CN110413870A - Method of Commodity Recommendation, device and server - Google Patents
Method of Commodity Recommendation, device and server Download PDFInfo
- Publication number
- CN110413870A CN110413870A CN201811549358.4A CN201811549358A CN110413870A CN 110413870 A CN110413870 A CN 110413870A CN 201811549358 A CN201811549358 A CN 201811549358A CN 110413870 A CN110413870 A CN 110413870A
- Authority
- CN
- China
- Prior art keywords
- commodity
- browsing
- grouping
- commodities
- session data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the present invention provides a kind of Method of Commodity Recommendation, device and server, obtains historical session data, and historical session data include: the sequence and the corresponding timestamp of each commodity of browsing of the commodity of user's browsing；According to historical session data, obtain at least one grouping of commodities, each grouping of commodities includes: the first commodity and the second commodity, and the commodity amount that the browsing sequence of the second commodity is located at after the browsing sequence of the first commodity, and is spaced between the second commodity and the first commodity is less than preset value；According at least one grouping of commodities, each second commodity corresponding to each first commodity are ranked up, and obtain the corresponding Recommendations list of the first commodity；When determining Recommendations list, the user browsing behavior in historical session data is taken full advantage of, improves the accuracy of commercial product recommending result；When constructing grouping of commodities, it is contemplated that the commodity amount being spaced between the second commodity and the first commodity further improves the accuracy of commercial product recommending result.
Description
Technical field
The present embodiments relate to Internet technical field more particularly to a kind of Method of Commodity Recommendation, device and server.
Background technique
With the high speed development of internet, more and more people select shopping online, and user browses electric business net by terminal
It stands, commodity needed for selecting simultaneously are bought.Method of Commodity Recommendation is widely applied in electric business, and value is that excavating user potentially purchases
Demand is bought, avoids the user effort plenty of time from browsing a large amount of unrelated merchandise newss, user is helped quickly to find in magnanimity commodity
Really necessary commodity improve user's shopping experience.
Currently, the Method of Commodity Recommendation generallyd use, can be divided mainly into content-based recommendation and based on groups of users
Recommend.In content-based recommendation method, according to the interested commodity of user, calculate similar between other commodity and the commodity
Degree recommends it may interested similar commodity according to similarity to user.In recommended method based on groups of users, according to institute
There is user to give a mark the evaluation of commodity, the user with similar preference be grouped by clustering algorithm, obtains groups of users,
It is bought according to user in group, the commodity of browsing, into group, other users are recommended.
However, commodity amount reaches more than one hundred million scales, existing commercial product recommending side in the commercial product recommending scene of electric business platform
Method, commodity accuracy recommended to the user is lower, i.e., commodity recommended to the user may not be that user is interested, so that with
Family experience is lower.
Summary of the invention
The embodiment of the present invention provides a kind of Method of Commodity Recommendation, device and server, for improving commercial product recommending result
Accuracy promotes user experience.
In a first aspect, the embodiment of the present invention provides a kind of Method of Commodity Recommendation, comprising:
Historical session data are obtained, the sequence and browsing that the historical session data include: the commodity of user's browsing are respectively
The corresponding timestamp of commodity；
According to the historical session data, at least one grouping of commodities is obtained, each grouping of commodities includes: the first commodity
It is located at after the browsing sequence of first commodity with the browsing sequence of the second commodity, second commodity, and second quotient
The commodity amount being spaced between product and first commodity is less than preset value；
According at least one described grouping of commodities, corresponding each second commodity of each first commodity are arranged
Sequence obtains the corresponding Recommendations list of first commodity.
Optionally, described at least one grouping of commodities according to, to each first commodity corresponding each described
Two commodity are ranked up, and obtain the corresponding Recommendations list of first commodity, comprising:
For each grouping of commodities, according to the browsing frequency of first commodity, the browsing of grouping of commodities frequency
The commodity amount and first commodity being spaced between secondary, described second commodity and first commodity are to second quotient
The browsing stay time of each commodity between product obtains second commodity relative to first commodity and jumps correlation；
For each first commodity, corresponding each second quotient of first commodity is obtained according to each grouping of commodities
Product, and correlation is jumped relative to first commodity according to each second commodity, each second commodity are arranged
Sequence obtains the corresponding Recommendations list of first commodity.
Optionally, the browsing frequency according to first commodity, the browsing frequency of the grouping of commodities, described second
The commodity amount and first commodity being spaced between commodity and first commodity are to each commodity between second commodity
Browsing stay time, obtain second commodity relative to first commodity and jump correlation, comprising:
According to the browsing frequency of the browsing frequency of first commodity and the grouping of commodities, the second commodity phase is obtained
Probability is jumped for first commodity；
According to the commodity amount being spaced between second commodity and first commodity, it is opposite to obtain second commodity
First commodity jump coefficient；
According to first commodity to the browsing stay time of each commodity between second commodity, second quotient is obtained
Condition jumps duration for first commodity；
According to it is described jump probability, it is described jump coefficient and it is described jump duration, obtain second commodity relative to institute
That states the first commodity jumps correlation.
Optionally, the browsing stay time according to first commodity each commodity into second commodity obtains
Second commodity relative to first commodity jump duration before, further includes:
According to the timestamp in the historical session data, the clear of each commodity in the historical session data is obtained
Look at stay time.
Optionally, described according to the historical session data, obtain at least one grouping of commodities, comprising:
According to the timestamp, the historical session data are cut, obtain at least one session, each session
In include the sequence of commodity that user browses in the session；
For each session, each commodity and subsequent adjacent N number of commodity combination of two are obtained into the commodity group
It closes, N is the preset value.
Optionally, it is described each commodity and subsequent adjacent N number of commodity combination of two are obtained into the grouping of commodities after,
Further include:
According to the historical session data, the browsing frequency of each commodity is obtained, and obtains the browsing of each grouping of commodities
The frequency；
According to the browsing frequency of each commodity, the corresponding grouping of commodities of the commodity for being unsatisfactory for the first preset condition is deleted；
According to the browsing frequency of each grouping of commodities, the grouping of commodities for being unsatisfactory for the second preset condition is deleted.
Optionally, described according to the timestamp, before being cut to the historical session data, further includes:
Remove the noise data in the historical session data.
Optionally, the method also includes:
Goods browse request is obtained, the goods browse request is used to indicate user and requests the first commodity of browsing；
It is requested according to the goods browse, obtains the Recommendations list corresponding with first commodity；
Recommend the commodity in the Recommendations list to the user.
It optionally, further include timestamp in the goods browse request；
After the goods browse request for obtaining user's transmission, the method also includes:
First commodity and the timestamp are recorded in the historical session data.
Second aspect, the embodiment of the present invention provide a kind of device for recommending the commodity, comprising:
First obtains module, and for obtaining historical session data, the historical session data include: the commodity of user's browsing
Sequence and the corresponding timestamp of each commodity of browsing；
Second obtains module, for obtaining at least one grouping of commodities, each commodity according to the historical session data
Combination includes: the first commodity and the second commodity, and the browsing sequence of second commodity is located at the browsing sequence of first commodity
Later, and between second commodity and first commodity commodity amount being spaced is less than preset value；
Generation module, it is corresponding each described to each first commodity at least one grouping of commodities according to
Second commodity are ranked up, and obtain the corresponding Recommendations list of first commodity.
