relation with first user.

Optionally, the comment of the user to its actual purchase model of having bought according to the particular commodity object is believed Breath, and the model recommendation information of the particular commodity object provided to first user is provided, including：

Obtain the purchase user that there is model associations relation with first user；

According to default user's selection rule, have from first user in the user of purchase of model associations relation Choose a specific user；

Comment information according to the specific user to its actual purchase model, determines the model recommendation information.

Optionally, default user's selection rule includes：Choose the highest user of the model degree of correlation.

Optionally, the acquisition has the purchase user of model associations relation with first user, including：

According to the transaction record of the particular commodity object, obtain each of particular commodity object and bought user's User identifier；

Between the user identifier of first user carried according to the model recommendation request, the different user previously generated Model degree of correlation information and each user identifier for having bought user, obtain with first user there are model associations The purchase user of relation.

Obtain the model comment information collection of the particular commodity object；

According to the model comment information collection, obtain and each of the model comment information is provided to the particular commodity object A user identifier for having bought user；

Optionally, the model comment information collection for obtaining the particular commodity object, including：

Obtain the comment on commodity information collection of the particular commodity object；

By semantic analysis technology, concentrated from the comment on commodity information and extract the comment information in relation to model；

According to the comment information for the related model extracted, the model comment information collection is generated.

Optionally, the model includes size model；The model degree of correlation information includes size degree of correlation information；

Model degree of correlation information between the different user, generates in the following way：

For each merchandise items in the merchandise items storehouse prestored, following steps are performed：

Obtain the size comment information collection of the merchandise items；

For each size comment information of the merchandise items, the reality of the corresponding user of the size comment information is obtained Buy size in border；And size appropriate degree information is extracted from the size comment information；And according to the size extracted Appropriate degree information and the actual purchase size, determine applicable size of the user to the merchandise items；

According to the corresponding applicable size of each size comment information of the merchandise items, will have identical described suitable Add one with the size degree of correlation between the user of size.

Optionally, the size appropriate degree information and the actual purchase size that the basis is extracted, and determine the use Family is to the applicable sizes of the merchandise items, in the following way：

If the size appropriate degree information is bigger than normal for size, the user sets the applicable size of the merchandise items It is set to one yard smaller than the actual purchase size；

If the size appropriate degree information is less than normal for size, the user sets the applicable size of the merchandise items It is set to one yard bigger than the actual purchase size；

If the size appropriate degree information is suitable for size, the user sets the applicable size of the merchandise items It is set to the actual purchase size.

Optionally, the comment information according to the specific user to its actual purchase model, and determine the model Recommendation information, including：

Model appropriate degree information is extracted from comment information of the specific user to its actual purchase model；

According to the actual purchase model for the model appropriate degree information and the specific user extracted, the type is determined Number recommendation information.

Optionally, model appropriate degree letter is extracted in the comment information to its actual purchase model from the specific user Breath, in the following way：

By semantic analysis technology, model conjunction is extracted from comment information of the specific user to its actual purchase model Appropriate information.

Optionally, the model includes size model, and the model appropriate degree information includes size appropriate degree information, described Actual purchase model includes actual purchase size, and the model recommendation information includes size recommendation information；

The model appropriate degree information and the actual purchase model of the specific user that the basis is extracted, and determine The model recommendation information, in the following way：

If the size appropriate degree information is bigger than normal for size, according to the size than small one yard of the actual purchase size, Generate the size recommendation information；

If the size appropriate degree information is less than normal for size, according to the size than big one yard of the actual purchase size, Generate the size recommendation information；

If the size appropriate degree information is suitable for size, according to the actual purchase size, generates the size and push away Recommend information.

The standard model or user for obtaining first user specify model；

According to each type bought comment information of the user to its actual purchase model, generated the particular commodity object Number appropriate degree information；

Model is specified according to the model appropriate degree information and the standard model or user, determines that the model is recommended Information.

The standard model or user for obtaining first user specify model；

According to have purchased the standard model or what user specified model has bought comment information of the user to model, generation The model appropriate degree information of the particular commodity object；

Correspondingly, the application also provides a kind of recommendation apparatus of merchandise items model, including：

Request reception unit, recommends to ask for receiving the model for particular commodity object that the first subscription client is sent Ask；

Recommendation information determination unit, for the user of purchase according to the particular commodity object to its actual purchase model Comment information, determine to first user provide the particular commodity object model recommendation information；

Recommendation information loopback cell, for model recommendation information described in the first subscription client loopback.

Optionally, the recommendation information determination unit includes：

First user obtains subelement, is used for obtaining the purchase for having model associations relation with first user Family；

Specific user chooses subelement, for according to default user's selection rule, having type from first user A specific user is chosen in the user of purchase of number incidence relation；

Recommendation information determination subelement, for the comment information according to the specific user to its actual purchase model, really The fixed model recommendation information.

Optionally, the recommendation information determination subelement includes：

Model appropriate degree information extraction subelement, for the comment information from the specific user to its actual purchase model Middle extraction model appropriate degree information；

First recommendation information determination subelement, for according to the model appropriate degree information and the specific user extracted The actual purchase model, determines the model recommendation information.

Optionally, the recommendation information determination unit includes：

Initial model obtains subelement, and standard model or user for obtaining first user specify model；

Model appropriate degree information generates subelement, for according to each comment for having bought user to its actual purchase model Information, generates the model appropriate degree information of the particular commodity object；

Recommendation information determination subelement, for being referred to according to the model appropriate degree information and the standard model or user Sizing number, determines the model recommendation information.

Optionally, the recommendation information determination unit includes：

Model appropriate degree information generates subelement, have purchased the standard model for basis or user has specified model Comment information of the user to model is bought, generates the model appropriate degree information of the particular commodity object；

Correspondingly, the application also provides a kind of electronic equipment, including：

Display；

Processor；And

Memory, for storing the program for the recommendation method for realizing merchandise items model, which is powered and by described After processor runs the program of recommendation method of the merchandise items model, following step is performed：Receive the first subscription client hair The model recommendation request for particular commodity object sent；According to the user of purchase of the particular commodity object to its actual purchase The comment information of model is bought, the model recommendation information of the particular commodity object provided to first user is provided；To institute State model recommendation information described in the first subscription client loopback.

Correspondingly, the application also provides a kind of choosing method of merchandise items model, including：

The model recommendation request for particular commodity object is sent to server；

Receive the model recommendation information of the particular commodity object of the server loopback；

According to the model recommendation information, the selection model of the particular commodity object is chosen.

Optionally, the model recommendation request includes the user identifier of the first user；

The recommendation that the model recommendation information includes the standard model of first user and provided for first user Model.

Optionally, the model of the particular commodity object, which chooses the page, includes the corresponding first page face element of the standard model The corresponding second page element of plain and described proposed model.

Optionally, it is described according to the model recommendation information, and the selection model of the particular commodity object is chosen, wrap Include：

The corresponding standard model of the first page element is chosen, as tentatively selected selection model；

The change procedure of selection model is visually represented with the variation pattern of the predetermined display parameter of page elements, by institute Stating the selection model of particular commodity object, from the tentatively selected selection model to be changed to the second page element corresponding Proposed model.

Optionally, the model recommendation request specifies model including user；

The model recommendation information includes the proposed model provided for the first user.

Optionally, the model of the particular commodity object chooses the page and specifies the corresponding first page of model including the user Surface element and the corresponding second page element of the proposed model.

Optionally, it is described according to the model recommendation information, and the selection model of the particular commodity object is chosen, use Following manner：

The change procedure of selection model is visually represented with the variation pattern of the predetermined display parameter of page elements, by institute The selection model for stating particular commodity object specifies model to be changed to the corresponding recommendation type of the second page element from the user Number.

Optionally, the predetermined display parameter includes background colour；

The variation pattern of the predetermined display parameter with page elements visually represents the change procedure of selection model, In the following way：

According to default transformation period, the background colour of the first page element is gradually moved to the second page face element Element, the background colour as the second page element.