Optionally, the generation module is specifically used for:
For each grouping of commodities, according to the browsing frequency of first commodity, the browsing of grouping of commodities frequency
The commodity amount and first commodity being spaced between secondary, described second commodity and first commodity are to second quotient
The browsing stay time of each commodity between product obtains second commodity relative to first commodity and jumps correlation；
For each first commodity, corresponding each second quotient of first commodity is obtained according to each grouping of commodities
Product, and correlation is jumped relative to first commodity according to each second commodity, each second commodity are arranged
Sequence obtains the corresponding Recommendations list of first commodity.
Optionally, the generation module is specifically used for:
According to the browsing frequency of the browsing frequency of first commodity and the grouping of commodities, the second commodity phase is obtained
Probability is jumped for first commodity；
According to the commodity amount being spaced between second commodity and first commodity, it is opposite to obtain second commodity
First commodity jump coefficient；
According to first commodity to the browsing stay time of each commodity between second commodity, second quotient is obtained
Condition jumps duration for first commodity；
According to it is described jump probability, it is described jump coefficient and it is described jump duration, obtain second commodity relative to institute
That states the first commodity jumps correlation.
Optionally, the second acquisition module is also used to: according to the timestamp in the historical session data, being obtained
The browsing stay time of each commodity in the historical session data.
Optionally, the second acquisition module is specifically used for: according to the timestamp, carrying out to the historical session data
Cutting, obtains at least one session, includes the sequence for the commodity that user browses in the session in each session；
For each session, each commodity and subsequent adjacent N number of commodity combination of two are obtained into the commodity group
It closes, N is the preset value.
Optionally, the second acquisition module is also used to:
According to the historical session data, the browsing frequency of each commodity is obtained, and obtains the browsing of each grouping of commodities
The frequency；
According to the browsing frequency of each commodity, the corresponding grouping of commodities of the commodity for being unsatisfactory for the first preset condition is deleted；
According to the browsing frequency of each grouping of commodities, the grouping of commodities for being unsatisfactory for the second preset condition is deleted.
Optionally, the second acquisition module is also used to:
Remove the noise data in the historical session data.
Optionally, described device further include: recommending module；
The first acquisition module is also used to: obtaining goods browse request, the goods browse request is used to indicate user
Request the first commodity of browsing；
The recommending module, for according to the goods browse request, obtain it is corresponding with first commodity described in push away
Items list is recommended, and recommends the commodity in the Recommendations list to the user.
It optionally, further include timestamp in the goods browse request；
The first acquisition module is also used to: the historical session number is recorded in first commodity and the timestamp
In.
The third aspect, the embodiment of the present invention provide a kind of server, comprising: at least one processor and memory；
The memory stores computer executed instructions；
At least one described processor executes the computer executed instructions of memory storage so that it is described at least one
Processor executes such as the described in any item methods of first aspect.
Fourth aspect, the embodiment of the present invention provide a kind of computer readable storage medium, the computer-readable storage medium
It is stored with computer executed instructions in matter, when processor executes the computer executed instructions, realizes as first aspect is any
Method described in.
Method of Commodity Recommendation, device and server provided in an embodiment of the present invention obtain historical session data, the history
Session data includes: the sequence and the corresponding timestamp of each commodity of browsing of the commodity of user's browsing；According to the historical session
Data, obtain at least one grouping of commodities, and each grouping of commodities includes: the first commodity and the second commodity, second commodity
Browsing sequence be located at after the browsing sequence of first commodity, and be spaced between second commodity and first commodity
Commodity amount be less than preset value；It is corresponding each described to each first commodity according at least one described grouping of commodities
Second commodity are ranked up, and obtain the corresponding Recommendations list of first commodity；Determining the corresponding recommendation of the first commodity
When items list, the user browsing behavior in historical session data is taken full advantage of, improves the accuracy of commercial product recommending result；
In addition, when constructing grouping of commodities, it is contemplated that the commodity amount being spaced between the second commodity and the first commodity, that is to say, that same
When consider that adjacent hop transfers the registration of Party membership, etc. from one unit to another and interval jumps relationship so that more to the user browsing behavior use of information in historical session data
Add sufficiently, further improves the accuracy of commercial product recommending result.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention without any creative labor, may be used also for those of ordinary skill in the art
To obtain other drawings based on these drawings.
Fig. 1 is a kind of schematic diagram for application scenarios that the embodiment of the present invention may be applicable in；
Fig. 2 is the flow diagram one of Method of Commodity Recommendation provided in an embodiment of the present invention；
Fig. 3 is commercial product recommending process schematic provided in an embodiment of the present invention；
Fig. 4 determines the corresponding Recommendations list of the first commodity according to each grouping of commodities to be provided in an embodiment of the present invention
Flow diagram；
Fig. 5 is the flow diagram provided in an embodiment of the present invention for obtaining and jumping the phase same sex；
Fig. 6 is the flow diagram provided in an embodiment of the present invention according to historical session data acquisition grouping of commodities；
Fig. 7 is the flow diagram two of Method of Commodity Recommendation provided in an embodiment of the present invention；
Fig. 8 is the structural schematic diagram one of the device for recommending the commodity provided in an embodiment of the present invention；
Fig. 9 is the structural schematic diagram two of the device for recommending the commodity provided in an embodiment of the present invention；
Figure 10 is the hardware structural diagram of server provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Description and claims of this specification and term " first ", " second ", " third " " in above-mentioned attached drawing
The (if present)s such as four " are to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should manage
The data that solution uses in this way are interchangeable under appropriate circumstances, so that the embodiment of the present invention described herein for example can be to remove
Sequence other than those of illustrating or describe herein is implemented.In addition, term " includes " and " having " and theirs is any
Deformation, it is intended that cover it is non-exclusive include, for example, containing the process, method of a series of steps or units, system, production
Product or equipment those of are not necessarily limited to be clearly listed step or unit, but may include be not clearly listed or for this
A little process, methods, the other step or units of product or equipment inherently.
Commercial product recommending scene of the embodiment of the present invention suitable for internet area, specifically, browsed according to user
Current commodity, may other interested commodity to user recommended user.Below with reference to Fig. 1 to the possible of the embodiment of the present invention
Application scenarios are introduced.Fig. 1 is a kind of schematic diagram for application scenarios that the embodiment of the present invention may be applicable in.As shown in Figure 1, this
The commercial product recommending system that inventive embodiments provide includes terminal and server.Wherein, terminal is to input information for user and show
Show any electronic equipment of output result, including but not limited to: computer, smart phone, laptop, platform computer, intelligence
Energy wearable device etc..Server is the electronic equipment for executing commercial product recommending.
Method of Commodity Recommendation provided in an embodiment of the present invention is applicable to need to carry out any scene of commercial product recommending, including
But it is not limited to: commercial product recommending, commercial product recommending of physical stores of electric business platform etc..In a kind of possible application scenarios, with electric business
For the commercial product recommending of platform, user browses commodity by terminal access electric business platform, and in electric business platform, and terminal is clear by commodity
Request of looking at is sent to server, and server requests the possible interested commodity of prediction user according to goods browse, to user terminal
Recommended, so that user carries out commodity selection, purchase.
It should be noted that the commodity in the embodiment of the present invention refer to that sensu lato commodity, commodity can be tangible quotient
Product, or intangible goods can also be electronic data commodity.Tangible goods for example can be electric business platform or entity quotient
The commodity that product are sold；Intangible goods for example can be service class commodity, such as insurance products, financial product etc.；Electronic data quotient
Product include but is not limited to music commodity, video commodity, news commodity etc..
In the application scenarios of the embodiment of the present invention, the scene based on commodity shopping can be, can also be clear based on commodity
The scene look at.For the scene based on commodity shopping, illustratively, when electric business platform is bought goods, user clicks clear user
Look at a certain commodity when, the server of electric business platform recommends it interested according to the historical viewings behavior of user to user
Other commodity.For the scene based on goods browse, such as: user is when browsing music data, when user selects browsing a certain
When song, recommendation server recommends it may other interested music according to the historical viewings behavior of user to user.
The embodiment of the present invention takes full advantage of historical session data when determining the corresponding Recommendations list of the first commodity
In user browsing behavior, improve the accuracy of commercial product recommending result；In addition, when constructing grouping of commodities, it is contemplated that second
The commodity amount being spaced between commodity and the first commodity, that is to say, that while considering that adjacent hop transfers the registration of Party membership, etc. from one unit to another and interval jumps relationship,
So that it is more abundant to the user browsing behavior use of information in historical session data, further improve commercial product recommending result
Accuracy.
Technical solution of the present invention is described in detail with specifically embodiment below.These specific implementations below
Example can be combined with each other, and the same or similar concept or process may be repeated no more in some embodiments.
Fig. 2 is the flow diagram one of Method of Commodity Recommendation provided in an embodiment of the present invention, and the method for the present embodiment can be with
It is executed by the server in Fig. 1, as shown in Fig. 2, the method for the present embodiment, comprising:
S201: obtain historical session data, the historical session data include: user browsing commodity sequence and
Browse the corresponding timestamp of each commodity.
Wherein, historical session data are that server identification user browses the record data formed after commodity behavior.Specifically,
Server can create a session for each user, and the behavior of commodity is browsed for recording the user；Server can also be
The browsing behavior of each user in different time period creates different sessions, such as: it is created for the browsing behavior in user's morning
A session is built, for browsing behavior one session of creation in user's afternoon.It should be understood that server can also have others
The mode of session is created, the present invention is not especially limit this.
Specifically, user's browsing commodity behavior specifically can be user and click some commodity in the client of electric business platform
Behavior, when user is when electric business platform client clicks some commodity, server identifies that the user is browsing the commodity,
And by the goods browse behavior record of user into historical session data.
In the present embodiment, historical session data may include: that the sequence of the commodity of user's browsing and user browse each quotient
The corresponding timestamp of product.
Fig. 3 is commercial product recommending process schematic provided in an embodiment of the present invention, is illustrated below with reference to Fig. 3 citing.One
In kind optional embodiment, as shown in figure 3, the form of historical session data can for [commodity A, time 1], [commodity B, when
Between 2], [commodity C, time 3], [commodity D, time 4] ..., that is, user has browsed commodity A in the time 1, the time 2 browse
Commodity B, have browsed commodity C in the time 3, have browsed commodity D, etc. in the time 4.
It should be noted that the historical session data obtained in the step, can be the corresponding historical session of all users
Data can also be the corresponding historical session data of certain customers, and the present invention is not especially limit this.It is a kind of optional
Embodiment in, can in advance according to the feature of user to all users carry out group division, when need to certain user carry out
When commercial product recommending, it can only obtain the corresponding historical session data of the user in the affiliated group of the user and analyze.
In a kind of optional embodiment, the historical session data are by cutting to goods browse log information
It obtains, it includes user in a conversation procedure, each historical session data at this that each historical session data, which correspond to user,
The sequence and the corresponding timestamp of each commodity of browsing of the commodity browsed in secondary conversation procedure.
Specifically, user is when browsing commodity, server can the goods browse behavior to user record, form browsing
Log information.Due to including the multiple conversation procedure of user in view log, in the present embodiment, it can pass through
View log is cut, multiple historical session data are obtained.More specifically, according to user browse commodity timestamp and
The preset period carries out cutting to user browsing behavior and generates historical session data.Such as: it is clicked in setting time threshold value T1
Commodity belong to same session, can value T1=60min.It should be understood that by cutting after each historical session data in
Sequence and the corresponding timestamp of each commodity of browsing including the commodity that user browses in a session.
S202: according to the historical session data, at least one grouping of commodities is obtained, each grouping of commodities includes:
The browsing sequence of one commodity and the second commodity, second commodity is located at after the browsing sequence of first commodity, and described
The commodity amount being spaced between second commodity and first commodity is less than preset value.
Specifically, the user according to recorded in historical session data browses commodity behavior, at least one commodity group is obtained
It closes, wherein each grouping of commodities instruction is to jump relationship between commodity when user browses commodity.
In the present embodiment, each grouping of commodities includes: the first commodity and the second commodity, the browsing sequence of second commodity
After the browsing sequence of first commodity, that is to say, that each grouping of commodities indicated is that user is browsing first
After commodity, the second commodity have also been browsed.
It is illustrated continuing with Fig. 3, in a kind of optional embodiment, historical session data according to Fig.3, this
The grouping of commodities that embodiment is got can be { (commodity A, commodity B), (commodity B, commodity C), (commodity C, commodity D) }.
It should be understood that since user usually has similitude or association between the commodity browsed in the same session
Property, although for example, having recorded user in historical session data to browse the sequence of commodity is commodity A, commodity B, commodity C, quotient
Product D, but due to the similitude of commodity B, commodity C and commodity D, user be likely to browse after browsing commodity A commodity C or
Commodity D.
Therefore, the embodiment of the present invention excavates possible grouping of commodities also according to historical session data, will be under satisfaction
The first commodity and the second commodity for stating condition form grouping of commodities: the browsing sequence of second commodity is located at first commodity
Browsing sequence after, and the commodity amount being spaced between second commodity and first commodity be less than preset value.Also
It is to say, when constructing grouping of commodities, not only considers adjacent to jump relationship, it is also contemplated that interval jumps relationship, so that obtaining
Grouping of commodities it is more abundant so that more accurate according to the commercial product recommending result that grouping of commodities obtains.
It is illustrated continuing with Fig. 3, in a kind of optional embodiment, when the preset value is 3, as shown in figure 3,
The embodiment of the present invention obtain grouping of commodities can also for (commodity A, commodity B), (commodity A, commodity C), (commodity A, commodity D),
(commodity B, commodity C), (commodity B, commodity D), (commodity C, commodity D) }.
It should be noted that the embodiment of the present invention according to the method for historical session data acquisition grouping of commodities for not making to have
Body limits, as long as the grouping of commodities obtained meets following conditions: the browsing sequence of second commodity is located at described first
After the browsing sequence of commodity, and the commodity amount being spaced between second commodity and first commodity is less than preset value.
It should be understood that the embodiment of the present invention is not especially limited the specific value of the preset value, it can basis
Actual conditions are rationally arranged, and the example above is merely illustrative.
S203: according at least one described grouping of commodities, each second commodity corresponding to each first commodity
It is ranked up, obtains the corresponding Recommendations list of first commodity.
It should be understood that jumping relationship, example between commodity when user browses commodity since each grouping of commodities indicates
Such as: (commodity A, commodity B) is indicated after user browses commodity A and is jumped to commodity B, illustrates that commodity B is the interested commodity of user,
It therefore, can be to user's Recommendations B when user browses commodity A.