Correspondingly, the application also provides a kind of selecting device of merchandise items model, including：

Request transmitting unit, for sending the model recommendation request for particular commodity object to server；

Information receiving unit, the model recommendation information of the particular commodity object for receiving the server loopback；

Model chooses unit, for according to the model recommendation information, choosing the selection model of the particular commodity object.

Display；

Processor；And

Memory, for storing the program for the choosing method for realizing merchandise items model, which is powered and by described After processor runs the program of the choosing method of the merchandise items model, following step is performed：Sent to server for specific The model recommendation request of merchandise items；Receive the model recommendation information of the particular commodity object of the server loopback；Root According to the model recommendation information, the selection model of the particular commodity object is chosen.

Correspondingly, the application also provides a kind of selecting system of merchandise items model, including：According to any of the above-described Merchandise items model recommendation apparatus；And the selecting device of the merchandise items model according to any of the above-described.

Obtain the proposed model of particular commodity object；

The change procedure of selection model is visually represented with the variation pattern of the predetermined display parameter of page elements, in institute The model for stating particular commodity object is chosen in the page, and the selection model of the particular commodity object is corresponded to from first page element Tentatively selected selection model be changed to the corresponding proposed model of second page element.

Optionally, the predetermined display parameter includes background colour；

Proposed model acquiring unit, for obtaining the proposed model of particular commodity object；

Model chooses unit, for visually representing selection type with the variation pattern of the predetermined display parameter of page elements Number change procedure, the particular commodity object model choose the page in, by the selection model of the particular commodity object The corresponding proposed model of second page element is changed to from the corresponding tentatively selected selection model of first page element.

Compared with prior art, the recommendation method for the merchandise items model that the application provides, by receiving the first user visitor The model recommendation request for particular commodity object that family end is sent, and according to the purchase user couple of the particular commodity object The comment information of its actual purchase model, determines the model recommendation of the particular commodity object provided to first user Breath, then, then to model recommendation information described in the first subscription client loopback.The merchandise items type provided using the application Number recommendation method, can be provided according to having bought user the comment information of its actual purchase model newly being bought from trend user Model recommendation information, recommends efficiency and recommends the effect of precision so as to reach and improve marque.

Brief description of the drawings

Fig. 1 is a kind of flow chart of the embodiment of the recommendation method for merchandise items model that the application provides；

Fig. 2 is a kind of schematic diagram of the embodiment of the recommendation apparatus for merchandise items model that the application provides；

Fig. 3 is the schematic diagram of the embodiment for a kind of electronic equipment that the application provides；

Fig. 4 is a kind of flow chart of the embodiment of the choosing method for merchandise items model that the application provides；

Fig. 5 is a kind of interface schematic diagram of the embodiment of the choosing method for merchandise items model that the application provides；

Fig. 6 is a kind of schematic diagram of the embodiment of the selecting device for merchandise items model that the application provides；

Fig. 7 is the schematic diagram of the embodiment for a kind of electronic equipment that the application provides；

Fig. 8 is a kind of schematic diagram of the selecting system for merchandise items model that the application provides；

Fig. 9 is the flow chart of the embodiment of the choosing method for another merchandise items model that the application provides；

Figure 10 is the schematic diagram of the embodiment of the selecting device for another merchandise items model that the application provides.

Embodiment

Many details are elaborated in the following description in order to fully understand the application.But the application can Much to implement different from other manner described here, those skilled in the art can be in the feelings without prejudice to the application intension Similar popularization is done under condition, therefore the application is from the limitation of following public specific implementation.

In this application, there is provided the recommendation method, apparatus and electronic equipment of a kind of merchandise items model, a kind of commodity pair As the choosing method, device and electronic equipment of model, a kind of selecting system of merchandise items model, and another commodity pair As the choosing method and device of model.It is described in detail one by one in the following embodiments.

The recommendation method for the merchandise items model that the application provides, the basic thought of its core are：According to particular commodity pair Elephant has bought comment information of the user to its actual purchase model, and the model that particular commodity object is provided to purchase user is recommended Information.Due to can newly buy user's offer type from trend according to comment information of the user to its actual purchase model has been bought Number recommendation information, thus, improve marque and recommend efficiency and recommend precision.

Please refer to Fig.1, it is the flow chart of the recommendation embodiment of the method for the merchandise items model of the application.The method bag Include following steps：

Step S101：Receive the model recommendation request for particular commodity object of the first subscription client transmission.

The particular commodity object includes nonstandard class merchandise items.The merchandise items refer to that the digitization of commodity is expressed Form, for example, the commodity shown on electric business platform are merchandise items, and noncommodity is in itself.The nonstandard class commodity, i.e.,： The commodity for unified professional standard and the specification manufacture promulgated not in accordance with country.With the dress ornament class business such as clothes, shoes or cap Exemplified by product, since these commodity have the characteristics that size specification and disunity, color there may be aberration, they belong to Nonstandard class commodity.

The model can be the model of different angle, for example, size model, color model, material model or performance type Number etc..By taking dress ornament class commodity as an example, its size model is specifically as follows trumpet, medium size, large size etc.；Its color model specifically can be with For white, red, blueness etc.；Its material model is specifically as follows cotton, fiber crops, silk etc.；Its style model be specifically as follows long sleeves, Cotta, sleeveless etc..During stock control, the elementary cell of disengaging metering usually using model as stock, therefore, model Referred to as keeper unit (SKU, Stock Keeping Unit).

First user refers to, the user interested of the model recommendation information to particular commodity object.Described first uses Family sends the model recommendation request for particular commodity object by the first subscription client.First subscription client includes But mobile communication equipment is not limited to, i.e.,：Usually said mobile phone or smart mobile phone, further includes PC, PAD, iPad etc. Terminal device.

Model recommendation request from the first subscription client will at least include the mark of particular commodity object, and the application carries The recommendation method of the merchandise items model of confession, will determine particular commodity object according to the mark of particular commodity object.It is in addition, described Model recommendation request also needs to include at least one of data below：The user identifier of first user, first user To the specified model of the particular commodity object (i.e.：User specifies model).

The user identifier is used for one user of unique mark.When the model recommendation request does not include the particular commodity When the user of object specifies model, the standard type that first user is applicable in can be determined according to the user identifier of first user Number, i.e.,：The normal model of user.The recommendation method for the merchandise items model that the application provides, can fit according to first user Standard model or the user specify model, carry out model recommendation.

After receiving model recommendation request, it is possible into next step, according to the purchase of the particular commodity object Comment information of the user to its actual purchase model is bought, determines model recommendation information.

Step S103：Comment information of the user to its actual purchase model has been bought according to the particular commodity object, The model recommendation information of the particular commodity object provided to first user is provided.

The recommendation method of the merchandise items model provided using the application, provide model recommendation information for the first user one A important prerequisite is：The particular commodity object has at least had one to buy user.It is described to have bought user, may be with First user has the user of model associations relation, it is also possible to is without model associations relation with first user User.

Judge whether there is model associations relation between user, a variety of embodiments can be used, be given below two kinds Optional embodiment.

If having identical standard model between mode one, user, it can determine that between user that there is model associations pass System, if for example, the standard size of user A and user B is L codes, user A and user B have model associations relation.

If mode two, multiple users bought certain commodity of same model at the same time, can determine that between these users With model associations relation, if for example, user A, user B and user C have purchased the same money jacket of L codes at the same time, these three are used There is model associations relation between family.

Further, can also according between user buy same model identical commodity number, set user between The degree of association, if for example, user A, user B and user C have purchased the same money jacket of L codes at the same time, user A and user B are purchased at the same time 40 yards of same money shoes have been bought, then there is model associations relation and mutual association between user A, user B and user C Spend for 1, it is 2 to have model associations relation and the mutual degree of association between user A and user B.

It should be noted that above-mentioned judge whether there is the degree of association between model associations relation and definite user between user Various forms of changes, all simply changes of embodiment, all without departing from the core of the application, therefore all in the application Protection domain within.

The recommendation method for the merchandise items model that the application provides, can buy whether user includes and first according to described User has the concrete condition of the user of model associations relation, and different embodiments is respectively adopted and realizes this step S103.Under Different embodiments under both of these case are made an explanation.

First, when the user of purchase of the particular commodity object does not include having model associations relation with first user User when, according to the standard model of first user or specified model and user can have been bought to its actual purchase model Comment information, determine to first user provide the particular commodity object model recommendation information.

As a kind of optional scheme, according to the standard model of the first user or specified model and user has been bought to it The comment information of actual purchase model, determines the model recommendation information provided to the first user, can be in the following way：According to big Majority has bought user to the evaluation information of model (i.e.：Model appropriate degree information), determine to push away to the model that the first user provides Recommend information, for example, it is most of (such as：More than 50%) it is that size is bigger than normal to the evaluation information of model to have bought user, and first uses The standard model at family or specified model L codes, then the model recommendation information provided to the first user may include：Size is bigger than normal, pushes away It is M codes to recommend size.

As another optional scheme, according to the standard model of the first user or specified model and user couple has been bought The comment information of its actual purchase model, determines the model recommendation information provided to the first user, can also be in the following way：Root According to most of users for having bought standard model that the first user is applicable in or specified model to the evaluation information of model (i.e.：Type Number appropriate degree information), the model recommendation information provided to the first user is determined, for example, the standard model of the first user or specifying Model L codes, it is most of (such as：More than 50%) user for having bought L codes is that size is bigger than normal to the evaluation information of the model, then The model recommendation information provided to the first user may include：Size is bigger than normal, and recommendation size is M codes.

When the model recommendation request include the particular commodity object mark and first user user identifier, During without specifying model including user, the recommendation method for the merchandise items model that the application provides can be used according to described first first The user identifier at family determines the standard model of first user, and then, mark and first further according to particular commodity object are used The standard model at family, and comment information of the user to its actual purchase model has been bought, the type provided to the first user is provided Number recommendation information.

Realize the function that standard model that user is applicable in is determined according to user identifier, a variety of specific embodiment parties can be used Formula, is given below several optional embodiments.

Mode one, have recorded the various type informations that user is applicable in, these types in the user characteristic data of the first user Number it is considered as standard model, for example, the type information such as jacket size, trousers size, shoes size, then, according to the first user User identifier, you can the standard model of the user is obtained from its user characteristic data.

Mode two, have recorded the body-shape informations such as height, the weight of the first user in the user characteristic data of the first user, So, the body-shape information of the user can be obtained according to the user identifier of the first user, by these body-shape informations and to be chosen The size details (waistline, bust, sleeve length etc.) of dress ornament class commodity are contrasted, you can the user is calculated to needed for the clothes Model, this model can also be considered as standard model.

Mode three, do not record the body-shape informations such as height, weight in the user characteristic data of the first user, does not record yet Type information, but other users that there is model associations relation with the user are have recorded, in the user characteristics number of other users It has recorded the information such as height, weight or model in, then, it can first be obtained according to the user identifier of the first user and had with it There are other users of model associations relation, and then refer to the information such as the height, weight or model of other users, obtain first The type information of user, the standard model as the first user.

When the model recommendation request includes mark and the user of the particular commodity object of the particular commodity object When specifying model, the recommendation method for the merchandise items model that the application provides can be directly according to these two aspects information, and purchased Comment information of the user to its actual purchase model is bought, the type of the particular commodity object provided to first user is provided Number recommendation information.

Above section is illustrated when the user of purchase of particular commodity object does not include having model associations with the first user During the user of relation, the adoptable embodiments of step S103.

Below by the user for having bought user and having included there is model associations relation with the first user of particular commodity object In the case of processing procedure illustrate.

2nd, when the user that bought of the particular commodity object includes having model associations relation with first user During user, can according to first user have model associations relation some (or some) user to its actual purchase model Comment information, determine to first user provide the particular commodity object model recommendation information.

In the present embodiment, according to the first user have model associations relation some user to its actual purchase model Comment information, determine to the first user provide model recommendation information.Realize has model associations according to the first user Some specific user of relation determines the model recommendation provided to the first user to the comment information of its actual purchase model The function of breath, this step S103 may include following specific steps：

Step S1031：Obtain the purchase user that there is model associations relation with first user.

Will be according to the comment of some specific user to its actual purchase model with the first user with model associations relation Information, determines the model recommendation information provided to the first user, it is necessary first to which obtain has model associations relation with the first user Purchase user.For example, it is assumed that there are 20 with user of first user with model associations relation, particular commodity pair have purchased The user of elephant has 100, and there are only 3 users in this 20 users of model associations relation have purchased the spy with the first user Determine merchandise items, then this 3 users are the purchase user for having model associations relation with the first user.

The purchase user that there is model associations relation with first user is obtained, a variety of specific embodiment parties can be used Formula, is given below two kinds of optional embodiments.

Mode one, the processing procedure of which are as described below：First, according to the transaction record of particular commodity object, obtain Each user identifier for having bought user of particular commodity object；Then, according to the user identifier of the first user, previously generate Model degree of correlation information and each user identifier for having bought user between different user, obtain and are closed with the first user with model The purchase user of connection relation.

Since mode one is according to the transaction record of particular commodity object, obtains and closed with first user with model associations The purchase user of system, therefore, the purchase user with the first user with model associations relation that employing mode one is got, " all " including particular commodity object have the purchase user of model associations relation with the first user.

Mode two, the processing procedure of which are as described below：First, the model comment information of particular commodity object is obtained Collection；Then, according to model comment information collection, obtain and each of model comment information is provided to particular commodity object bought user User identifier；Finally, according to the model degree of correlation information between the user identifier of the first user, the different user previously generated and Each user identifier for having bought user, obtains the purchase user for having model associations relation with the first user.

The model comment information collection refers to, the set of the comment on commodity information related with model.By having bought specific business The user of product object forms the model comment information collection of the merchandise items to the comment information of the related model of the merchandise items.

Since mode two is according to the model comment information collection of particular commodity object, obtain has model with first user The purchase user of incidence relation, therefore, the purchase with the first user with model associations relation that employing mode two is got User is bought, " part " and purchase user of first user with model associations relation that may be only including particular commodity object, I.e.：Purchase user that model is commented on, that there is model associations relation with the first user was only carried out to particular commodity object.

Employing mode two is wanted to be handled, it is necessary to obtain the model comment information collection of particular commodity object first.It is specific real The step of Shi Shi, the model comment information collection of acquisition particular commodity object, it may include following specific steps：1) obtain described specific The comment on commodity information collection of merchandise items；2) by semantic analysis technology, extracted from comment on commodity information concentration related The comment information of model；3) according to the comment information for the related model extracted, the model comment information collection is generated.It is right below These steps are briefly described.

1) the comment on commodity information collection of the particular commodity object is obtained.

User can be criticized or be discussed for the merchandise items, i.e., after a merchandise items have been bought：Provide commodity Comment information.Comment on commodity information is to have purchased the subjective opinion of user, and having purchased user can be from all angles (such as：Size, color, Material, style, comfort level etc.) comment on having purchased merchandise items.By taking clothes as an example, commodity of the user to certain part clothes have been purchased Comment information is as shown in table 1：

Table 1, comment on commodity information list

The each comment on commodity information structure that the merchandise items that user is identified as merchandise items " 1 " have been purchased in table 1 should The comment information collection of merchandise items.

Since comment on commodity information includes having purchased the comment that user from different perspectives made commodity, and model comment information Only include having purchased the comment that user makes marque, therefore, the model comment information collection is usually the comment on commodity The a subset of information collection, for example, model size model, in table 1 comment be identified as 2,3,4,5,76,78,79 commodity and comment Include size comment information by information, wherein, the size comment information that comment is identified as 2,3,4 constitutes the rulers of merchandise items 1 Code comment information collection.