On the basis of above-mentioned steps, after getting each grouping of commodities, it can be respectively obtained each according to each grouping of commodities
Corresponding all second commodity of first commodity, using these second commodity as the corresponding Recommendations list of the first commodity.When with
When family browses the first commodity, then recommend these the second commodity to user, to ensure that commodity recommended to the user are user's senses
Interest, the accuracy of Recommendations is improved, user experience is promoted.
Continue with and be illustrated in conjunction with Fig. 3, when the grouping of commodities got can for (commodity A, commodity B),
(commodity A, commodity C), (commodity A, commodity D), (commodity B, commodity C), (commodity B, commodity D), (commodity C, commodity D) } when, such as scheme
Shown in 3, it can determine that corresponding each second commodity of commodity A are commodity B, commodity C, commodity D, that is, the corresponding recommendation quotient of commodity A
Product list is { commodity B, commodity C, commodity D }, and similarly, the corresponding Recommendations list of commodity B is { commodity C, commodity D }, commodity C
Corresponding Recommendations list is { commodity D }.
It therefore, then can be to user's Recommendations B, commodity C, commodity D when user browses commodity A；When user browses quotient
It, then can be to user's Recommendations C, commodity D when product B；It, then can be to user's Recommendations D when user browses commodity C.
It, can also be to each second quotient after determining corresponding each second commodity of each first commodity in the present embodiment
Product are ranked up, using each second commodity after sequence as Recommendations list.
It should be noted that the embodiment of the present invention is not especially limited the method to sort to each second commodity,
It can be ranked up according to the correlation that jumps between each second commodity and first commodity, it can also be according to each second commodity
Attribute be ranked up, such as: the attention rate of commodity, the sales volume of commodity, the positive rating of commodity, price of commodity etc..
In addition, Method of Commodity Recommendation provided in an embodiment of the present invention, determines Recommendations list according to historical session data
Process can be online execute or off-line execution.In a kind of optional embodiment, server is according to preset update
Period, the S201 to S203 in off-line execution the present embodiment obtain the corresponding Recommendations list of each commodity and are stored,
When server, which recognizes user, browses the commodity, recommended according to the Recommendations list of storage to user.It is another optional
In embodiment, when server recognizes user's browsing a certain commodity, the online S201 to S203 executed in the present embodiment is obtained
The corresponding Recommendations list of the commodity, and recommend to user.
Method of Commodity Recommendation provided in an embodiment of the present invention, obtains historical session data, and the historical session data include:
The sequence and the corresponding timestamp of each commodity of browsing of the commodity of user's browsing；According to the historical session data, obtain at least
One grouping of commodities, each grouping of commodities include: the first commodity and the second commodity, and the browsing sequence of second commodity is located at
After the browsing sequence of first commodity, and the commodity amount being spaced between second commodity and first commodity is less than
Preset value；According at least one described grouping of commodities, corresponding each second commodity of each first commodity are arranged
Sequence obtains the corresponding Recommendations list of first commodity；When determining the corresponding Recommendations list of the first commodity, sufficiently
The user browsing behavior in historical session data is utilized, improves the accuracy of commercial product recommending result；In addition, in building commodity
When combination, it is contemplated that the commodity amount being spaced between the second commodity and the first commodity, that is to say, that while considering adjacent to jump pass
System and interval jump relationship, so that it is more abundant to the user browsing behavior use of information in historical session data, further mention
The high accuracy of commercial product recommending result.
On the basis of the above embodiments, one kind below with reference to a specific embodiment detailed description S203 is optional
Embodiment.Fig. 4 determines the corresponding Recommendations list of the first commodity according to each grouping of commodities to be provided in an embodiment of the present invention
Flow diagram, as shown in figure 4, this method comprises:
S401: be directed to each grouping of commodities, according to the browsing frequency of first commodity, the grouping of commodities it is clear
The commodity amount being spaced between the frequency, second commodity and first commodity and first commodity are look to described
The browsing stay time of each commodity between two commodity obtains second commodity relative to first commodity and jumps correlation
Property.
Specifically, determining that the second commodity jump phase relative to the first commodity according to each grouping of commodities in the present embodiment
Guan Xing, wherein jump correlation for characterizing after user browses the first commodity and jump to a possibility that browsing the second commodity.This reality
Apply in example, jumping correlation can obtain according to parameters described below: the browsing frequencys of first commodity, the grouping of commodities it is clear
The commodity amount being spaced between the frequency, second commodity and first commodity and first commodity are look to described
The browsing stay time of each commodity between two commodity.
A kind of optional embodiment for jumping correlation is obtained below with reference to a specific example detailed description.Fig. 5
The flow diagram of the phase same sex is jumped for acquisition provided in an embodiment of the present invention, as shown in figure 5, this method comprises:
S4011: according to the browsing frequency of the browsing frequency of first commodity and the grouping of commodities, described second is obtained
Commodity jump probability relative to first commodity.
For convenience, the present embodiment for grouping of commodities (commodity A, commodity B) to be described, i.e., the first commodity are
Commodity A, the second commodity are commodity B.
Specifically, historical session data can be the corresponding historical session data of multiple users, and each user can be with
Corresponding multiple historical session data.According to all historical session data, the browsing frequency for obtaining commodity A can be counted；Then,
According to historical session data, after obtaining all possible grouping of commodities, can count to obtain grouping of commodities (commodity A, commodity B)
Browse the frequency.
Assuming that the browsing frequency of commodity A is count (A), the browsing frequency of grouping of commodities (commodity A, commodity B) is count
(AB), then in a kind of optional embodiment, commodity B can be with relative to the probability that jumps of commodity A are as follows: and P (B | A)=count
(AB)/count(A)。
S4012: according to the commodity amount being spaced between second commodity and first commodity, second quotient is obtained
Condition jumps coefficient to first commodity.
Still with grouping of commodities (commodity A, commodity B) for, it is assumed that commodity B and commodity A is spaced in historical session data
Commodity amount be λ, that is, user has browsed λ commodity after having browsed commodity A, has then browsed commodity B, then commodity B
Jumping coefficient and can be indicated using F (λ) relative to commodity A.
Wherein, F () representative function converts, in a kind of optional embodiment,
S4013: according to first commodity to the browsing stay time of each commodity between second commodity, described in acquisition
Second commodity jump duration relative to first commodity.
It optionally, can also include: to be obtained according to the timestamp in the historical session data before S4013
The browsing stay time of each commodity in the historical session data.
Specifically, when the browsing stay time of each commodity can use the browsing of two neighboring commodity in historical session data
Between the difference dt that stabsi=ti+1-tiTo identify.Wherein, tiFor the browsing time stamp of i-th of commodity in historical session data.
In specific implementation process, for the last one commodity in historical session data, due to the commodity do not have it is subsequent clear
Look at commodity, then the browsing stay time of the commodity can be set to fixed value C1.In addition, in order to avoid due to user's cause specific
The too long situation of the browsing stay time of caused individual goods occurs, can the browsing stay time to each commodity can be set
Threshold value C2, as the browsing stay time dt of certain commodityiWhen more than threshold value C2, then the browsing stay time that the commodity are directly arranged is
C2.That is, the browsing stay time of each commodity can be indicated using following formula:
It should be understood that the value of C1 and C2 can be rationally arranged in conjunction with actual conditions, such as: C1=
10s, C2=300s.