2) by semantic analysis technology, concentrated from the comment on commodity information and extract the comment information in relation to model.

The model comment information collection is that the information subset extracted is concentrated from comment on commodity information.It is being embodied When, the comment information in relation to model can be extracted from comment on commodity information by semantic analysis technology, for example, being commented in table 1 The comment information for the related model being identified as included by the comment on commodity information of " 2 " is " fit ", and comment is identified as the commodity of " 3 " The comment information of related model included by comment information is " but to buy than usual one yard big ", and comment is identified as the business of " 4 " The comment information of related model included by product comment information is " number is less than normal, is bought by usual code, somewhat small ".

3) according to the comment information for the related model extracted, the model comment information collection is generated.

Previous step is concentrated the comment information of the related model extracted by this step from the comment on commodity information, is formed The model comment information collection.

No matter use aforesaid way one or mode two to realize this step S103, be both needed to using the different user previously generated Between model degree of correlation information.Model degree of correlation information between the different user previously generated, can be that any two is used Degree of correlation between family in terms of model, degree of correlation that can also be between multiple users in terms of model.With model Exemplified by size model, the size degree of correlation information between any two user is as shown in table 2：

Mark	Association user	The size degree of correlation
			1	User A, user B	23
2	User A, user C	10
			3	User A, user X	2
4	User A, user Y	1
			…	…	…
10001	User Y, user B	13
			10002	User Y, user C	8
10003	User Y, user X	5
			10004	User Y, user Z	1
…	…	…

Size degree of correlation information between table 2, different user

The information structure of degree of correlation in table 2 between a plurality of different user in terms of size is described to be previously generated not With the size degree of correlation information between user.

Below by taking size model as an example, a kind of optional side for generating the model degree of correlation information between the different user is provided Case.In the present embodiment, the size degree of correlation information between the different user, can generate, i.e., in the following way：For advance Each merchandise items in the merchandise items storehouse of storage, perform following specific steps：1) size for obtaining the merchandise items is commented By information collection；2) each size comment information of the merchandise items is directed to, obtains the corresponding user of the size comment information Actual purchase size；And size appropriate degree information is extracted from the size comment information；And according to extracting Size appropriate degree information and the actual purchase size, determine applicable size of the user to the merchandise items；3) basis The corresponding applicable size of each size comment information of the merchandise items, by with the identical user for being applicable in size Between the size degree of correlation add one.

Firstly the need of explanation, above-mentioned steps 1 to step 3 are needed for each in the merchandise items storehouse prestored Merchandise items perform once respectively.Elaborate respectively to above-mentioned steps 1 to step 3 below.

1) the size comment information collection of the merchandise items is obtained.

For any one merchandise items, it is necessary first to obtain and purchased the size comment letter that user makes the merchandise items These size comment informations, are formed the size comment information collection of the merchandise items by breath.

2) each size comment information of the merchandise items is directed to, obtains the corresponding user's of the size comment information Actual purchase size；And size appropriate degree information is extracted from the size comment information；And according to the ruler extracted Code appropriate degree information and the actual purchase size, determine applicable size of the user to the merchandise items.

Each size comment information of the merchandise items got for previous step is commented, it is necessary to further obtain size By the actual purchase size of the corresponding user of information, in the specific implementation, can be extracted from the corresponding order of size comment information Go out to have purchased the commodity size of user's actual purchase.

It is also desirable to size appropriate degree information is extracted from size comment information.In the specific implementation, language can be passed through Adopted analytical technology extracts size appropriate degree information from size comment information, for example, comment is identified as the commodity of " 2 " in table 1 Size appropriate degree information included by object comment information is " suitable ", and comment is identified as the merchandise items comment information institute of " 3 " Including size appropriate degree information be " size is bigger than normal ", comment is identified as size included by the merchandise items comment information of " 4 " Appropriate degree information is " size is less than normal ".

Then, it is possible to which, according to the size appropriate degree information extracted and the size for having purchased user's actual purchase, determining should Applicable size of the user to the merchandise items is purchased.

In the specific implementation, according to the size appropriate degree information and the actual purchase size extracted, the use is determined The step of family is to the applicable sizes of the merchandise items, can be in the following way：If the size appropriate degree information is inclined for size Greatly, then the user is arranged to the applicable size of the merchandise items to one yard smaller than the actual purchase size；It is if described Size appropriate degree information is less than normal for size, then is arranged to the user the applicable size of the merchandise items than the reality It is one yard big to buy size；If the size appropriate degree information is suitable for size, the user fits the merchandise items The actual purchase size is arranged to size.

3) according to the corresponding applicable size of each size comment information of the article, it is described will have it is identical described The size degree of correlation being applicable between the user of size adds one.

By previous step get it is each purchased user to the applicable size of merchandise items after, this step is according to these Sizing information is applicable in, one will be added with the identical size degree of correlation being applicable between the different user of size, for example, user A and use Family B have purchased a kind of merchandise items, and be identical applicable size for the merchandise items, then by user A and user B it Between the size degree of correlation add one；If user A and user B have purchased another merchandise items at the same time again, and for the merchandise items It is identical applicable size again, then the size degree of correlation between user A and user B is added one again.

Each merchandise items that above-mentioned steps 1 are directed in the merchandise items storehouse prestored to step 3 perform once respectively Afterwards, that is, the size degree of correlation information between different user as shown in Table 2 is generated.The model generated between the different user is related After spending information, you can using aforesaid way one or mode two, using the model degree of correlation information between the different user previously generated, Obtain the purchase user that there is model associations relation with first user.

Step S1033：According to default user's selection rule, there is model associations relation from first user Buy and a specific user is chosen in user.

Default user's selection rule includes but not limited to：The highest user of the model degree of correlation is chosen, use is this After user has been purchased in selection rule selection, the recommendation that the user carries out the comment information of its actual purchase model size is referred to, So as to obtain preferable recommendation effect.Certainly, in the specific implementation, other users selection rule can also be set, for example, from big Optional user etc. in the user of default Minimum relevance weight threshold value.

Step S1035：Comment information according to the specific user to its actual purchase model, determines that the model is recommended Information.

The specific user buys the actual purchase model of the particular commodity object, can be from the user to the merchandise items Trading order form in extract and obtain.

The model recommendation information, it may include the concrete model recommended to the first user, may also include to the specific business The size evaluation information of product object is (i.e.：Model appropriate degree information), for example, the size of the commodity is bigger than normal or the types such as size is less than normal Number appropriate degree information.In addition, the model recommendation information, specifically may also include the standard model of the first user, for the first use Family is with reference to use.

When it is implemented, this step S1035 may include following specific steps：1) from the specific user to its actual purchase Model appropriate degree information is extracted in the comment information of model；2) according to the model appropriate degree information and the specific user extracted The actual purchase model, determine the model recommendation information.

Wherein, model appropriate degree information is extracted from comment information of the specific user to its actual purchase model, can In the following way：By semantic analysis technology, extracted from comment information of the specific user to its actual purchase model Model appropriate degree information.For example, comment is identified as the size appropriate degree letter included by the merchandise items comment information of " 2 " in table 1 Cease for " suitable ", it is " size is bigger than normal " to comment on the size appropriate degree information being identified as included by the merchandise items comment information of " 3 ", It is " size is less than normal " to comment on the size appropriate degree information being identified as included by the merchandise items comment information of " 4 ".

In the present embodiment, the model size model, correspondingly, the model appropriate degree information is size appropriate degree Information, the model recommendation information are size recommendation information, the actual purchase model actual purchase size.It is being embodied When, this step S1035 can be realized in the following way：If the size appropriate degree information is bigger than normal for size, can be used to first Recommend the size of small one yard of the actual purchase size than specific user in family；If the size appropriate degree information is less than normal for size, It can recommend the size of big one yard of the actual purchase size than specific user to the first user；If the size appropriate degree information is ruler Code is suitable, then can recommend the actual purchase size of specific user to the first user.

For example, user X wants the merchandise items that merchandise items in purchase table 1 are identified as " 1 "；Pass through table 2, user X The size degree of correlation with user A is 2, and the size degree of correlation of user X and user Y is 5；Understand that there is size with user X by table 1 In the user of the degree of correlation, only user A and user Y have purchased the article that merchandise items are identified as " 1 "；Due to user X and user Y The size degree of correlation be up to 5, therefore, size can be recommended to user X to the comment informations of the commodity according to user Y；Due to Size appropriate degree information included by the comment on commodity information of family Y is " size is less than normal " and the size of its actual purchase is L, because This, recommends XL codes to user X.

Step S105：To model recommendation information described in the first subscription client loopback.

The model recommendation information that above-mentioned steps are got is recycled to the first subscription client by this step, for the first user visitor Family end is with reference to use.

In the above-described embodiment, there is provided a kind of recommendation method of merchandise items model, corresponding, the application A kind of recommendation apparatus of merchandise items model is also provided.The device is corresponding with the embodiment of the above method.

Fig. 2 is refer to, it is the schematic diagram of the recommendation apparatus embodiment of the merchandise items model of the application.Since device is real Apply example and be substantially similar to embodiment of the method, so describing fairly simple, referring to the part explanation of embodiment of the method in place of correlation .Device embodiment described below is only schematical.

A kind of recommendation apparatus of the merchandise items model of the present embodiment, including：

Request reception unit 101, pushes away for receiving the model for particular commodity object that the first subscription client is sent Recommend request；

Recommendation information determination unit 103, for according to the user of purchase of the particular commodity object to its actual purchase The comment information of model, determines the model recommendation information of the particular commodity object provided to first user；

Recommendation information loopback cell 105, for model recommendation information described in the first subscription client loopback.

Optionally, the recommendation information determination unit 105 includes：

Optionally, the recommendation information determination subelement includes：

Optionally, the recommendation information determination unit 105 includes：

Please refer to Fig.3, it is the schematic diagram of the electronic equipment embodiment of the application.Since apparatus embodiments are substantially similar to Embodiment of the method, so describing fairly simple, the relevent part can refer to the partial explaination of embodiments of method.Described below Apparatus embodiments are only schematical.

The a kind of electronic equipment of the present embodiment, the electronic equipment include：Processor 101；And memory 103, for depositing The program of the recommendation method of merchandise items model is realized in storage, which is powered and runs the commodity pair by the processor 101 As the recommendation method of model program after, perform following step：Receive the transmission of the first subscription client is directed to particular commodity pair The model recommendation request of elephant；Comment information of the user to its actual purchase model has been bought according to the particular commodity object, The model recommendation information of the particular commodity object provided to first user is provided；Returned to first subscription client Send the model recommendation information.

The recommendation method, apparatus and electronic equipment for the merchandise items model that the application provides, by receiving the first user visitor The model recommendation request for particular commodity object that family end is sent, and according to the purchase user couple of the particular commodity object The comment information of its actual purchase model, determines the model recommendation of the particular commodity object provided to first user Breath, then, then to model recommendation information described in the first subscription client loopback.The merchandise items type provided using the application Number recommendation method, apparatus and electronic equipment, can be automatic to the comment information of its actual purchase model according to user has been bought Model recommendation information is provided to new purchase user, recommends efficiency so as to reach and improve marque and recommends the effect of precision.

Corresponding with the recommendation method of above-mentioned merchandise items model, the application also provides a kind of choosing of merchandise items model Take method.Please refer to Fig.4, a kind of flow chart of the choosing method embodiment of its merchandise items model provided for the application, this The embodiment part identical with first embodiment content repeats no more, and refers to the appropriate section in embodiment one.The application carries A kind of choosing method of the merchandise items model supplied includes：

Step S101：The model recommendation request for particular commodity object is sent to server.

The model recommendation request includes at least the mark of the particular commodity object, in addition, the model recommendation request It may also include：The user identifier of first user and the user of particular commodity object specify at least one of model.

The user identifier is used for one user of unique mark.When the model recommendation request includes user identifier, use Family client will obtain the user identifier of active user first, and identify to form model recommendation request according to the user；Server After receiving the request, the user identifier carried according to request is determined into the standard model that user is applicable in, i.e.,：The normal type of user Number, then, server end can carry out model recommendation according to the standard model that user is applicable in.

As a kind of optional scheme, user can select the concrete model of merchandise items first, then, can be by clicking on class Like modes such as " size recommendation tables please be check " buttons, the model recommendation request for particular commodity object is sent to server, should Model recommendation request carries user and specifies model；Server end will request to generate the type that the merchandise items are adapted to the user according to this Number (such as：Size) recommendation information, and to requesting party's loopback recommendation information.

Step S103：Receive the model recommendation information of the particular commodity object of the server loopback.

The model recommendation information, it may include the concrete model recommended to the first user.When model recommendation request includes the During the user identifier of one user, model recommendation information may also include the standard model of the first user.

In addition, the model recommendation information, may also include to the model evaluation information of particular commodity object (i.e.：Model is closed Appropriate information), for example, the size of the commodity is bigger than normal or size is less than normal etc..

Step S105：According to the model recommendation information, the selection model of the particular commodity object is chosen.

The choosing method for the merchandise items model that the application provides, is receiving the model recommendation information of server loopback Afterwards, the marque for being adapted to user will be chosen automatically according to these information.

As a kind of optional scheme, if model recommendation request includes the user identifier of the first user, model recommendation Breath includes the standard model and proposed model of the first user, and the model selection page of the particular commodity object includes standard The corresponding first page element of model and the corresponding second page element of proposed model, then this step S105 may include following specific Step：First, the corresponding standard model of first page element is chosen, as tentatively selected selection model；Then, with page The variation pattern of the predetermined display parameter of surface element visually represents to choose the change procedure of model, by particular commodity object Choose model and be changed to the corresponding proposed model of second page element from the tentatively selected selection model.

The predetermined display parameter refers to, the parameter preset being had an impact to the visual effect of page elements.It is described default Display parameters include but not limited at least one of parameter shown below：Background colour, border color, size, can also be other Display parameters, for example, font size, font color etc., these display parameters can produce shadow to the visual effect of page elements Ring.

The variation pattern of the predetermined display parameter with page elements visually represents the change procedure of selection model, Refer to the predetermined display parameter by varying initial selected model and the corresponding page elements of proposed model so that Yong Huneng Enough selection change procedures visually perceived from initial selected model to proposed model.

By taking predetermined display parameter is background colour as an example, the variation pattern of the predetermined display parameter with page elements from regarding The change procedure of selection model is represented in feel, can be in the following way：According to default transformation period, by the first page face element The background colour of element is gradually moved to the second page element, the background colour as the second page element.

The default transformation period refers to, the background colour of the first page element is moved fully to the second page The time that surface element is consumed.The default transformation period, can be set according to specific business need, for example, being set It is set to 5 seconds etc..

Fig. 5 is refer to, a kind of interface signal of the choosing method embodiment of its merchandise items model provided for the application Figure, by Fig. 5 can direct feel to the above embodiment model variation effect.As seen from Figure 5, user will buy a T-shirt, When the details of the merchandise items are checked in user's selection, subscription client can be sent for the merchandise items from trend server Model (size) recommendation request, the request include active user (i.e.：First user) user identifier；Server is according to the request Generate model (size) recommendation information that the merchandise items are adapted to the user, and to requesting party's loopback recommendation information；Server The model recommendation information of loopback includes：Standard size：XL codes, recommend size：XXL codes, and model appropriate degree information：Size is inclined It is small.