After the browsing stay time for getting each commodity in historical session data, it can be stopped according to the browsing of each commodity
Duration is stayed, determines in grouping of commodities the first commodity to the browsing stay time of commodity each between the second commodity.
Be exemplified below, it is assumed that historical session data be [commodity A, time 1], [commodity C, time 2], [commodity B, when
Between 3], then the corresponding browsing stay time of commodity A be dtA2- time=time 1, the corresponding browsing stay time of commodity C are dtC
The 3- time=time 2.It, then can be by t=dt for grouping of commodities (commodity A, commodity B)A+dtCBe determined as commodity A to commodity B it
Between each commodity browsing stay time.
Furthermore it is also possible to the browsing stay time to above-mentioned acquisition models, in a kind of optional embodiment, T=H
(t), wherein H () indicates modeling functions, such as:Or H (x)=log (x).
S4014: according to it is described jump probability, it is described jump coefficient and it is described jump duration, obtain the second commodity phase
Correlation is jumped for first commodity.
Specifically, each grouping of commodities is directed to, according to the second commodity obtained in above-mentioned steps relative to the first commodity
Probability P (B | A) is jumped, coefficient F (λ) is jumped, jumps duration T, available second commodity jump phase relative to the first commodity
Guan Xing.
In a kind of optional embodiment, jumping correlation can be indicated using following formula:
S (B | A)=P (B | A) * F (λ) * T
S402: being directed to each first commodity, and it is corresponding each described to obtain first commodity according to each grouping of commodities
Second commodity, and correlation is jumped relative to first commodity according to each second commodity, to each second commodity
It is ranked up, obtains the corresponding Recommendations list of first commodity.
In the present embodiment, for corresponding each second commodity of each first commodity, can according to each second commodity relative to
First commodity jump correlation, are ranked up to each second commodity, obtain the corresponding Recommendations list of the first commodity, guarantee
The reasonability and accuracy of commodity sequence in Recommendations list.
It is exemplified below, it is assumed that according to grouping of commodities { (commodity A, commodity B), (commodity A, commodity C), (commodity A, commodity
D) }, the corresponding Recommendations of commodity A obtained are respectively commodity B, commodity C, commodity D.Also, the commodity determined according to S401
B, commodity C and commodity D is respectively s (B | A), s (C | A), s (D | A) relative to the correlation that jumps of commodity A, then can be according to s (B
| A), s (C | A), s (D | A) commodity B, commodity C and commodity D are ranked up according to descending sequence, by the quotient after sequence
Product are as Recommendations list.
In the present embodiment, determining that the second commodity when jumping correlation, jump relative to the first commodity in addition to considering
Probability, jump system, it is also contemplated that jump duration, that is, consider stay time when user browses commodity.It should be understood that with
The stay time information of family browsing commodity is able to reflect user to the preference of commodity, therefore, comprehensively considers user and browses quotient
The stay time of product, so that the preference for jumping correlation and being more in line with user determined, promotes the accurate of commercial product recommending result
Property, and promote user experience.
On the basis of the above embodiments, one kind below with reference to a specific embodiment detailed description S202 is optional
Embodiment.Fig. 6 is the flow diagram provided in an embodiment of the present invention according to historical session data acquisition grouping of commodities, is such as schemed
Shown in 6, this method comprises:
S601: according to the timestamp, cutting the historical session data, obtains at least one session, each institute
State the sequence of the commodity browsed in the session in session including user.
Specifically, may include the multiple conversation procedure of user in each historical session data, it therefore, can in the present embodiment
To cut to historical session data, the session sequence of user is generated.When specific cutting, the commodity time is browsed according to user
Stamp carries out cutting to user browsing behavior and generates user conversation sequence, and the commodity clicked in setting time threshold value T1 belong to same
Session, can value T1=60min.
It wherein, include the sequence for the commodity that user browses in a session in each session after cutting.
It optionally, can also include: in the removal historical session data before being cut to historical session data
Noise data.Specifically, the commodity data that preset condition is unsatisfactory in historical session data is deleted, such as: goods number
Be not inconsistent normally, browsing time stamp it is abnormal etc..
S602: being directed to each session, and each commodity and subsequent adjacent N number of commodity combination of two are obtained the quotient
Product combination, N are the preset value.
Specifically, for the commodity sequence in each session after cutting, by each commodity and subsequent adjacent N number of commodity
Combination of two obtains the grouping of commodities.
It is exemplified below, it is assumed that the commodity sequence in some session is { commodity A, commodity B, commodity C, commodity D, commodity
E, commodity F }, N value is 4, then the process for obtaining grouping of commodities is as follows:
Commodity A and subsequent 4 commodity (commodity B, commodity C, commodity D, commodity E) combination of two are obtained for commodity A
It is (commodity A, commodity B), (commodity A, commodity C), (commodity A, commodity D), (commodity A, commodity E) to grouping of commodities.
Commodity B and subsequent 4 commodity (commodity C, commodity D, commodity E, commodity F) combination of two are obtained for commodity B
It is (commodity B, commodity C), (commodity B, commodity D), (commodity B, commodity E), (commodity B, commodity F) to grouping of commodities.
Commodity C, commodity D, the acquisition process of the corresponding grouping of commodities of commodity E are similar, and details are not described herein again.
S603: according to the historical session data, the browsing frequency of each commodity is obtained, and obtains each grouping of commodities
Browse the frequency.
S604: according to the browsing frequency of each commodity, the corresponding grouping of commodities of the commodity for being unsatisfactory for the first preset condition is deleted
It removes；According to the browsing frequency of each grouping of commodities, the grouping of commodities for being unsatisfactory for the second preset condition is deleted.
In the present embodiment, after getting grouping of commodities, the grouping of commodities of low frequency time can be deleted, thus remaining high frequency time
Grouping of commodities.Specifically, given threshold T2, determines that the browsing frequency of commodity is less than the low frequency commodity of T2, will include low frequency
The grouping of commodities of commodity is deleted.Then, given threshold T3, the grouping of commodities by the browsing frequency of grouping of commodities less than T3 are deleted.
So that remaining grouping of commodities is the grouping of commodities of high frequency time, in turn, determined according to the grouping of commodities of high frequency time
Recommendations list is more accurate.
It should be noted that the embodiment of the present invention is not made to have for the method for obtaining the commodity frequency and the grouping of commodities frequency
Body limits, and in a kind of optional embodiment, can use frequent item set algorithm, such as: it is obtained using a frequent set algorithm
The browsing frequency of each commodity obtains the browsing frequency of each grouping of commodities using frequency binomial set algorithm.
Fig. 7 is the flow diagram two of Method of Commodity Recommendation provided in an embodiment of the present invention, as shown in fig. 7, the present invention is real
The method for applying example, comprising:
S701: obtaining goods browse request, and the goods browse request is used to indicate user and requests the first commodity of browsing.
Specifically, the behavior that can click commodity picture according to user obtains goods browse request, other can also be passed through
Mode obtains goods browse request, and the embodiment of the present invention is not especially limited.
S702: requesting according to the goods browse, obtains the Recommendations list corresponding with first commodity.
S703: Xiang Suoshu user recommends the commodity in the Recommendations list.
The embodiment of the present invention is when recognizing user and browsing the first commodity, according to first obtained in above-described embodiment
The corresponding Recommendations list of commodity carries out commercial product recommending to user, ensure that the accuracy of commercial product recommending result, help to mention
Rise user experience.
In addition, further including timestamp in the goods browse request, getting commodity in a kind of optional embodiment
After browse request, can also include:
S704: first commodity and the timestamp are recorded in the historical session data.
It is recorded by the browsing commodity behavior to user, in case the subsequent historical session data according to continuous renewal are more
New Recommendations list, ensure that the real-time and accuracy of Recommendations list, promotes the experience of user.
Fig. 8 is the structural schematic diagram one of the device for recommending the commodity provided in an embodiment of the present invention, as shown in figure 8, the present invention is real
Apply the device for recommending the commodity 800 of example offer, comprising: first, which obtains module 801, second, obtains module 802 and generation module 803.
Wherein, first module 801 is obtained, for obtaining historical session data, the historical session data include: that user is clear
The sequence and the corresponding timestamp of each commodity of browsing for the commodity look at；
Second obtains module 802, each described for obtaining at least one grouping of commodities according to the historical session data
Grouping of commodities includes: the first commodity and the second commodity, and the browsing sequence of second commodity is located at the browsing of first commodity
After sequence, and the commodity amount being spaced between second commodity and first commodity is less than preset value；
Generation module 803, for according at least one described grouping of commodities, each institute corresponding to each first commodity
It states the second commodity to be ranked up, obtains the corresponding Recommendations list of first commodity.
The device for recommending the commodity of the present embodiment, can be used for executing Method of Commodity Recommendation as shown in Figure 2, realization principle and
Technical effect is similar, and details are not described herein again.
Fig. 9 is the structural schematic diagram two of the device for recommending the commodity provided in an embodiment of the present invention, the base of embodiment shown in Fig. 8
On plinth, the device for recommending the commodity provided in this embodiment further include: recommending module 804.
Optionally, the generation module 803 is specifically used for:
For each grouping of commodities, according to the browsing frequency of first commodity, the browsing of grouping of commodities frequency
The commodity amount and first commodity being spaced between secondary, described second commodity and first commodity are to second quotient
The browsing stay time of each commodity between product obtains second commodity relative to first commodity and jumps correlation；
For each first commodity, corresponding each second quotient of first commodity is obtained according to each grouping of commodities
Product, and correlation is jumped relative to first commodity according to each second commodity, each second commodity are arranged
Sequence obtains the corresponding Recommendations list of first commodity.
Optionally, the generation module 803 is specifically used for:
According to the browsing frequency of the browsing frequency of first commodity and the grouping of commodities, the second commodity phase is obtained
Probability is jumped for first commodity；
According to the commodity amount being spaced between second commodity and first commodity, it is opposite to obtain second commodity
First commodity jump coefficient；
According to first commodity to the browsing stay time of each commodity between second commodity, second quotient is obtained
Condition jumps duration for first commodity；
According to it is described jump probability, it is described jump coefficient and it is described jump duration, obtain second commodity relative to institute
That states the first commodity jumps correlation.
Optionally, the second acquisition module 802 is also used to: according to the timestamp in the historical session data,
Obtain the browsing stay time of each commodity in the historical session data.
Optionally, the second acquisition module 802 is specifically used for: according to the timestamp, to the historical session data
It is cut, obtains at least one session, include the sequence for the commodity that user browses in the session in each session；
For each session, each commodity and subsequent adjacent N number of commodity combination of two are obtained into the commodity group
It closes, N is the preset value.
Optionally, the second acquisition module 802 is also used to:
According to the historical session data, the browsing frequency of each commodity is obtained, and obtains the browsing of each grouping of commodities
The frequency；
According to the browsing frequency of each commodity, the corresponding grouping of commodities of the commodity for being unsatisfactory for the first preset condition is deleted；
According to the browsing frequency of each grouping of commodities, the grouping of commodities for being unsatisfactory for the second preset condition is deleted.
Optionally, the second acquisition module 802 is also used to:
Remove the noise data in the historical session data.
Optionally, the first acquisition module 801 is also used to: obtaining goods browse request, the goods browse request is used
The first commodity of browsing are requested in instruction user；
The recommending module 804 obtains corresponding with first commodity described for being requested according to the goods browse
Recommendations list, and recommend the commodity in the Recommendations list to the user.
It optionally, further include timestamp in the goods browse request；
The first acquisition module 801 is also used to: the history meeting is recorded in first commodity and the timestamp
It talks about in data.
The device for recommending the commodity provided in this embodiment can be used for executing the commercial product recommending side in any of the above-described embodiment of the method
Method, it is similar that the realization principle and technical effect are similar, and details are not described herein again.
Figure 10 is the hardware structural diagram of server provided in an embodiment of the present invention, and as described in Figure 10, the present embodiment mentions
The server 1000 of confession, comprising: at least one processor 1001 and memory 1002.Wherein, processor 1001, memory 1002
And it is connected by bus 1003.
During specific implementation, at least one processor 1001 executes the computer that the memory 1002 stores and executes
Instruction, so that at least one processor 1001 executes the Method of Commodity Recommendation in any of the above-described embodiment of the method.
The specific implementation process of processor 1001 can be found in above method embodiment, implementing principle and technical effect class
Seemingly, details are not described herein again for the present embodiment.
In above-mentioned embodiment shown in Fig. 10, it should be appreciated that processor can be central processing unit (English:
Central Processing Unit, referred to as: CPU), can also be other general processors, digital signal processor (English:
Digital Signal Processor, referred to as: DSP), specific integrated circuit (English: Application Specific
Integrated Circuit, referred to as: ASIC) etc..General processor can be microprocessor or the processor is also possible to
Any conventional processor etc..Hardware processor can be embodied directly in conjunction with the step of invention disclosed method to have executed
At, or in processor hardware and software module combination execute completion.
Memory may include high speed RAM memory, it is also possible to and it further include non-volatile memories NVM, for example, at least one
Magnetic disk storage.
Bus can be industry standard architecture (Industry Standard Architecture, ISA) bus, outer
Portion's apparatus interconnection (Peripheral Component, PCI) bus or extended industry-standard architecture (Extended
Industry Standard Architecture, EISA) bus etc..Bus can be divided into address bus, data/address bus, control
Bus etc..For convenient for indicating, the bus in illustrations does not limit only a bus or a type of bus.
The embodiment of the present invention also provides a kind of computer readable storage medium, stores in the computer readable storage medium
There are computer executed instructions, when processor executes the computer executed instructions, realizes in any of the above-described embodiment of the method
Method of Commodity Recommendation.
Above-mentioned computer readable storage medium, above-mentioned readable storage medium storing program for executing can be by any kind of volatibility or non-
Volatile storage devices or their combination realize that, such as static random access memory (SRAM), electrically erasable is only
It reads memory (EEPROM), Erasable Programmable Read Only Memory EPROM (EPROM), programmable read only memory (PROM) is read-only to deposit
Reservoir (ROM), magnetic memory, flash memory, disk or CD.Readable storage medium storing program for executing can be general or specialized computer capacity
Any usable medium enough accessed.