The model change procedure being related to below to Fig. 5 illustrates.The small figures of a of Fig. 5 are shown particular commodity object Selection model be tentatively chosen to be standard model (i.e.：The normal size of the user) effect, at this time, the corresponding page of standard model The background colour of surface element is the background colour completely filled；Over time, the background of the corresponding page elements of standard model Change takes place in color, and gradually to recommending the corresponding page elements of size to move, the small figures of b are shown the background colour of the page elements Model just starts effect during change, and at this time, the background colour of the corresponding page elements of standard model is the background colour being partially filled with； Elapse as time go on, a part for the background colour of the corresponding page elements of standard model, which has moved to, recommends size pair The page elements answered, the small figures of c show the effect of model change at this time, at this time, the background of the corresponding page elements of proposed model Color is the background colour being partially filled with；With the further passage of time, when reaching default transformation period, standard model corresponds to Page elements background colour be moved fully to recommend the corresponding page elements of size, at this time, recommend size corresponding The background colour of page elements is the background colour completely filled, and the small figures of d, which are shown, finally selects the selection model of particular commodity object The effect of proposed model is taken as, and also show model appropriate degree information (" size is less than normal, it is proposed that selects this "), it is straight for user Connect with reference to use.By this four pictures of the small figures of a to the small figures of d as it can be seen that by the way that the background colour of first page element is gradually moved to Two page elements, eventually as the mode of the background colour of second page element, can make user visually perceive selection type Number change procedure.

As another optional scheme, if model recommendation request specifies model including user, model recommendation information includes Proposed model, and the model of the particular commodity object chooses the page and specifies the corresponding first page face element of model including the user During the corresponding second page element of plain and described proposed model, then this step S105 can be in the following way：With page elements The variation pattern of predetermined display parameter visually represents to choose the change procedure of model, by the selection of the particular commodity object Model specifies model to be changed to the corresponding proposed model of the second page element from the user.

It should be noted that this step S105 can also use other embodiments, for example, recommendation type can directly be chosen Selection model number as particular commodity object, rather than first selection standard model is as the model tentatively selected, then with the page The variation pattern of the predetermined display parameter of element visually represents to choose change procedure of model etc..In practical applications, may be used According to specific business need choose embodiment, these above-mentioned different modes, all simply change of embodiment, All without departing from the core of the application, therefore all within the protection domain of the application.

In the above-described embodiment, there is provided a kind of choosing method of merchandise items model, corresponding, the application A kind of acquisition device of merchandise items model is also provided.The device is corresponding with the embodiment of the above method.

Fig. 6 is refer to, it is the schematic diagram of the acquisition device embodiment of the merchandise items model of the application.Since device is real Apply example and be substantially similar to embodiment of the method, so describing fairly simple, referring to the part explanation of embodiment of the method in place of correlation .Device embodiment described below is only schematical.

A kind of acquisition device of the merchandise items model of the present embodiment, including：

Request transmitting unit 101, for sending the model recommendation request for particular commodity object to server；

Information receiving unit 103, the model recommendation of the particular commodity object for receiving the server loopback Breath；

Model chooses unit 105, for according to the model recommendation information, choosing the selection type of the particular commodity object Number.

Fig. 7 is refer to, it is the schematic diagram of the electronic equipment embodiment of the application.Since apparatus embodiments are substantially similar to Embodiment of the method, so describing fairly simple, the relevent part can refer to the partial explaination of embodiments of method.Described below Apparatus embodiments are only schematical.

The a kind of electronic equipment of the present embodiment, the electronic equipment include：Display 101；Processor 102；And memory 103, for storing the program for the choosing method for realizing merchandise items model, which is powered and is transported by the processor 102 After the program of the choosing method of the row merchandise items model, following step is performed：Sent to server and be directed to particular commodity object Model recommendation request；Receive the model recommendation information of the particular commodity object of the server loopback；According to the type Number recommendation information, chooses the selection model of the particular commodity object.

The embodiment of the present application additionally provides a kind of selecting system of merchandise items model, as shown in figure 8, the system is including upper State the recommendation apparatus 101 of the merchandise items model described in embodiment and the selecting device 102 of merchandise items model.The commodity pair As the recommendation apparatus 101 of model is usually deployed in server, but it is not limited to server or can realizes the business Any equipment of the recommendation method of product object model；The selecting device 102 of the merchandise items model is usually deployed in mobile logical Interrogate the terminal devices such as equipment, PC, PAD, iPad.

For example, the selecting device 102 of merchandise items model is deployed on smart mobile phone, can be sent to server for spy Determine the model recommendation request of merchandise items, and receive the model recommendation of the particular commodity object of the server loopback Breath；The device 101 of the recommendation of the merchandise items model is disposed on the server, by receiving the transmission of the first subscription client For the model recommendation request of particular commodity object；According to the user of purchase of the particular commodity object to its actual purchase type Number comment information, determine to first user provide the particular commodity object model recommendation information；To described Model recommendation information described in one subscription client loopback；The selecting device 102 of the merchandise items model receives model recommendation After information, the selection model of the particular commodity object will be chosen automatically according to model recommendation information.

Using the selecting system of the application merchandise items model, the operation for choosing model is performed without user, simplifies use Family operates, and shortens model and chooses the duration of process, and the effect that model chooses efficiency is improved so as to reach.

Corresponding with the choosing method of above-mentioned merchandise items model, the application also provides another merchandise items model Choosing method.Fig. 9 is refer to, the flow of the choosing method embodiment of its another merchandise items model provided for the application Figure, the present embodiment part identical with the embodiment content of the choosing method of above-mentioned merchandise items model repeats no more, refers to Appropriate section in the embodiment.The choosing method for another merchandise items model that the application provides includes：

Step S101：Obtain the proposed model of particular commodity object.

Step S103：The change of selection model is visually represented with the variation pattern of the predetermined display parameter of page elements Process, in the model of the particular commodity object chooses the page, by the selection model of the particular commodity object from first page The corresponding tentatively selected selection model of surface element is changed to the corresponding proposed model of second page element.

The predetermined display parameter includes at least one of parameter shown below：Background colour, border color, size.

The tentatively selected selection model, can be the standard model of user, or the model that user specifies.

As a preferred solution, the predetermined display parameter includes background colour；It is described to be shown with the default of page elements Show that the variation pattern of parameter visually represents to choose the change procedure of model, can be in the following way：According to default change Time, is gradually moved to the second page element, as the second page face element by the background colour of the first page element The background colour of element.

For example, when user chooses the model of the merchandise items in the details page of some merchandise items, specify first Its usually applicable model, the model are the tentatively selected selection model, when the client of display details page obtains To the merchandise items provided to the user proposed model when, by the tentatively selected corresponding first page of selection model The background colour of element is gradually moved to the corresponding second page element of proposed model, as the background colour of second page element, makes Obtaining user can visually perceive from initial model of selecting to the selection change procedure of proposed model.

The acquisition methods for another merchandise items model that the application provides, by with the predetermined display parameter of page elements Variation pattern visually represent choose model change procedure, particular commodity object model choose the page in, by spy The selection model for determining merchandise items is changed to second page element from the corresponding tentatively selected selection model of first page element The corresponding proposed model；This processing mode so that user can be visually perceived from initial selected model to pushing away Recommend the selection change procedure of model；Therefore, user experience can be effectively improved.

In the above-described embodiment, there is provided the choosing method of another merchandise items model, corresponding, this Shen The acquisition device of another merchandise items model is please also provided.The device is corresponding with the embodiment of the above method.

Figure 10 is refer to, it is the schematic diagram of the acquisition device embodiment of another merchandise items model of the application.By Embodiment of the method is substantially similar in device embodiment, so describing fairly simple, related part is referring to embodiment of the method Part illustrates.Device embodiment described below is only schematical.

The acquisition device of another merchandise items model of the present embodiment, including：

Proposed model acquiring unit 101, for obtaining the proposed model of particular commodity object；

Model chooses unit 103, for visually representing to select with the variation pattern of the predetermined display parameter of page elements The change procedure of model is taken, in the model of the particular commodity object chooses the page, by the selection of the particular commodity object Model is changed to the corresponding recommendation type of second page element from the corresponding tentatively selected selection model of first page element Number.

Although the application is disclosed as above with preferred embodiment, it is not for limiting the application, any this area skill Art personnel are not being departed from spirit and scope, can make possible variation and modification, therefore the guarantor of the application Shield scope should be subject to the scope that the application claim is defined.

In a typical configuration, computing device includes one or more processors (CPU), input/output interface, net Network interface and memory.

Memory may include computer-readable medium in volatile memory, random access memory (RAM) and/or The forms such as Nonvolatile memory, such as read-only storage (ROM) or flash memory (flash RAM).Memory is computer-readable medium Example.

1st, computer-readable medium can be by any side including permanent and non-permanent, removable and non-removable media Method or technology realize that information stores.Information can be computer-readable instruction, data structure, the module of program or other numbers According to.The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), dynamic random access memory (DRAM), other kinds of random access memory (RAM), read-only storage (ROM), electrically erasable programmable read-only memory (EEPROM), fast flash memory bank or other memory techniques, read-only optical disc are read-only Memory (CD-ROM), digital versatile disc (DVD) or other optical storages, magnetic cassette tape, tape magnetic rigid disk storage or Other magnetic storage apparatus or any other non-transmission medium, the information that can be accessed by a computing device available for storage.According to Herein defines, and computer-readable medium does not include non-temporary computer readable media (transitory media), such as modulates Data-signal and carrier wave.

2nd, it will be understood by those skilled in the art that embodiments herein can be provided as method, system or computer program production Product.Therefore, the application can use the embodiment in terms of complete hardware embodiment, complete software embodiment or combination software and hardware Form.Moreover, the application can use the computer for wherein including computer usable program code in one or more can use The computer program product that storage medium is implemented on (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) Form.

Claims

A kind of 1. recommendation method of merchandise items model, it is characterised in that including：

Receive the model recommendation request for particular commodity object of the first subscription client transmission；

Comment information of the user to its actual purchase model has been bought according to the particular commodity object, has been determined to described first The model recommendation information for the particular commodity object that user provides；

To model recommendation information described in the first subscription client loopback.
2. the recommendation method of merchandise items model according to claim 1, it is characterised in that described to have bought user and include There is the purchase user of model associations relation with first user.
3. the recommendation method of merchandise items model according to claim 2, it is characterised in that described according to the specific business Product object has bought comment information of the user to its actual purchase model, and determines the spy provided to first user Determine the model recommendation information of merchandise items, including：

Obtain the purchase user that there is model associations relation with first user；

According to default user's selection rule, chosen from the user of purchase that there is model associations relation with first user One specific user；

Comment information according to the specific user to its actual purchase model, determines the model recommendation information.
4. the recommendation method of merchandise items model according to claim 3, it is characterised in that the default user chooses Rule includes：Choose the highest user of the model degree of correlation.
5. the recommendation method of merchandise items model according to claim 3, it is characterised in that the acquisition and described first User has the purchase user of model associations relation, including：

According to the transaction record of the particular commodity object, each user for having bought user of the particular commodity object is obtained Mark；

Type between the user identifier of first user carried according to the model recommendation request, the different user previously generated Number degree of correlation information and each user identifier for having bought user, obtain has model associations relation with first user Purchase user.
6. the recommendation method of merchandise items model according to claim 3, it is characterised in that the acquisition and described first User has the purchase user of model associations relation, including：

Obtain the model comment information collection of the particular commodity object；

According to the model comment information collection, obtain the particular commodity object is provided the model comment information it is each Buy the user identifier of user；

Type between the user identifier of first user carried according to the model recommendation request, the different user previously generated Number degree of correlation information and each user identifier for having bought user, obtain has model associations relation with first user Purchase user.
7. the recommendation method of merchandise items model according to claim 6, it is characterised in that described to obtain the specific business The model comment information collection of product object, including：

Obtain the comment on commodity information collection of the particular commodity object；

By semantic analysis technology, concentrated from the comment on commodity information and extract the comment information in relation to model；

According to the comment information for the related model extracted, the model comment information collection is generated.
8. the recommendation method of the merchandise items model according to claim 5 or 6, it is characterised in that：

The model includes size model；The model degree of correlation information includes size degree of correlation information；

Model degree of correlation information between the different user, generates in the following way：

For each merchandise items in the merchandise items storehouse prestored, following steps are performed：

Obtain the size comment information collection of the merchandise items；

For each size comment information of the merchandise items, the actual purchase of the corresponding user of the size comment information is obtained Buy size；And size appropriate degree information is extracted from the size comment information；It is and suitable according to the size extracted Information and the actual purchase size are spent, determines applicable size of the user to the merchandise items；

According to the corresponding applicable size of each size comment information of the merchandise items, there will be the identical applicable ruler The size degree of correlation between the user of code adds one.
9. the recommendation method of merchandise items model according to claim 8, it is characterised in that the ruler that the basis is extracted Code appropriate degree information and the actual purchase size, and determine applicable size of the user to the merchandise items, using such as Under type：

If the size appropriate degree information is bigger than normal for size, the user is arranged to the applicable size of the merchandise items It is one yard smaller than the actual purchase size；

If the size appropriate degree information is less than normal for size, the user is arranged to the applicable size of the merchandise items It is one yard bigger than the actual purchase size；

If the size appropriate degree information is suitable for size, the user is arranged to the applicable size of the merchandise items The actual purchase size.
10. the recommendation method of merchandise items model according to claim 3, it is characterised in that described according to described specific User determines the model recommendation information to the comment information of its actual purchase model, including：

Model appropriate degree information is extracted from comment information of the specific user to its actual purchase model；

According to the actual purchase model for the model appropriate degree information and the specific user extracted, determine that the model pushes away Recommend information.
11. the recommendation method of merchandise items model according to claim 10, it is characterised in that described from the specific use Family is to extracting model appropriate degree information in the comment information of its actual purchase model, in the following way：

By semantic analysis technology, model appropriate degree is extracted from comment information of the specific user to its actual purchase model Information.
12. the recommendation method of merchandise items model according to claim 10, it is characterised in that：

The model includes size model, and the model appropriate degree information includes size appropriate degree information, the actual purchase type Number include actual purchase size, the model recommendation information includes size recommendation information；

The model appropriate degree information and the actual purchase model of the specific user that the basis is extracted, and determine described Model recommendation information, in the following way：

If the size appropriate degree information is bigger than normal for size, according to the size than small one yard of the actual purchase size, generation The size recommendation information；

If the size appropriate degree information is less than normal for size, according to the size than big one yard of the actual purchase size, generation The size recommendation information；

If the size appropriate degree information is suitable for size, according to the actual purchase size, the size recommendation is generated Breath.
13. the recommendation method of merchandise items model according to claim 1, it is characterised in that described according to described specific Merchandise items have bought comment information of the user to its actual purchase model, and determine to described in first user offer The model recommendation information of particular commodity object, including：

The standard model or user for obtaining first user specify model；

Comment information of the user to its actual purchase model is bought according to each, the model for generating the particular commodity object is closed Appropriate information；

Model is specified according to the model appropriate degree information and the standard model or user, determines the model recommendation information.
14. the recommendation method of merchandise items model according to claim 1, it is characterised in that described according to described specific Merchandise items have bought comment information of the user to its actual purchase model, and determine to described in first user offer The model recommendation information of particular commodity object, including：

The standard model or user for obtaining first user specify model；

According to have purchased the standard model or what user specified model has bought comment information of the user to model, described in generation The model appropriate degree information of particular commodity object；

Model is specified according to the model appropriate degree information and the standard model or user, determines the model recommendation information.
15. the recommendation method of merchandise items model according to claim 1, it is characterised in that the model recommendation information Including at least one of proposed model and model appropriate degree information.
16. the recommendation method of merchandise items model according to claim 15, it is characterised in that the model recommendation information Further include the standard model of first user.
17. the recommendation method of merchandise items model according to claim 1, it is characterised in that the particular commodity object Including clothes, shoes or cap.
18. the recommendation method of merchandise items model according to claim 1, it is characterised in that the model includes size Model, color model, material model or style model.
A kind of 19. recommendation apparatus of merchandise items model, it is characterised in that including：

Request reception unit, the model recommendation request for particular commodity object sent for receiving the first subscription client；

Recommendation information determination unit, comments its actual purchase model for the user that bought according to the particular commodity object By information, the model recommendation information of the particular commodity object provided to first user is determined；

Recommendation information loopback cell, for model recommendation information described in the first subscription client loopback.
20. the recommendation apparatus of merchandise items model according to claim 19, it is characterised in that the recommendation information determines Unit includes：

First user obtains subelement, for obtaining the purchase user for having model associations relation with first user；

Specific user chooses subelement, for according to default user's selection rule, being closed from first user with model A specific user is chosen in the user of purchase of connection relation；

Recommendation information determination subelement, for the comment information according to the specific user to its actual purchase model, determines institute State model recommendation information.
21. the recommendation apparatus of merchandise items model according to claim 20, it is characterised in that the recommendation information determines Subelement includes：

Model appropriate degree information extraction subelement, for being carried from comment information of the specific user to its actual purchase model Take model appropriate degree information；

First recommendation information determination subelement, for according to the model appropriate degree information and the specific user extracted Actual purchase model, determines the model recommendation information.
22. the recommendation apparatus of merchandise items model according to claim 19, it is characterised in that the recommendation information determines Unit includes：

Initial model obtains subelement, and standard model or user for obtaining first user specify model；

Model appropriate degree information generates subelement, for being believed according to each comment for having bought user to its actual purchase model Breath, generates the model appropriate degree information of the particular commodity object；

Recommendation information determination subelement, for specifying type according to the model appropriate degree information and the standard model or user Number, determine the model recommendation information.
23. the recommendation apparatus of merchandise items model according to claim 19, it is characterised in that the recommendation information determines Unit includes：

Initial model obtains subelement, and standard model or user for obtaining first user specify model；

Model appropriate degree information generates subelement, have purchased the standard model for basis or user specifies the purchase of model User generates the model appropriate degree information of the particular commodity object to the comment information of model；

Recommendation information determination subelement, for specifying type according to the model appropriate degree information and the standard model or user Number, determine the model recommendation information.
24. a kind of electronic equipment, it is characterised in that including：

Display；

Processor；And

Memory, for storing the program for the recommendation method for realizing merchandise items model, equipment energization simultaneously passes through the processing After device runs the program of recommendation method of the merchandise items model, following step is performed：Receive the transmission of the first subscription client For the model recommendation request of particular commodity object；According to the user of purchase of the particular commodity object to its actual purchase type Number comment information, determine to first user provide the particular commodity object model recommendation information；To described Model recommendation information described in one subscription client loopback.
A kind of 25. choosing method of merchandise items model, it is characterised in that including：

The model recommendation request for particular commodity object is sent to server；

Receive the model recommendation information of the particular commodity object of the server loopback；

According to the model recommendation information, the selection model of the particular commodity object is chosen.
26. the choosing method of merchandise items model according to claim 25, it is characterised in that：

The model recommendation request includes the user identifier of the first user；

The proposed model that the model recommendation information includes the standard model of first user and provided for first user.
27. the choosing method of merchandise items model according to claim 26, it is characterised in that the particular commodity object Model choose the page include the corresponding first page element of the standard model and the corresponding second page of the proposed model Element.
28. the choosing method of merchandise items model according to claim 27, it is characterised in that described according to the model Recommendation information, and the selection model of the particular commodity object is chosen, including：

The corresponding standard model of the first page element is chosen, as tentatively selected selection model；

The change procedure of selection model is visually represented with the variation pattern of the predetermined display parameter of page elements, by the spy The selection model for determining merchandise items is changed to the corresponding recommendation of the second page element from the tentatively selected selection model Model.
29. the choosing method of merchandise items model according to claim 25, it is characterised in that：

The model recommendation request specifies model including user；

The model recommendation information includes the proposed model provided for the first user.
30. the choosing method of merchandise items model according to claim 29, it is characterised in that the particular commodity object Model choose the page specify the corresponding first page element of model and the proposed model corresponding second including the user Page elements.
31. the choosing method of merchandise items model according to claim 30, it is characterised in that described according to the model Recommendation information, and the selection model of the particular commodity object is chosen, in the following way：

The change procedure of selection model is visually represented with the variation pattern of the predetermined display parameter of page elements, by the spy The selection model for determining merchandise items specifies model to be changed to the corresponding proposed model of the second page element from the user.
32. the choosing method of the merchandise items model according to claim 28 or 31, it is characterised in that the default display Parameter includes at least one of parameter shown below：Background colour, border color, size.
33. the choosing method of merchandise items model according to claim 32, it is characterised in that：

The predetermined display parameter includes background colour；

The variation pattern of the predetermined display parameter with page elements is visually represented to choose the change procedure of model, used Following manner：

According to default transformation period, the background colour of the first page element is gradually moved to the second page element, Background colour as the second page element.
34. the choosing method of merchandise items model according to claim 25, it is characterised in that：

The model recommendation information includes model appropriate degree information；

The method further includes：

Show the model appropriate degree information.
A kind of 35. selecting device of merchandise items model, it is characterised in that including：

Request transmitting unit, for sending the model recommendation request for particular commodity object to server；

Information receiving unit, the model recommendation information of the particular commodity object for receiving the server loopback；

Model chooses unit, for according to the model recommendation information, choosing the selection model of the particular commodity object.
36. a kind of electronic equipment, it is characterised in that including：

Display；

Processor；And

Memory, for storing the program for the choosing method for realizing merchandise items model, equipment energization simultaneously passes through the processing After device runs the program of the choosing method of the merchandise items model, following step is performed：Sent to server and be directed to particular commodity The model recommendation request of object；Receive the model recommendation information of the particular commodity object of the server loopback；According to institute Model recommendation information is stated, chooses the selection model of the particular commodity object.
A kind of 37. selecting system of merchandise items model, it is characterised in that including：According to the business described in the claims 19 The recommendation apparatus of product object model；And the selecting device of the merchandise items model according to the claims 35.
A kind of 38. choosing method of merchandise items model, it is characterised in that including：

Obtain the proposed model of particular commodity object；

The change procedure of selection model is visually represented with the variation pattern of the predetermined display parameter of page elements, in the spy The model for determining merchandise items is chosen in the page, and the selection model of the particular commodity object is corresponding just from first page element The selected selection model of step is changed to the corresponding proposed model of second page element.
39. the choosing method of the merchandise items model according to claim 38, it is characterised in that the predetermined display parameter Include at least one of parameter shown below：Background colour, border color, size.
40. the choosing method of merchandise items model according to claim 39, it is characterised in that：

The predetermined display parameter includes background colour；

The variation pattern of the predetermined display parameter with page elements is visually represented to choose the change procedure of model, used Following manner：

According to default transformation period, the background colour of the first page element is gradually moved to the second page element, Background colour as the second page element.
A kind of 41. selecting device of merchandise items model, it is characterised in that including：

Proposed model acquiring unit, for obtaining the proposed model of particular commodity object；

Model chooses unit, for visually representing to choose model with the variation pattern of the predetermined display parameter of page elements Change procedure, in the model of the particular commodity object chooses the page, by the selection model of the particular commodity object from the The corresponding tentatively selected selection model of one page elements is changed to the corresponding proposed model of second page element.