A kind of illustrative readable storage medium storing program for executing is coupled to processor, to enable a processor to from the readable storage medium storing program for executing
Information is read, and information can be written to the readable storage medium storing program for executing.Certainly, readable storage medium storing program for executing is also possible to the composition portion of processor
Point.Processor and readable storage medium storing program for executing can be located at specific integrated circuit (Application Specific Integrated
Circuits, referred to as: ASIC) in.Certainly, processor and readable storage medium storing program for executing can also be used as discrete assembly and be present in equipment
In.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above-mentioned each method embodiment can lead to
The relevant hardware of program instruction is crossed to complete.Program above-mentioned can be stored in a computer readable storage medium.The journey
When being executed, execution includes the steps that above-mentioned each method embodiment to sequence；And storage medium above-mentioned include: ROM, RAM, magnetic disk or
The various media that can store program code such as person's CD.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (12)
1. a kind of Method of Commodity Recommendation characterized by comprising
Historical session data are obtained, the historical session data include: the sequence and each commodity of browsing of the commodity of user's browsing
Corresponding timestamp；
According to the historical session data, at least one grouping of commodities is obtained, each grouping of commodities includes: the first commodity and
Two commodity, second commodity browsing sequence be located at after the browsing sequence of first commodity, and second commodity with
The commodity amount being spaced between first commodity is less than preset value；
According at least one described grouping of commodities, corresponding each second commodity of each first commodity are ranked up,
Obtain the corresponding Recommendations list of first commodity.
2. the method according to claim 1, wherein described at least one grouping of commodities according to, to each
Corresponding each second commodity of first commodity are ranked up, and obtain the corresponding Recommendations list of first commodity,
Include:
For each grouping of commodities, according to the browsing frequency of first commodity, the browsing frequency of the grouping of commodities, institute
The commodity amount being spaced between the second commodity and first commodity and first commodity are stated between second commodity
The browsing stay time of each commodity obtains second commodity relative to first commodity and jumps correlation；
For each first commodity, corresponding each second commodity of first commodity are obtained according to each grouping of commodities,
And correlation is jumped relative to first commodity according to each second commodity, each second commodity are ranked up,
Obtain the corresponding Recommendations list of first commodity.
3. according to the method described in claim 2, it is characterized in that, the browsing frequency according to first commodity, described
The commodity amount and described first being spaced between the browsing frequency of grouping of commodities, second commodity and first commodity
Commodity obtain second commodity relative to first commodity to the browsing stay time of each commodity between second commodity
Jump correlation, comprising:
According to the browsing frequency of the browsing frequency of first commodity and the grouping of commodities, obtain second commodity relative to
First commodity jump probability；
According to the commodity amount being spaced between second commodity and first commodity, it is relatively described to obtain second commodity
First commodity jump coefficient；
According to first commodity to the browsing stay time of each commodity between second commodity, the second commodity phase is obtained
Duration is jumped for first commodity；
According to it is described jump probability, it is described jump coefficient and it is described jump duration, obtain second commodity relative to described the
One commodity jump correlation.
4. according to the method described in claim 3, it is characterized in that, it is described according to first commodity into second commodity
The browsing stay time of each commodity, obtain second commodity relative to first commodity jump duration before, further includes:
According to the timestamp in the historical session data, the browsing for obtaining each commodity in the historical session data is stopped
Stay duration.
5. the method according to claim 1, wherein described according to the historical session data, acquisition at least one
A grouping of commodities, comprising:
According to the timestamp, the historical session data are cut, at least one session is obtained, is wrapped in each session
Include the sequence for the commodity that user browses in the session；
For each session, each commodity and subsequent adjacent N number of commodity combination of two are obtained into the grouping of commodities, N
For the preset value.
6. according to the method described in claim 5, it is characterized in that, described by each commodity and subsequent adjacent N number of commodity two
Two combinations obtain after the grouping of commodities, further includes:
According to the historical session data, the browsing frequency of each commodity is obtained, and obtains the browsing frequency of each grouping of commodities；
According to the browsing frequency of each commodity, the corresponding grouping of commodities of the commodity for being unsatisfactory for the first preset condition is deleted；
According to the browsing frequency of each grouping of commodities, the grouping of commodities for being unsatisfactory for the second preset condition is deleted.
7. according to the method described in claim 5, it is characterized in that, described according to the timestamp, to the historical session number
According to before being cut, further includes:
Remove the noise data in the historical session data.
8. the method according to claim 1, wherein further include:
Goods browse request is obtained, the goods browse request is used to indicate user and requests the first commodity of browsing；
It is requested according to the goods browse, obtains the Recommendations list corresponding with first commodity；
Recommend the commodity in the Recommendations list to the user.
9. the method according to claim 1, wherein further including timestamp in goods browse request；
After the goods browse request for obtaining user's transmission, further includes:
First commodity and the timestamp are recorded in the historical session data.
10. a kind of device for recommending the commodity characterized by comprising
First obtains module, and for obtaining historical session data, the historical session data include: the sequence of the commodity of user's browsing
Column and the corresponding timestamp of each commodity of browsing；
Second obtains module, for obtaining at least one grouping of commodities, each grouping of commodities according to the historical session data
It include: the first commodity and the second commodity, the browsing sequence of second commodity is located at after the browsing sequence of first commodity,
And the commodity amount being spaced between second commodity and first commodity is less than preset value；
Generation module, at least one grouping of commodities according to, to each first commodity corresponding each described second
Commodity are ranked up, and obtain the corresponding Recommendations list of first commodity.
11. a kind of server characterized by comprising at least one processor and memory；
The memory stores computer executed instructions；
At least one described processor executes the computer executed instructions of the memory storage, so that at least one described processing
Device executes method as described in any one of claim 1 to 9.
12. a kind of computer readable storage medium, which is characterized in that be stored with computer in the computer readable storage medium
It executes instruction, when processor executes the computer executed instructions, realizes method as described in any one of claim 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811549358.4A CN110413870B (en) | 2018-12-18 | 2018-12-18 | Commodity recommendation method and device and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811549358.4A CN110413870B (en) | 2018-12-18 | 2018-12-18 | Commodity recommendation method and device and server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110413870A true CN110413870A (en) | 2019-11-05 |
| CN110413870B CN110413870B (en) | 2021-12-31 |
Family
ID=68357479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811549358.4A Active CN110413870B (en) | 2018-12-18 | 2018-12-18 | Commodity recommendation method and device and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110413870B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111311353A (en) * | 2020-01-19 | 2020-06-19 | 河北工业职业技术学院 | User purchase intention evaluation method and system and terminal equipment |
| CN111339829A (en) * | 2020-01-19 | 2020-06-26 | 海通证券股份有限公司 | User identity authentication method, device, computer equipment and storage medium |
| CN111415219A (en) * | 2020-03-02 | 2020-07-14 | 青岛聚好联科技有限公司 | Commodity recommendation method and device based on family and community shopping big data |
| CN112102002A (en) * | 2020-09-16 | 2020-12-18 | 珠海格力电器股份有限公司 | Commodity recommendation method and device based on knowledge graph, computer and storage medium |
| CN112258264A (en) * | 2020-09-30 | 2021-01-22 | 中智关爱通（上海）科技股份有限公司 | Welfare commodity selection method and system |
| CN112783842A (en) * | 2019-11-07 | 2021-05-11 | 北京沃东天骏信息技术有限公司 | Log collection method and device |
| CN113240489A (en) * | 2021-05-18 | 2021-08-10 | 广州卓铸网络科技有限公司 | Article recommendation method and device based on big data statistical analysis |
| CN113362143A (en) * | 2021-07-01 | 2021-09-07 | 海南炳祥投资咨询有限公司 | Internet sales recommendation method and system based on big data |
| CN113763110A (en) * | 2021-02-08 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Article recommendation method and device |
| CN116629984A (en) * | 2023-07-24 | 2023-08-22 | 中信证券股份有限公司 | Product information recommendation method, device, equipment and medium based on embedded model |
| CN117668361A (en) * | 2023-12-06 | 2024-03-08 | 湖南远图网络科技有限公司 | Big data-based pushing method and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011122575A1 (en) * | 2010-03-30 | 2011-10-06 | 楽天株式会社 | Product recommendation device, product recommendation method, program, and recording medium |
| CN102629360A (en) * | 2012-03-13 | 2012-08-08 | 浙江大学 | Effective dynamic commodity recommendation method and commodity recommendation system |
| CN103839167A (en) * | 2012-11-21 | 2014-06-04 | 大连灵动科技发展有限公司 | Commodity candidate set recommendation method |
| CN103839169A (en) * | 2012-11-21 | 2014-06-04 | 大连灵动科技发展有限公司 | Personalized commodity recommendation method based on frequency matrix and text similarity |
| CN103942193A (en) * | 2013-01-17 | 2014-07-23 | 腾讯科技（深圳）有限公司 | Information pushing method and device |
| CN105894332A (en) * | 2016-04-22 | 2016-08-24 | 深圳市永兴元科技有限公司 | Commodity recommendation method, device and system based on user behavior analysis |
| CN107103487A (en) * | 2017-03-02 | 2017-08-29 | 浙江兰德纵横网络技术股份有限公司 | A kind of advertisement sending method based on big data |
-
2018
- 2018-12-18 CN CN201811549358.4A patent/CN110413870B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011122575A1 (en) * | 2010-03-30 | 2011-10-06 | 楽天株式会社 | Product recommendation device, product recommendation method, program, and recording medium |
| CN102629360A (en) * | 2012-03-13 | 2012-08-08 | 浙江大学 | Effective dynamic commodity recommendation method and commodity recommendation system |
| CN103839167A (en) * | 2012-11-21 | 2014-06-04 | 大连灵动科技发展有限公司 | Commodity candidate set recommendation method |
| CN103839169A (en) * | 2012-11-21 | 2014-06-04 | 大连灵动科技发展有限公司 | Personalized commodity recommendation method based on frequency matrix and text similarity |
| CN103942193A (en) * | 2013-01-17 | 2014-07-23 | 腾讯科技（深圳）有限公司 | Information pushing method and device |
| CN105894332A (en) * | 2016-04-22 | 2016-08-24 | 深圳市永兴元科技有限公司 | Commodity recommendation method, device and system based on user behavior analysis |
| CN107103487A (en) * | 2017-03-02 | 2017-08-29 | 浙江兰德纵横网络技术股份有限公司 | A kind of advertisement sending method based on big data |
Non-Patent Citations (1)
| Title |
|---|
| 丁哲: "基于移动用户浏览行为的推荐模型", 《电子科技大学学报》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112783842A (en) * | 2019-11-07 | 2021-05-11 | 北京沃东天骏信息技术有限公司 | Log collection method and device |
| CN112783842B (en) * | 2019-11-07 | 2024-03-01 | 北京沃东天骏信息技术有限公司 | Log collection method and device |
| CN111311353B (en) * | 2020-01-19 | 2023-06-30 | 河北工业职业技术学院 | User purchase intention evaluation method, system and terminal equipment |
| CN111339829A (en) * | 2020-01-19 | 2020-06-26 | 海通证券股份有限公司 | User identity authentication method, device, computer equipment and storage medium |
| CN111311353A (en) * | 2020-01-19 | 2020-06-19 | 河北工业职业技术学院 | User purchase intention evaluation method and system and terminal equipment |
| CN111415219A (en) * | 2020-03-02 | 2020-07-14 | 青岛聚好联科技有限公司 | Commodity recommendation method and device based on family and community shopping big data |
| CN111415219B (en) * | 2020-03-02 | 2023-08-25 | 青岛聚好联科技有限公司 | Commodity recommendation method and device based on big data of home and community shopping |
| CN112102002A (en) * | 2020-09-16 | 2020-12-18 | 珠海格力电器股份有限公司 | Commodity recommendation method and device based on knowledge graph, computer and storage medium |
| CN112258264A (en) * | 2020-09-30 | 2021-01-22 | 中智关爱通（上海）科技股份有限公司 | Welfare commodity selection method and system |
| CN113763110A (en) * | 2021-02-08 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Article recommendation method and device |
| CN113240489B (en) * | 2021-05-18 | 2024-02-09 | 广州卓铸网络科技有限公司 | Article recommendation method and device based on big data statistical analysis |
| CN113240489A (en) * | 2021-05-18 | 2021-08-10 | 广州卓铸网络科技有限公司 | Article recommendation method and device based on big data statistical analysis |
| CN113362143A (en) * | 2021-07-01 | 2021-09-07 | 海南炳祥投资咨询有限公司 | Internet sales recommendation method and system based on big data |
| CN116629984A (en) * | 2023-07-24 | 2023-08-22 | 中信证券股份有限公司 | Product information recommendation method, device, equipment and medium based on embedded model |
| CN116629984B (en) * | 2023-07-24 | 2024-02-06 | 中信证券股份有限公司 | Product information recommendation method, device, equipment and medium based on embedded model |
| CN117668361A (en) * | 2023-12-06 | 2024-03-08 | 湖南远图网络科技有限公司 | Big data-based pushing method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110413870B (en) | 2021-12-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110413870A (en) | Method of Commodity Recommendation, device and server | |
| CN110941740B (en) | Video recommendation method and computer-readable storage medium | |
| TWI529642B (en) | Promotion method and equipment of product information | |
| JP5881153B2 (en) | Recommendation of correlated information | |
| CN106407210B (en) | A kind of methods of exhibiting and device of business object | |
| US9767208B1 (en) | Recommendations for creation of content items | |
| CN105469263A (en) | Commodity recommendation method and device | |
| US9639848B1 (en) | Diffusion prediction based on indicator scoring | |
| JP2005135071A (en) | Method and device for calculating trust values on purchase | |
| CN111461841A (en) | Article recommendation method, device, server and storage medium | |
| CN111144874B (en) | Payment mode recommendation method, device and equipment | |
| McStay | 3.1 Micro-moments, liquidity, intimacy and automation: Developments in programmatic Ad-tech | |
| CN111861605A (en) | Business object recommendation method | |
| EP2850544A1 (en) | A user recommendation method and device | |
| WO2022111095A1 (en) | Product recommendation method and apparatus, computer storage medium, and system | |
| CN110570271A (en) | information recommendation method and device, electronic equipment and readable storage medium | |
| US11392751B1 (en) | Artificial intelligence system for optimizing informational content presentation | |
| CN111680213A (en) | Information recommendation method, data processing method and device | |
| CN107220881B (en) | Method and device for ranking e-commerce popularity based on time and space | |
| US10096045B2 (en) | Tying objective ratings to online items | |
| CN117056619A (en) | Method and device for determining user behavior characteristics | |
| JP6761068B2 (en) | Decision device, decision method and decision program | |
| CN109688178A (en) | Recommended method, device and equipment | |
| CN111708945A (en) | Product recommendation method and device, electronic equipment and computer storage medium | |
| Johari et al. | The Hybrid Recommender System of the Indonesian Online Market Products using IMDb weight rating and TF-IDF |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |