CN104599160A - Commodity recommendation method and commodity recommendation device - Google Patents
Commodity recommendation method and commodity recommendation device Download PDFInfo
- Publication number
- CN104599160A CN104599160A CN201510064107.7A CN201510064107A CN104599160A CN 104599160 A CN104599160 A CN 104599160A CN 201510064107 A CN201510064107 A CN 201510064107A CN 104599160 A CN104599160 A CN 104599160A
- Authority
- CN
- China
- Prior art keywords
- commodity
- user
- interested
- stage
- bundle
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A commodity recommendation method includes the steps of extracting a user's interested commodity from a commodity database which a user is interested in; obtaining a universality purchase cycle of the interested commodity; extracting historical records of similar commodities, purchased by the user, of the interested commodity according to user's commodity purchase records, and computing latest time when the user purchases the similar commodities according to the historical records; setting recommendation time for the interested commodity according to the universality purchase cycle and the latest time; recommending the interested commodity to the user according to the recommendation time. By the method, hit rate of commodity push information can be increased, and network and computer resource utilization rate can be increased effectively. Additionally, the invention further provides a commodity recommendation device corresponding to the commodity recommendation method, and a method and a device for recommending another commodity in an interested commodity bundle.
Description
Technical field
Network Information Mining Technology field of the present invention, particularly a kind of Method of Commodity Recommendation and device.
Background technology
Along with the development of Internet technology and e-commerce technology, shopping online becomes more and more universal.In order to be guided the shopping online behavior of Internet user, E-commerce businessman pushes all kinds of merchandise news to Internet user through various channels, such as, by various internet, applications client and each big hot topic website etc.
Existing merchandise news pushes scheme, and the general shopping record by collecting Internet user, obtains the interested commodity of user, and recommend similar commodity to user.
Although the commodity that existing scheme is recommended have certain hit rate, namely the commodity pushed information quantity successfully guiding user to produce corresponding Shopping Behaviors has accounted for the certain proportion of commodity pushed information total amount, but there is its inherent shortcoming in existing scheme: existing scheme is to user it is recommended that the similar commodity of commodity bought of user, and user probably no longer considers recommended commodity because of the commodity bought.Thus existing scheme can produce a large amount of invalid pushed information, waste network and computer resource.
Summary of the invention
Based on this, be necessary to provide a kind of and improve commercial product recommending hit rate thus effectively utilize Method of Commodity Recommendation and the device of network and computational resource.
A kind of Method of Commodity Recommendation, comprises the following steps:
The commodity interested of user are extracted from user's commodity interested storehouse;
Obtain the ubiquity Buying Cycle of described commodity interested;
Extract according to the commodity purchasing record of described user the historical record that described user buys the similar commodity of described commodity interested, count according to described historical record the up-to-date time that described user buys described similar commodity;
The recommendation time of commodity interested according to described ubiquity Buying Cycle and described up-to-date set of time;
Described commodity interested are recommended to described user according to the described recommendation time.
A kind of Method of Commodity Recommendation, comprises the following steps:
The commodity bundle interested of user is extracted from user's commodity interested storehouse, this commodity bundle interested comprises the multiple commodity belonging to multiple stage, have successively timing between the plurality of stage, the purchase sequential of the commodity of previous stage is prior to the purchase sequential of the commodity of the latter half;
The ubiquity obtaining the two benches commodity that sequential is adjacent in described commodity bundle interested buys interval duration;
Extract according to the commodity purchasing record of described user the historical record that described user buys the similar commodity of the commodity in described commodity bundle interested, according to described historical record determine described user buy up-to-date time of described similar commodity and described similar commodity in described multiple stage belonging to latest stage;
With the described up-to-date time for starting point, buy according to described ubiquity the recommendation time that interval duration arranges each stage commodity after latest stage described in described commodity bundle interested;
According to the described recommendation time to each stage commodity after described user recommends described latest stage.
A kind of device for recommending the commodity, comprising:
Commodity extraction module, for extracting the commodity interested of user from user's commodity interested storehouse;
Buying Cycle acquisition module, for obtaining the ubiquity Buying Cycle of described commodity interested;
Up-to-date time statistical module, for extracting according to the commodity purchasing record of described user the historical record that described user buys the similar commodity of described commodity interested, counts according to described historical record the up-to-date time that described user buys described similar commodity;
First recommends set of time module, for the recommendation time of commodity interested according to described ubiquity Buying Cycle and described up-to-date set of time;
First recommending module, for recommending described commodity interested according to the described recommendation time to described user.
A kind of device for recommending the commodity, comprising:
Commodity bundle extraction module, for extracting the commodity bundle interested of user from user's commodity interested storehouse, this commodity bundle interested comprises the multiple commodity belonging to multiple stage, have successively timing between the plurality of stage, the purchase sequential of the commodity of previous stage is prior to the purchase sequential of the commodity of the latter half;
Buy interval duration acquisition module, buy interval duration for the ubiquity obtaining the two benches commodity that sequential is adjacent in described commodity bundle interested;
Up-to-date time and stage acquisition module, for extracting according to the commodity purchasing record of described user the historical record that described user buys the similar commodity of the commodity in described commodity bundle interested, according to described historical record determine described user buy up-to-date time of described similar commodity and described similar commodity in described multiple stage belonging to latest stage;
Second recommends set of time module, for the described up-to-date time for starting point, buy according to described ubiquity the recommendation time that interval duration arranges each stage commodity after latest stage described in described commodity bundle interested;
Second recommending module, for according to the described recommendation time to each stage commodity after described user recommends described latest stage.
The first Method of Commodity Recommendation of above-mentioned recommendation commodity interested and device, the commodity interested of user are extracted from user's commodity interested storehouse, obtain the ubiquity Buying Cycle of commodity interested, the recommendation time of the up-to-date set of time commodity interested of similar commodity is bought according to this ubiquity Buying Cycle and user, and recommend commodity interested according to this recommendation time to user, the ubiquity Buying Cycle of up-to-date time from user to user and these commodity that said method and device recommend the recommendation time of commodity interested to buy these commodity according to obtains, this recommendation time and user need the time again buying these type of commodity to match, instead of at any time randomly to user's Recommendations, thus improve the hit rate of commodity pushed information, effectively improve the utilization factor of network and computer resource.
The second Method of Commodity Recommendation of commodity and device in above-mentioned recommendation commodity bundle interested, the commodity bundle interested of user is extracted from user's commodity interested storehouse, determine that user have purchased the similar commodity of which stage commodity of commodity bundle interested, and determine the latest stage that bought similar commodity are corresponding and up-to-date time buying, and buy according to the ubiquity of the adjacent two benches commodity of sequential in commodity bundle interested the recommendation time that interval duration and this up-to-date time buying arrange each stage commodity in commodity bundle interested after latest stage, and according to this recommendation time to each stage commodity after user recommends latest stage, this recommendation time and user need the time of each stage commodity bought after latest stage to match, instead of at any time randomly to user's Recommendations, thus improve the hit rate of commodity pushed information, effectively improve the utilization factor of network and computer resource.
Accompanying drawing explanation
Fig. 1 is the part-structure block diagram that can run the equipment of the Method of Commodity Recommendation of the application in an embodiment;
Fig. 2 is the schematic flow sheet of the Method of Commodity Recommendation in an embodiment;
Fig. 3 is the schematic flow sheet of the step of the ubiquity Buying Cycle adding up commodity in an embodiment;
Fig. 4 is the schematic flow sheet of the Method of Commodity Recommendation in an embodiment;
Fig. 5 is the schematic flow sheet of the Method of Commodity Recommendation in an embodiment;
Fig. 6 is the schematic flow sheet of the step of the universality purchase interval duration adding up the adjacent two benches commodity of sequential in commodity bundle in an embodiment;
Fig. 7 is the schematic flow sheet of the Method of Commodity Recommendation in an embodiment;
Fig. 8 is the schematic flow sheet building the step of commodity bundle model in an embodiment;
Fig. 9 is the structural representation of the device for recommending the commodity in an embodiment;
Figure 10 is the structural representation of the device for recommending the commodity in an embodiment;
Figure 11 is the structural representation of the device for recommending the commodity in an embodiment;
Figure 12 is the structural representation of the device for recommending the commodity in an embodiment;
Figure 13 is the structural representation of the device for recommending the commodity in an embodiment;
Figure 14 is the structural representation of the device for recommending the commodity in an embodiment;
Figure 15 is the structural representation of the device for recommending the commodity in an embodiment;
Figure 16 is the structural representation of the device for recommending the commodity in an embodiment.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
Fig. 1 is the part-structure block diagram that can run the equipment of the Method of Commodity Recommendation of the application in an embodiment.As shown in Figure 1, in one embodiment, this server comprises the processor, storage medium, internal memory and the network interface that are connected by system bus.Wherein, network interface is used for network service; Store operating system, database and the software instruction for realizing the Method of Commodity Recommendation described in the application in storage medium, database is for storing user's merchandise news interested, commodity ubiquity Buying Cycle information and commodity purchasing record etc.; Internal memory is used for data cached; Processor is coordinated the work between all parts and is performed above-mentioned software instruction to realize the Method of Commodity Recommendation described in the application.Structure shown in Fig. 1, it is only the block diagram of the part-structure relevant to the application's scheme, do not form the restriction to the equipment that the application's scheme is applied thereon, concrete equipment can comprise than parts more or less shown in figure, or combine some parts, or there is different parts layouts.
As shown in Figure 2, in one embodiment, a kind of Method of Commodity Recommendation, comprises the following steps:
Step S202, extracts the commodity interested of user from user's commodity interested storehouse.
Step S204, obtains the ubiquity Buying Cycle of commodity interested.
In one embodiment, the ubiquity Buying Cycle of commodity composite target that to be the multiple user of tabulate statistics obtain for Buying Cycle of commodity.In one embodiment, ubiquity Buying Cycle of commodity is the mean value or weighted mean value that calculate for Buying Cycle of commodity according to multiple user, wherein, buying weights corresponding to the many users of this commodity amount can higher than buying weights corresponding to the few user of this commodity amount.
In one embodiment, can prestore the ubiquity Buying Cycle data of all kinds of commodity, step S204 can from the ubiquity Buying Cycle of this Buying Cycle extracting data commodity interested prestored.
In one embodiment, above-mentioned Method of Commodity Recommendation also comprises the step of the ubiquity Buying Cycle of statistics commodity, and as shown in Figure 3, in one embodiment, this step comprises the following steps:
Step S302, the average Buying Cycle that multiple user buys all kinds of commodity is counted from purchaser record database, wherein, add up a certain user to buy the step of the average Buying Cycle of all kinds of commodity and comprise: the name keyword extracting the commodity in this user's purchaser record, name keyword and the commodity class name preset are matched, by commodity class corresponding for the commodity class name that indication of goods is with its name keyword matches, the time buying of the same class commodity bought according to this user calculates the average Buying Cycle that this user buys such commodity.
In one embodiment, part or all of purchaser record in purchaser record database can be extracted as sample data, the average Buying Cycle of the purchase counting wherein each user according to the sample data extracted wherein every class I goods.
In one embodiment, the step extracting the name keyword of the commodity in a certain purchaser record comprises: the trade name in purchaser record is carried out participle and obtains multiple words, extract noun wherein and noun phrase, if extract multiple noun or noun phrase, then the plurality of noun or noun phrase are marked, get the highest noun of scoring or the noun phrase name keyword as commodity.In one embodiment, the step that multiple noun or noun phrase are marked is comprised: the plurality of noun or noun phrase can be mated with the picture and information attribute value (such as commercial size, material etc.) etc. of commodity, mark according to noun or the picture of noun phrase and commodity and the matching degree of information attribute value.Such as, " children's vest " name keyword as commodity is extracted from trade name " spring and autumn pure cotton knitting shirt children's garment children vest ".
In one embodiment, the matching relationship between noun can be pre-set, preserve noun relationship match storehouse.Such as, can arrange between approximate word semantically and there is matching relationship.In one embodiment, the coupling noun that the name keyword of commodity is corresponding can be searched in noun relationship match storehouse, if a certain commodity class name is contained among this coupling noun, then judge that this commodity class name and this name keyword match.In one embodiment, if name keyword is identical with commodity class name, then also can judge that this name keyword and this commodity class name match.
In one embodiment, time buying of the same class commodity bought according to this user calculates this user and buys the step of the average Buying Cycle of such commodity and comprise: sorted according to the time buying by the purchaser record of the purchase same class commodity of this user, obtain the predetermined number bar purchaser record that sorting position is adjacent, and obtain the time buying span interval duration of first time buying and last time buying (namely in this predetermined number bar purchaser record) of this predetermined number bar purchaser record, the ratio calculating this time buying span and predetermined number buys the average Buying Cycle of such commodity as user.In another embodiment, time buying of the same class commodity bought according to this user calculates this user and buys the step of the average Buying Cycle of such commodity and comprise: sorted according to the time buying by the purchaser record of the purchase same class commodity of this user, intercept time buying span in the purchaser record sequence after sequence and exceed a cross-talk sequence of threshold value, add up purchaser record quantity and time buying span that this sub-series of packets contains, the ratio calculating this time buying span and this purchaser record quantity buys the average Buying Cycle of such commodity as this user.
Step S304, calculates the ubiquity Buying Cycle of such commodity according to the average Buying Cycle of each user of same class commodity.
In one embodiment, the mean value of the average Buying Cycle of each user of same class commodity or the weighted mean value ubiquity Buying Cycle as such commodity can be calculated.
Step S206, extracts according to the commodity purchasing record of user the historical record that user buys the similar commodity of commodity interested, counts according to this historical record the up-to-date time that this user buys these similar commodity.
In one embodiment, the step buying the historical record of the similar commodity of commodity interested according to the commodity purchasing record extraction user of user comprises: the commodity purchasing record extracting user from purchaser record database, extract the name keyword of the commodity in each commodity purchasing record, if the class name of the name keyword of commodity and the default commodity class belonging to commodity interested matches, then the commodity purchasing record of correspondence is extracted as the historical record that user buys the similar commodity of commodity interested.
Step S208, buys the recommendation time of the up-to-date set of time commodity interested of the similar commodity of commodity interested according to ubiquity Buying Cycle of commodity interested and user.
In one embodiment, up-to-date time user can being bought the similar commodity of commodity interested is starting point, using the ubiquity Buying Cycle of commodity interested and the product of a preset weights as time increment, the recommendation time arranging commodity interested is time point corresponding after this starting point increases this time increment.Such as, preset weights is 1 or 0.8 etc.
Step S210, recommends commodity interested according to the recommendation time to user.
The relevant information of commodity interested can be pushed, such as commodity purchasing web page interlinkage, trade name, commodity price and commodity picture etc. to user.
As shown in Figure 4, in one embodiment, above-mentioned Method of Commodity Recommendation is further comprising the steps of:
Step S402, obtains the individual Buying Cycle of user for commodity interested.
In one embodiment, the individual Buying Cycle that user buys the similar commodity of commodity interested can be calculated, using this individual's Buying Cycle as the individual Buying Cycle of user for commodity interested.
Step S404, if user is shorter than the ubiquity Buying Cycle of commodity interested for individual Buying Cycle of commodity interested, then according to the recommendation time of this individual's Buying Cycle commodity interested in advance.
In one embodiment, the duration of the recommendation time advance of commodity interested can be that gap between this ubiquity Buying Cycle and this individual's Buying Cycle is long-pending with the weights preset.
In the present embodiment, if user is shorter than the ubiquity Buying Cycle of commodity interested for individual Buying Cycle of commodity interested, the recommendation time of commodity interested is then shifted to an earlier date according to this individual's Buying Cycle, recommendation time and user can be made to need the time buying these commodity interested more identical, thus improve the hit rate of commodity pushed information.
As shown in Figure 5, a kind of Method of Commodity Recommendation, comprises the following steps:
Step S502, the commodity bundle interested of user is extracted from user's commodity interested storehouse, this commodity bundle interested comprises the multiple commodity belonging to multiple stage, has successively timing between the plurality of stage, and the purchase sequential of the commodity of previous stage is prior to the purchase sequential of the commodity of the latter half.
In one embodiment, the commodity that multiple stages of comprising of the corresponding complete event process of commodity bundle are corresponding.Such as, comprise the commodity bundle of house fitting-up commodity, in this commodity bundle, comprise the commodity belonging to multiple stage needs of comprising in household fixtures process and use.Again such as, comprise the commodity bundle of fitness program commodity, this body-building calculates and comprises multiple stage, the commodity that the multiple stage needs comprising fitness program in this commodity bundle are used.
Step S504, the ubiquity obtaining the adjacent two benches commodity of sequential in commodity bundle interested buys interval duration.
Such as, if commodity bundle comprises the commodity of three phases, this three phases according to sequential from elder generation to rear sequence is: first stage, subordinate phase, phase III, then first stage and subordinate phase are the two benches that sequential is adjacent, and subordinate phase and phase III are the two benches that sequential is adjacent.
In one embodiment, the ubiquity of the two benches commodity that sequential is adjacent buys interval duration is the composite target that purchase interval duration that the multiple user of tabulate statistics buys these two benches commodity obtains.In one embodiment, it is the mean value that obtains of the purchase interval duration calculation buying these two benches commodity according to multiple user or weighted mean value that the ubiquity of two benches commodity buys interval duration, wherein, the weights that weights corresponding to the many users of this two benches commodity amount can be corresponding higher than the user that quantity purchase is few are bought.
In one embodiment, the ubiquity that can prestore the adjacent two benches commodity of sequential of extensive stock bag buys interval duration, and step S504 can buy the ubiquity purchase interval duration of the adjacent two benches commodity of sequential of interval duration extracting data commodity bundle interested from the ubiquity of the sequential adjacent two benches commodity of the extensive stock bag prestored.
In one embodiment, above-mentioned Method of Commodity Recommendation also comprises the step of the universality purchase interval duration of the adjacent two benches commodity of sequential in statistics commodity bundle, and as shown in Figure 6, this step comprises the following steps:
Step S602, extracts the purchaser record with users classification from purchaser record database.
Step S604, extract the name keyword of the commodity in the purchaser record of each user, mark the stage of the commodity bundle belonging to each commodity, if the commodity in certain stage match in the name keyword of certain commodity and commodity bundle, then mark this stage that these commodity belong to this commodity bundle.
In one embodiment, the matching relationship between noun can be pre-set, preserve noun relationship match storehouse.Such as, can arrange between approximate word semantically and there is matching relationship.In one embodiment, if the class name of the default commodity class belonging to a certain commodity in certain stage preserves matching relationship in this noun relationship match storehouse in the name keyword of certain commodity and commodity bundle, even in the name keyword of certain commodity and commodity bundle, the class name of the default commodity class belonging to a certain commodity in certain stage matches, then can judge that the commodity in the name keyword of these commodity and this stage of commodity bundle match.
Step S606, adds up the average purchase interval duration that each user buys the sequential adjacent two benches commodity of commodity bundle.
In one embodiment, mean value that a user repeatedly buys the purchase interval duration of the sequential adjacent two benches commodity of commodity bundle buys this sequential adjacent two benches commodity average purchase interval duration as this user can be calculated.Such as, the record of the commodity of a certain for the purchase of a certain user commodity bundle is obtained purchaser record sequence according to the time buying by first extremely rear sequence: (t1, first stage), (t2, subordinate phase), (t3, phase III), (t4, first stage), (t5, subordinate phase), (t6, phase III), wherein each purchaser record is referred to as the data pair in stage belonging to time buying and commodity, then the average purchase duration of this user first stage and subordinate phase commodity of buying this commodity bundle can be calculated as: [(t2-t1)+(t5-t4)]/2.
Step S608, the ubiquity according to the adjacent two benches commodity of average purchase interval this sequential of duration calculation of each user of sequential adjacent two benches commodity buys interval duration.
In one embodiment, the ubiquity purchase interval duration that can calculate the sequential adjacent two benches commodity of a certain commodity bundle is mean value or the weighted mean value that each user buys the average purchase interval duration of this sequential adjacent two benches commodity of this commodity bundle.
Step S506, extract according to the commodity purchasing record of user the historical record that user buys the similar commodity of the commodity in commodity bundle interested, according to this historical record determine user buy up-to-date time of these similar commodity and this similar commodity in multiple stages of commodity bundle interested belonging to latest stage.
In one embodiment, the step buying the historical record of the similar commodity of the commodity in commodity bundle interested according to the commodity purchasing record extraction user of user comprises: the commodity purchasing record extracting user from purchaser record database, extract the name keyword of the commodity in each commodity purchasing record, if the class name of the default commodity class belonging to arbitrary commodity in a commodity purchasing record in the name keyword of commodity and commodity bundle interested matches, then this commodity purchasing record is extracted as the historical record that user buys the similar commodity of commodity interested.And further, the stage of stage corresponding to this commodity purchasing record residing for the commodity that match in commodity bundle interested can be marked.
And further, buying stage corresponding to each bar historical record of the similar commodity of commodity interested by comparing user, the latest stage of these similar commodity belonging in multiple stages of commodity bundle interested can be obtained.
Step S508, with the above-mentioned up-to-date time for starting point, buys according to the ubiquity of the adjacent two benches commodity of sequential in commodity bundle interested the recommendation time that interval duration arranges each stage commodity in commodity bundle interested after above-mentioned latest stage.
In one embodiment, can the above-mentioned up-to-date time be starting point, using above-mentioned latest stage and its next stage commodity ubiquity buy the product of interval duration and a preset weights as time increment, arrange latest stage next stage commodity the recommendation time; And set gradually this next stage with the recommendation time of after-stage commodity, the wherein recommendation time of these each stage commodity later, with the recommendation time of the commodity on last stage in this stage for starting point, the ubiquity increasing this stage and commodity on last stage buys a product of interval duration and a preset weights.
Step S510, according to recommendation time in each stage to each stage commodity after user recommends latest stage.
As shown in Figure 7, in one embodiment, above-mentioned Method of Commodity Recommendation, further comprising the steps of:
Step S702, acquisition user buys interval duration for the individual of the adjacent two benches commodity of sequential in commodity bundle interested.
In one embodiment, the purchaser record of user can be extracted from purchaser record database; Extract the name keyword of the commodity in the purchaser record of this user, mark the stage of each commodity belonging in multiple stages of commodity bundle interested, if the commodity in certain stage match in the name keyword of certain commodity and commodity bundle interested, then mark this stage that these commodity belong to commodity bundle interested; Add up the average purchase interval duration that this user buys the sequential adjacent two benches commodity of commodity bundle interested, obtain above-mentioned individual and buy interval duration.
Step S704, if the above-mentioned individual of sequential adjacent two benches commodity buys the above-mentioned ubiquity purchase interval duration that interval duration is shorter than this sequential adjacent two benches commodity, then according to the recommendation time of each stage commodity after latest stage in above-mentioned individual's purchase interval duration in advance commodity bundle interested.
In one embodiment, the duration of the recommendation time advance of each stage commodity can be bought for the ubiquity of this stage and its preceding paragraph commodity the gap that interval duration and corresponding individual buy between the duration of interval and amasss with the weights preset are.
In the present embodiment, if the above-mentioned individual of sequential adjacent two benches commodity buys the above-mentioned ubiquity purchase interval duration that interval duration is shorter than this sequential adjacent two benches commodity, then buy according to above-mentioned individual the recommendation time that interval duration shifts to an earlier date each stage commodity in commodity bundle interested after latest stage, recommendation time and user can be made to need the time buying each stage commodity in this commodity bundle interested more identical, thus improve the hit rate of commodity pushed information.
In one embodiment, above-mentioned Method of Commodity Recommendation is further comprising the steps of: extract the commodity that in the above-mentioned latest stage commodity of commodity bundle interested, user not yet buys; The commodity recommending user not yet to buy to user in real time.
The step of the commodity in real time recommending user not yet to buy to user comprises: the commodity recommending user not yet to buy as the recommendation time to user current time.
In commodity bundle, the commodity amount in a stage can be multiple, in the present embodiment, the commodity of the latest stage of the commodity bundle interested that user buys may be all the commodity needed user's present stage, the recommendation time that current time can be suitable for, thus commercial product recommending user in this latest stage not yet bought in real time is to user, can improve the hit rate of commodity pushed information.
In one embodiment, before step S502, above-mentioned Method of Commodity Recommendation also comprises the step building commodity bundle model, and as shown in Figure 8, this step comprises the following steps:
Step S802, extracts the purchaser record with users classification from purchaser record database.
Step S804, extract the commodity that each user buys the same period, multiple commodity that the number of times simultaneously occurred in the commodity bought in each user same period exceedes threshold value are divided into same stage commodity, thus summarize multiple stage and commodity corresponding to the plurality of stage.
Step S806, counts according to the purchaser record of each user the purchase interval duration that each user buys each stage commodity.
Step S808, if each user buys the purchase sequence consensus that identical and this each user of the purchase interval duration convergence of certain two benches commodity buys these two benches commodity, then these two benches commodity are divided into the two benches commodity of same commodity bundle, thus obtain by the commodity bundle of multiple stage commodity composition, storing commodity bag data.
In one embodiment, can will comprise the commodity bundle interested being set to user with the commodity bundle of the similar commodity of the commodity interested of user, even comprise commodity in commodity bundle, these commodity are the similar commodity of the commodity interested of user, then this commodity bundle can be set to the commodity bundle interested of this user.Further, the commodity bundle interested of user can be stored in user commodity storehouse interested.
The first Method of Commodity Recommendation of above-mentioned recommendation commodity interested can in conjunction with implementing with the second Method of Commodity Recommendation of commodity in recommendation commodity bundle interested.A kind of Method of Commodity Recommendation, comprises the step in the step in the first Method of Commodity Recommendation in above-mentioned any embodiment and the second Method of Commodity Recommendation in above-mentioned any embodiment, then this Method of Commodity Recommendation also belongs to the scope of the application's protection.
As shown in Figure 9, in one embodiment, a kind of device for recommending the commodity, comprises commodity extraction module 902, Buying Cycle acquisition module 904, up-to-date time statistical module 906, first recommendation set of time module 908 and the first recommending module 910, wherein:
Commodity extraction module 902 for extracting the commodity interested of user from user's commodity interested storehouse.
Buying Cycle acquisition module 904 is for obtaining the ubiquity Buying Cycle of commodity interested.
In one embodiment, the ubiquity Buying Cycle of commodity composite target that to be the multiple user of tabulate statistics obtain for Buying Cycle of commodity.In one embodiment, ubiquity Buying Cycle of commodity is the mean value or weighted mean value that calculate for Buying Cycle of commodity according to multiple user, wherein, buying weights corresponding to the many users of this commodity amount can higher than buying weights corresponding to the few user of this commodity amount.
As shown in Figure 10, in one embodiment, the above-mentioned device for recommending the commodity also comprises general Buying Cycle statistical module 1002 and general Buying Cycle memory module 1004, wherein: general Buying Cycle statistical module 1002 is for adding up the ubiquity Buying Cycle of commodity; Ubiquity Buying Cycle memory module 1004 is for storing the ubiquity Buying Cycle data of all kinds of commodity, and Buying Cycle acquisition module 904 can from the ubiquity Buying Cycle of this Buying Cycle extracting data commodity interested prestored.
In one embodiment, general Buying Cycle statistical module 1002 buys the average Buying Cycle of all kinds of commodity for counting multiple user from purchaser record database, wherein, add up a certain user to buy the process of the average Buying Cycle of all kinds of commodity and comprise: the name keyword extracting the commodity in this user's purchaser record, name keyword and the commodity class name preset are matched, by commodity class corresponding for the commodity class name that indication of goods is with its name keyword matches, the time buying of the same class commodity bought according to this user calculates the average Buying Cycle that this user buys such commodity.
In one embodiment, general Buying Cycle statistical module 1002 can extract part or all of purchaser record in purchaser record database as sample data, the average Buying Cycle of the purchase counting wherein each user according to the sample data extracted wherein every class I goods.
In one embodiment, the process that general Buying Cycle statistical module 1002 extracts the name keyword of the commodity in a certain purchaser record comprises: the trade name in purchaser record is carried out participle and obtains multiple words, extract noun wherein and noun phrase, if extract multiple noun or noun phrase, then the plurality of noun or noun phrase are marked, get the highest noun of scoring or the noun phrase name keyword as commodity.In one embodiment, general Buying Cycle statistical module 1002 comprises the process that multiple noun or noun phrase are marked: the plurality of noun or noun phrase can be mated with the picture and information attribute value (such as commercial size, material etc.) etc. of commodity, marks according to noun or the picture of noun phrase and commodity and the matching degree of information attribute value.
In one embodiment, the above-mentioned device for recommending the commodity also comprises noun matching relationship memory module (not shown), for arranging the matching relationship between noun, preserves noun relationship match storehouse.Such as, can arrange between approximate word semantically and there is matching relationship.In one embodiment, general Buying Cycle statistical module 1002 can search coupling noun corresponding to the name keyword of commodity in noun relationship match storehouse, if a certain commodity class name is contained among this coupling noun, then judge that this commodity class name and this name keyword match.In one embodiment, if name keyword is identical with commodity class name, then general Buying Cycle statistical module 1002 also can judge that this name keyword and this commodity class name match.
In one embodiment, time buying of the same class commodity that general Buying Cycle statistical module 1002 is bought according to this user calculates this user and buys the process of the average Buying Cycle of such commodity and comprise: sorted according to the time buying by the purchaser record of the purchase same class commodity of this user, obtain the predetermined number bar purchaser record that sorting position is adjacent, and obtain the time buying span interval duration of first time buying and last time buying (namely in this predetermined number bar purchaser record) of this predetermined number bar purchaser record, the ratio calculating this time buying span and predetermined number buys the average Buying Cycle of such commodity as user.In another embodiment, time buying of the same class commodity that general Buying Cycle statistical module 1002 is bought according to this user calculates this user and buys the process of the average Buying Cycle of such commodity and comprise: sorted according to the time buying by the purchaser record of the purchase same class commodity of this user, intercept time buying span in the purchaser record sequence after sequence and exceed a cross-talk sequence of threshold value, add up purchaser record quantity and time buying span that this sub-series of packets contains, the ratio calculating this time buying span and this purchaser record quantity buys the average Buying Cycle of such commodity as this user.
Further, general Buying Cycle statistical module 1002 also calculates the ubiquity Buying Cycle of such commodity for average Buying Cycle of each user according to same class commodity.
In one embodiment, general Buying Cycle statistical module 1002 can calculate the mean value of the average Buying Cycle of each user of same class commodity or the weighted mean value ubiquity Buying Cycle as such commodity.
Up-to-date time statistical module 906, for extracting according to the commodity purchasing record of user the historical record that user buys the similar commodity of commodity interested, counts according to this historical record the up-to-date time that this user buys these similar commodity.
In one embodiment, the process that up-to-date time statistical module 906 buys the historical record of the similar commodity of commodity interested according to the commodity purchasing record extraction user of user comprises: the commodity purchasing record extracting user from purchaser record database, extract the name keyword of the commodity in each commodity purchasing record, if the class name of the name keyword of commodity and the default commodity class belonging to commodity interested matches, then the commodity purchasing record of correspondence is extracted as the historical record that user buys the similar commodity of commodity interested.
First recommends set of time module 908 for buying the recommendation time of the up-to-date set of time commodity interested of the similar commodity of commodity interested according to ubiquity Buying Cycle of commodity interested and user.
In one embodiment, the first up-to-date time of recommending set of time module 908 user can be bought the similar commodity of commodity interested was starting point, using the ubiquity Buying Cycle of commodity interested and the product of a preset weights as time increment, the recommendation time arranging commodity interested is time point corresponding after this starting point increases this time increment.
First recommending module 910 is for recommending commodity interested according to the recommendation time to user.
In one embodiment, the first recommending module 910 can push the relevant information of commodity interested to user, such as commodity purchasing web page interlinkage, trade name, commodity price and commodity picture etc.
As shown in figure 11, in one embodiment, the above-mentioned device for recommending the commodity also comprises individual's Buying Cycle acquisition module 1102 and first and recommends time regulating module 1104, wherein:
Individual's Buying Cycle acquisition module 1102 is for obtaining the individual Buying Cycle of user for commodity interested.
In one embodiment, individual's Buying Cycle acquisition module 1102 can calculate the individual Buying Cycle that user buys the similar commodity of commodity interested, using this individual's Buying Cycle as the individual Buying Cycle of user for commodity interested.
If first recommends time regulating module 1104 to be shorter than the ubiquity Buying Cycle of commodity interested for individual Buying Cycle of commodity interested for user, then shift to an earlier date the recommendation time of commodity interested according to this individual Buying Cycle.
In one embodiment, the duration of the recommendation time advance of commodity interested can be that gap between this ubiquity Buying Cycle and this individual's Buying Cycle is long-pending with the weights preset.
In the present embodiment, if user is shorter than the ubiquity Buying Cycle of commodity interested for individual Buying Cycle of commodity interested, the recommendation time of commodity interested is then shifted to an earlier date according to this individual's Buying Cycle, recommendation time and user can be made to need the time buying these commodity interested more identical, thus improve the hit rate of commodity pushed information.
As shown in figure 12, a kind of device for recommending the commodity, comprises commodity bundle extraction module 1202, purchase interval duration acquisition module 1204, up-to-date time and stage acquisition module 1206, second and recommends set of time module 1208 and the second recommending module 1210, wherein:
Commodity bundle extraction module 1202 for extracting the commodity bundle interested of user from user's commodity interested storehouse, this commodity bundle interested comprises the multiple commodity belonging to multiple stage, have successively timing between the plurality of stage, the purchase sequential of the commodity of previous stage is prior to the purchase sequential of the commodity of the latter half.
In one embodiment, the commodity that multiple stages of comprising of the corresponding complete event process of commodity bundle are corresponding.Such as, comprise the commodity bundle of house fitting-up commodity, in this commodity bundle, comprise the commodity belonging to multiple stage needs of comprising in household fixtures process and use.Again such as, comprise the commodity bundle of fitness program commodity, this body-building calculates and comprises multiple stage, the commodity that the multiple stage needs comprising fitness program in this commodity bundle are used.
Buy interval duration acquisition module 1204 and buy interval duration for the ubiquity obtaining the adjacent two benches commodity of sequential in commodity bundle interested.
Such as, if commodity bundle comprises the commodity of three phases, this three phases according to sequential from elder generation to rear sequence is: first stage, subordinate phase, phase III, then first stage and subordinate phase are the two benches that sequential is adjacent, and subordinate phase and phase III are the two benches that sequential is adjacent.
In one embodiment, the ubiquity of the two benches commodity that sequential is adjacent buys interval duration is the composite target that purchase interval duration that the multiple user of tabulate statistics buys these two benches commodity obtains.In one embodiment, it is the mean value that obtains of the purchase interval duration calculation buying these two benches commodity according to multiple user or weighted mean value that the ubiquity of two benches commodity buys interval duration, wherein, the weights that weights corresponding to the many users of this two benches commodity amount can be corresponding higher than the user that quantity purchase is few are bought.
As shown in figure 13, in one embodiment, the above-mentioned device for recommending the commodity also comprises common purchase interval duration statistical module 1302 and common purchase interval duration memory module 1304, wherein: universality is bought interval duration module 1302 and bought interval duration for the universality of adding up the adjacent two benches commodity of sequential in commodity bundle; Universality is bought interval duration memory module 1304 and is bought interval duration for the ubiquity of the adjacent two benches commodity of sequential storing extensive stock bag, buys interval duration acquisition module 1204 can buy the adjacent two benches commodity of sequential of interval duration extracting data commodity bundle interested ubiquity purchase interval duration from the ubiquity of the sequential adjacent two benches commodity of the extensive stock bag prestored.
In one embodiment, universality buys interval duration module 1302 for extracting the purchaser record with users classification from purchaser record database.
Further, universality buys interval duration module 1302 also for extracting the name keyword of the commodity in the purchaser record of each user, mark the stage of the commodity bundle belonging to each commodity, if the commodity in certain stage match in the name keyword of certain commodity and commodity bundle, then mark this stage that these commodity belong to this commodity bundle.
In one embodiment, the above-mentioned device for recommending the commodity also comprises noun matching relationship memory module (not shown), for arranging the matching relationship between noun, preserves noun relationship match storehouse.Such as, can arrange between approximate word semantically and there is matching relationship.In one embodiment, if the class name of the default commodity class belonging to a certain commodity in certain stage preserves matching relationship in this noun relationship match storehouse in the name keyword of certain commodity and commodity bundle, even in the name keyword of certain commodity and commodity bundle, the class name of the default commodity class belonging to a certain commodity in certain stage matches, then universality is bought interval duration module 1302 and can be judged that the commodity in the name keyword of these commodity and this stage of commodity bundle match.
Further, universality buys interval duration module 1302 also buys the sequential adjacent two benches commodity of commodity bundle average purchase interval duration for adding up each user.
In one embodiment, universality is bought interval duration module 1302 and can be calculated mean value that a user repeatedly buys the purchase interval duration of the sequential adjacent two benches commodity of commodity bundle buys this sequential adjacent two benches commodity average purchase interval duration as this user.
Further, universality is bought interval duration module 1302 and is also bought interval duration for the ubiquity of the adjacent two benches commodity of average purchase interval this sequential of duration calculation of each user according to sequential adjacent two benches commodity.
In one embodiment, it is mean value or the weighted mean value that each user buys the average purchase interval duration of this sequential adjacent two benches commodity of this commodity bundle that the ubiquity that universality purchase interval duration module 1302 can calculate the sequential adjacent two benches commodity of a certain commodity bundle buys interval duration.
Up-to-date time and stage acquisition module 1206 for extracting according to the commodity purchasing record of user the historical record that user buys the similar commodity of the commodity in commodity bundle interested, according to this historical record determine user buy up-to-date time of these similar commodity and this similar commodity in multiple stages of commodity bundle interested belonging to latest stage.
In one embodiment, the process that up-to-date time and stage acquisition module 1206 buy the historical record of the similar commodity of the commodity in commodity bundle interested according to the commodity purchasing record extraction user of user comprises: the commodity purchasing record extracting user from purchaser record database, extract the name keyword of the commodity in each commodity purchasing record, if the class name of the default commodity class belonging to arbitrary commodity in a commodity purchasing record in the name keyword of commodity and commodity bundle interested matches, then this commodity purchasing record is extracted as the historical record that user buys the similar commodity of commodity interested.And further, up-to-date time and stage acquisition module 1206 can mark the stage of stage corresponding to this commodity purchasing record residing for the commodity that match in commodity bundle interested.
And further, up-to-date time and stage acquisition module 1206 buy stage corresponding to each bar historical record of the similar commodity of commodity interested by comparing user, can obtain the latest stage of these similar commodity belonging in multiple stages of commodity bundle interested.
Second recommend set of time module 1208 for the above-mentioned up-to-date time for starting point, the ubiquity according to the adjacent two benches commodity of sequential in commodity bundle interested buys the recommendation time that interval duration arranges each stage commodity in commodity bundle interested after above-mentioned latest stage.
In one embodiment, second recommends set of time module 1208 the above-mentioned up-to-date time to be starting point, using above-mentioned latest stage and its next stage commodity ubiquity buy the product of interval duration and a preset weights as time increment, arrange latest stage next stage commodity the recommendation time; And set gradually this next stage with the recommendation time of after-stage commodity, the wherein recommendation time of these each stage commodity later, with the recommendation time of the commodity on last stage in this stage for starting point, the ubiquity increasing this stage and commodity on last stage buys a product of interval duration and a preset weights.
Second recommending module 1210 for the recommendation time according to each stage to each stage commodity after user recommends latest stage.
As shown in figure 14, in one embodiment, the above-mentioned device for recommending the commodity also comprises individual and buys interval duration 1402 and the second recommendation time regulating module 1404, wherein:
Individual buys interval duration 1402 and buys interval duration for obtaining user for the individual of the adjacent two benches commodity of sequential in commodity bundle interested.
In one embodiment, individual buys the purchaser record that interval duration 1402 can extract user from purchaser record database; Extract the name keyword of the commodity in the purchaser record of this user, mark the stage of each commodity belonging in multiple stages of commodity bundle interested, if the commodity in certain stage match in the name keyword of certain commodity and commodity bundle interested, then mark this stage that these commodity belong to commodity bundle interested; Add up the average purchase interval duration that this user buys the sequential adjacent two benches commodity of commodity bundle interested, obtain above-mentioned individual and buy interval duration.
If second recommends time regulating module 1404 to buy for the above-mentioned individual of sequential adjacent two benches commodity the above-mentioned ubiquity purchase interval duration that interval duration is shorter than this sequential adjacent two benches commodity, then buy according to above-mentioned individual the recommendation time that interval duration shifts to an earlier date each stage commodity in commodity bundle interested after latest stage.
In one embodiment, the duration of the recommendation time advance of each stage commodity can be bought for the ubiquity of this stage and its preceding paragraph commodity the gap that interval duration and corresponding individual buy between the duration of interval and amasss with the weights preset are.
In the present embodiment, if the above-mentioned individual of sequential adjacent two benches commodity buys the above-mentioned ubiquity purchase interval duration that interval duration is shorter than this sequential adjacent two benches commodity, then buy according to above-mentioned individual the recommendation time that interval duration shifts to an earlier date each stage commodity in commodity bundle interested after latest stage, recommendation time and user can be made to need the time buying each stage commodity in this commodity bundle interested more identical, thus improve the hit rate of commodity pushed information.
As shown in figure 15, in one embodiment, the above-mentioned device for recommending the commodity also comprises does not buy commodity extraction module 1502 and the 3rd recommending module 1504, wherein:
Do not buy commodity extraction module 1502 for extract commodity bundle interested above-mentioned latest stage commodity in the commodity not yet bought of user; The commodity that 3rd recommending module 1504 recommends user not yet to buy to user in real time.
The process of the commodity that the 3rd recommending module 1504 in real time recommends user not yet to buy to user comprises: the commodity recommending user not yet to buy as the recommendation time to user current time.
In commodity bundle, the commodity amount in a stage can be multiple, in the present embodiment, the commodity of the latest stage of the commodity bundle interested that user buys may be all the commodity needed user's present stage, the recommendation time that current time can be suitable for, thus commercial product recommending user in this latest stage not yet bought in real time is to user, can improve the hit rate of commodity pushed information.
As shown in figure 16, in one embodiment, the above-mentioned device for recommending the commodity also comprises commodity bundle model construction module 1602 and commodity bundle interested arranges module 1604, wherein:
Commodity bundle model construction module 1602 for extracting the purchaser record with users classification from purchaser record database.
Further, commodity bundle model construction module 1602 is also for extracting the commodity that each user buys the same period, multiple commodity that the number of times simultaneously occurred in the commodity bought in each user same period exceedes threshold value are divided into same stage commodity, thus summarize multiple stage and commodity corresponding to the plurality of stage.
Further, commodity bundle model construction module 1602 also buys the purchase interval duration of each stage commodity for counting each user according to the purchaser record of each user.
Further, if commodity bundle model construction module 1602 also buys for each user the purchase sequence consensus that identical and this each user of the purchase interval duration convergence of certain two benches commodity buys these two benches commodity, then these two benches commodity are divided into the two benches commodity of same commodity bundle, thus obtain by the commodity bundle of multiple stage commodity composition, storing commodity bag data.
Commodity bundle interested arranges module 1604 for comprising the commodity bundle interested being set to user with the commodity bundle of the similar commodity of the commodity interested of user, even comprise commodity in commodity bundle, these commodity are the similar commodity of the commodity interested of user, then this commodity bundle can be set to the commodity bundle interested of this user.Further, commodity bundle interested arranges module 1604 and the commodity bundle interested of user can be stored in user commodity storehouse interested.
The first device for recommending the commodity of above-mentioned recommendation commodity interested can be merged into a device with the second device for recommending the commodity of commodity in recommendation commodity bundle interested.A kind of device for recommending the commodity, comprises the module in the module in the first device for recommending the commodity in above-mentioned any embodiment and the second device for recommending the commodity in above-mentioned any embodiment, then this device for recommending the commodity also belongs to the scope of the application's protection.
The first Method of Commodity Recommendation of above-mentioned recommendation commodity interested and device, the commodity interested of user are extracted from user's commodity interested storehouse, obtain the ubiquity Buying Cycle of commodity interested, the recommendation time of the up-to-date set of time commodity interested of similar commodity is bought according to this ubiquity Buying Cycle and user, and recommend commodity interested according to this recommendation time to user, the ubiquity Buying Cycle of up-to-date time from user to user and these commodity that said method and device recommend the recommendation time of commodity interested to buy these commodity according to obtains, this recommendation time and user need the time again buying these type of commodity to match, instead of at any time randomly to user's Recommendations, thus improve the hit rate of commodity pushed information, effectively improve the utilization factor of network and computer resource.
The second Method of Commodity Recommendation of commodity and device in above-mentioned recommendation commodity bundle interested, the commodity bundle interested of user is extracted from user's commodity interested storehouse, determine that user have purchased the similar commodity of which stage commodity of commodity bundle interested, and determine the latest stage that bought similar commodity are corresponding and up-to-date time buying, and buy according to the ubiquity of the adjacent two benches commodity of sequential in commodity bundle interested the recommendation time that interval duration and this up-to-date time buying arrange each stage commodity in commodity bundle interested after latest stage, and according to this recommendation time to each stage commodity after user recommends latest stage, this recommendation time and user need the time of each stage commodity bought after latest stage to match, instead of at any time randomly to user's Recommendations, thus improve the hit rate of commodity pushed information, effectively improve the utilization factor of network and computer resource.
The above embodiment only have expressed several embodiment of the present invention, and it describes comparatively concrete and detailed, but therefore can not be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.
Claims (10)
1. a Method of Commodity Recommendation, comprises the following steps:
The commodity interested of user are extracted from user's commodity interested storehouse;
Obtain the ubiquity Buying Cycle of described commodity interested;
Extract according to the commodity purchasing record of described user the historical record that described user buys the similar commodity of described commodity interested, count according to described historical record the up-to-date time that described user buys described similar commodity;
The recommendation time of commodity interested according to described ubiquity Buying Cycle and described up-to-date set of time;
Described commodity interested are recommended to described user according to the described recommendation time.
2. Method of Commodity Recommendation according to claim 1, is characterized in that, also comprise the step of the ubiquity Buying Cycle of statistics commodity, this step comprises the following steps:
The average Buying Cycle that multiple user buys all kinds of commodity is counted from purchaser record database, wherein, add up a certain user to buy the step of the average Buying Cycle of all kinds of commodity and comprise: the name keyword extracting the commodity in this user's purchaser record, the commodity class name that name keyword is preset is matched, by commodity class corresponding for the commodity class name that indication of goods is with its name keyword matches, the time buying of the same class commodity bought according to this user calculates the average Buying Cycle that this user buys such commodity;
The ubiquity Buying Cycle of such commodity is calculated according to the average Buying Cycle of each user of same class commodity.
3. Method of Commodity Recommendation according to claim 1, is characterized in that, also comprises step:
Obtain the individual Buying Cycle of described user for described commodity interested;
If described individual's Buying Cycle is shorter than the described ubiquity Buying Cycle, then shift to an earlier date the recommendation time of described commodity interested according to described individual's Buying Cycle.
4. a Method of Commodity Recommendation, comprises the following steps:
The commodity bundle interested of user is extracted from user's commodity interested storehouse, this commodity bundle interested comprises the multiple commodity belonging to multiple stage, have successively timing between the plurality of stage, the purchase sequential of the commodity of previous stage is prior to the purchase sequential of the commodity of the latter half;
The ubiquity obtaining the two benches commodity that sequential is adjacent in described commodity bundle interested buys interval duration;
Extract according to the commodity purchasing record of described user the historical record that described user buys the similar commodity of the commodity in described commodity bundle interested, according to described historical record determine described user buy up-to-date time of described similar commodity and described similar commodity in described multiple stage belonging to latest stage;
With the described up-to-date time for starting point, buy according to described ubiquity the recommendation time that interval duration arranges each stage commodity after latest stage described in described commodity bundle interested;
According to the described recommendation time to each stage commodity after described user recommends described latest stage.
5. Method of Commodity Recommendation according to claim 4, is characterized in that, described method also comprises the step of the universality purchase interval duration of the adjacent two benches commodity of sequential in statistics commodity bundle, and this step comprises the following steps:
The purchaser record with users classification is extracted from purchaser record database;
Extract the name keyword of the commodity in the purchaser record of each user, mark the stage of the commodity bundle belonging to each commodity, if the commodity in certain stage match in the name keyword of certain commodity and commodity bundle, then mark this stage that these commodity belong to this commodity bundle;
Add up the average purchase interval duration that each user buys the sequential adjacent two benches commodity of commodity bundle;
Ubiquity according to the adjacent two benches commodity of average purchase interval this sequential of duration calculation of each user of sequential adjacent two benches commodity buys interval duration.
6. Method of Commodity Recommendation according to claim 4, is characterized in that, further comprising the steps of:
Obtain described user and buy interval duration for the individual of the adjacent two benches commodity of sequential in described commodity bundle interested;
If the described individual of sequential adjacent two benches commodity buys the described ubiquity purchase interval duration that interval duration is shorter than this sequential adjacent two benches commodity, then according to the recommendation time of each stage commodity after described individual's purchase interval duration in advance latest stage described in described commodity bundle interested.
7. Method of Commodity Recommendation according to claim 4, is characterized in that, further comprising the steps of:
The commodity that described in the described latest stage commodity extracting described commodity bundle interested, user not yet buys;
The commodity not yet bought described in recommending to described user in real time.
8. a device for recommending the commodity, is characterized in that, comprising:
Commodity extraction module, for extracting the commodity interested of user from user's commodity interested storehouse;
Buying Cycle acquisition module, for obtaining the ubiquity Buying Cycle of described commodity interested;
Up-to-date time statistical module, for extracting according to the commodity purchasing record of described user the historical record that described user buys the similar commodity of described commodity interested, counts according to described historical record the up-to-date time that described user buys described similar commodity;
First recommends set of time module, for the recommendation time of commodity interested according to described ubiquity Buying Cycle and described up-to-date set of time;
First recommending module, for recommending described commodity interested according to the described recommendation time to described user.
9. a device for recommending the commodity, is characterized in that, comprising:
Commodity bundle extraction module, for extracting the commodity bundle interested of user from user's commodity interested storehouse, this commodity bundle interested comprises the multiple commodity belonging to multiple stage, have successively timing between the plurality of stage, the purchase sequential of the commodity of previous stage is prior to the purchase sequential of the commodity of the latter half;
Buy interval duration acquisition module, buy interval duration for the ubiquity obtaining the two benches commodity that sequential is adjacent in described commodity bundle interested;
Up-to-date time and stage acquisition module, for extracting according to the commodity purchasing record of described user the historical record that described user buys the similar commodity of the commodity in described commodity bundle interested, according to described historical record determine described user buy up-to-date time of described similar commodity and described similar commodity in described multiple stage belonging to latest stage;
Second recommends set of time module, for the described up-to-date time for starting point, buy according to described ubiquity the recommendation time that interval duration arranges each stage commodity after latest stage described in described commodity bundle interested;
Second recommending module, for according to the described recommendation time to each stage commodity after described user recommends described latest stage.
10. the device for recommending the commodity according to claim 9, is characterized in that, also comprises:
Do not buy commodity extraction module, for extract described commodity bundle interested described latest stage commodity described in the commodity not yet bought of user;
3rd recommending module, for the commodity not yet bought described in recommending to described user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510064107.7A CN104599160B (en) | 2015-02-06 | 2015-02-06 | Commodity recommendation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510064107.7A CN104599160B (en) | 2015-02-06 | 2015-02-06 | Commodity recommendation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104599160A true CN104599160A (en) | 2015-05-06 |
| CN104599160B CN104599160B (en) | 2020-09-08 |
Family
ID=53124914
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510064107.7A Active CN104599160B (en) | 2015-02-06 | 2015-02-06 | Commodity recommendation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104599160B (en) |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809637A (en) * | 2015-05-18 | 2015-07-29 | 北京京东尚科信息技术有限公司 | Commodity recommending method and system realized by computer |
| CN105205685A (en) * | 2015-07-15 | 2015-12-30 | 北京京东尚科信息技术有限公司 | Method and device for processing virtual commodity data |
| CN106157097A (en) * | 2016-08-22 | 2016-11-23 | 北京京东尚科信息技术有限公司 | Method of Commodity Recommendation and system |
| CN106169141A (en) * | 2015-05-22 | 2016-11-30 | 卡西欧计算机株式会社 | Shopping auxiliary device and shopping householder method |
| CN106970972A (en) * | 2017-03-24 | 2017-07-21 | 广州茶途网络科技有限公司 | A kind of commodity method for pushing and device analyzed based on big data |
| CN106981011A (en) * | 2017-03-20 | 2017-07-25 | 北京小米移动软件有限公司 | Recommendation method, device and the terminal of article |
| CN107590684A (en) * | 2017-08-21 | 2018-01-16 | 时趣互动（北京）科技有限公司 | A kind of method and device that data real time down reward voucher is bought based on user |
| CN107730336A (en) * | 2016-08-12 | 2018-02-23 | 苏宁云商集团股份有限公司 | Commodity method for pushing and device in a kind of online transaction |
| CN107784390A (en) * | 2017-10-19 | 2018-03-09 | 北京京东尚科信息技术有限公司 | Recognition methods, device, electronic equipment and the storage medium of subscriber lifecycle |
| CN107862566A (en) * | 2017-10-17 | 2018-03-30 | 杨明 | A kind of Method of Commodity Recommendation and system |
| CN108122156A (en) * | 2017-10-31 | 2018-06-05 | 深圳市博鑫创科科技有限公司 | A kind of shopping guide method and balance shopping cart based on balance shopping cart |
| CN109002450A (en) * | 2017-06-07 | 2018-12-14 | 北京京东尚科信息技术有限公司 | Information processing method, system and electronic equipment |
| CN109064261A (en) * | 2018-07-12 | 2018-12-21 | 北京京东金融科技控股有限公司 | The method, apparatus and computer readable storage medium of Recommendations |
| CN109447704A (en) * | 2018-10-26 | 2019-03-08 | 湖北大学 | Towards the client interests analysis method being newly sold |
| WO2019072194A1 (en) * | 2017-10-13 | 2019-04-18 | Midea Group Co., Ltd. | Method and system for providing personalized on-location information exchange |
| CN109872211A (en) * | 2017-12-04 | 2019-06-11 | 北京京东尚科信息技术有限公司 | A kind of method and apparatus of object recommendation |
| CN109952590A (en) * | 2016-11-30 | 2019-06-28 | 万事达卡国际公司 | The content delivery system and method for scene driving |
| CN109961351A (en) * | 2019-02-13 | 2019-07-02 | 平安科技（深圳）有限公司 | Information recommendation method, device, storage medium and computer equipment |
| CN110009457A (en) * | 2019-04-09 | 2019-07-12 | 昆山古鳌电子机械有限公司 | A kind of commercial product recommending system based on big data |
| CN110298697A (en) * | 2019-06-24 | 2019-10-01 | 恩亿科(北京)数据科技有限公司 | The method and apparatus for obtaining the shopping need information of user |
| CN110458643A (en) * | 2019-05-15 | 2019-11-15 | 北京邮电大学 | Repeated commodity information recommendation method and electronic equipment based on Fusion Features |
| CN110610402A (en) * | 2019-08-30 | 2019-12-24 | 恩亿科(北京)数据科技有限公司 | Commodity pushing method and computer storage medium |
| CN111027895A (en) * | 2019-05-16 | 2020-04-17 | 珠海随变科技有限公司 | Stock prediction and behavior data collection method, apparatus, device and medium for commodity |
| WO2020215751A1 (en) * | 2019-04-23 | 2020-10-29 | 北京三快在线科技有限公司 | Recommendation method, electronic device and readable storage medium |
| CN112381623A (en) * | 2020-12-04 | 2021-02-19 | 南京浪脆电子商务有限公司 | Electronic commerce platform commodity intelligent recommendation method based on big data analysis |
| CN112418969A (en) * | 2020-05-12 | 2021-02-26 | 上海哔哩哔哩科技有限公司 | Commodity matching method and device and computer equipment |
| CN113301094A (en) * | 2020-09-10 | 2021-08-24 | 阿里巴巴集团控股有限公司 | Information pushing method and device |
| CN113538141A (en) * | 2021-07-14 | 2021-10-22 | 中数通信息有限公司 | Product recommendation method based on customer information |
| CN115204940A (en) * | 2022-06-22 | 2022-10-18 | 北京森科医药有限公司 | Medicine selling method, device, equipment and computer readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010039519A1 (en) * | 2000-03-08 | 2001-11-08 | Richards David John | Cooperative buying system for purchasing consumer products using a computer network |
| US20070094066A1 (en) * | 2005-10-21 | 2007-04-26 | Shailesh Kumar | Method and apparatus for recommendation engine using pair-wise co-occurrence consistency |
| US20120036037A1 (en) * | 2010-08-03 | 2012-02-09 | Alibaba Group Holding Limited | Product reccommendation system |
| CN102479366A (en) * | 2010-11-25 | 2012-05-30 | 阿里巴巴集团控股有限公司 | Commodity recommending method and system |
| US20130041774A1 (en) * | 2010-03-30 | 2013-02-14 | Rakuten, Inc. | Product recommendation device, product recommendation method, program, and recording medium |
| CN103116581A (en) * | 2011-11-16 | 2013-05-22 | 阿里巴巴集团控股有限公司 | Recommendation method and recommendation device of electronic information |
| CN103325047A (en) * | 2012-03-22 | 2013-09-25 | 纽海信息技术（上海）有限公司 | Device and method for guiding online shopping |
-
2015
- 2015-02-06 CN CN201510064107.7A patent/CN104599160B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010039519A1 (en) * | 2000-03-08 | 2001-11-08 | Richards David John | Cooperative buying system for purchasing consumer products using a computer network |
| US20070094066A1 (en) * | 2005-10-21 | 2007-04-26 | Shailesh Kumar | Method and apparatus for recommendation engine using pair-wise co-occurrence consistency |
| US20130041774A1 (en) * | 2010-03-30 | 2013-02-14 | Rakuten, Inc. | Product recommendation device, product recommendation method, program, and recording medium |
| US20120036037A1 (en) * | 2010-08-03 | 2012-02-09 | Alibaba Group Holding Limited | Product reccommendation system |
| CN102479366A (en) * | 2010-11-25 | 2012-05-30 | 阿里巴巴集团控股有限公司 | Commodity recommending method and system |
| CN103116581A (en) * | 2011-11-16 | 2013-05-22 | 阿里巴巴集团控股有限公司 | Recommendation method and recommendation device of electronic information |
| CN103325047A (en) * | 2012-03-22 | 2013-09-25 | 纽海信息技术（上海）有限公司 | Device and method for guiding online shopping |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809637A (en) * | 2015-05-18 | 2015-07-29 | 北京京东尚科信息技术有限公司 | Commodity recommending method and system realized by computer |
| CN106169141A (en) * | 2015-05-22 | 2016-11-30 | 卡西欧计算机株式会社 | Shopping auxiliary device and shopping householder method |
| CN105205685A (en) * | 2015-07-15 | 2015-12-30 | 北京京东尚科信息技术有限公司 | Method and device for processing virtual commodity data |
| CN107730336A (en) * | 2016-08-12 | 2018-02-23 | 苏宁云商集团股份有限公司 | Commodity method for pushing and device in a kind of online transaction |
| CN110135915B (en) * | 2016-08-22 | 2023-05-02 | 北京京东尚科信息技术有限公司 | Commodity recommendation method and system |
| CN106157097A (en) * | 2016-08-22 | 2016-11-23 | 北京京东尚科信息技术有限公司 | Method of Commodity Recommendation and system |
| CN110135915A (en) * | 2016-08-22 | 2019-08-16 | 北京京东尚科信息技术有限公司 | Method of Commodity Recommendation and system |
| CN109952590A (en) * | 2016-11-30 | 2019-06-28 | 万事达卡国际公司 | The content delivery system and method for scene driving |
| CN106981011A (en) * | 2017-03-20 | 2017-07-25 | 北京小米移动软件有限公司 | Recommendation method, device and the terminal of article |
| CN106970972A (en) * | 2017-03-24 | 2017-07-21 | 广州茶途网络科技有限公司 | A kind of commodity method for pushing and device analyzed based on big data |
| CN109002450A (en) * | 2017-06-07 | 2018-12-14 | 北京京东尚科信息技术有限公司 | Information processing method, system and electronic equipment |
| CN107590684A (en) * | 2017-08-21 | 2018-01-16 | 时趣互动（北京）科技有限公司 | A kind of method and device that data real time down reward voucher is bought based on user |
| US10789638B2 (en) | 2017-10-13 | 2020-09-29 | Midea Group Co., Ltd. | Method and system for providing personalized on-location information exchange |
| WO2019072194A1 (en) * | 2017-10-13 | 2019-04-18 | Midea Group Co., Ltd. | Method and system for providing personalized on-location information exchange |
| CN107862566A (en) * | 2017-10-17 | 2018-03-30 | 杨明 | A kind of Method of Commodity Recommendation and system |
| CN107784390A (en) * | 2017-10-19 | 2018-03-09 | 北京京东尚科信息技术有限公司 | Recognition methods, device, electronic equipment and the storage medium of subscriber lifecycle |
| CN108122156A (en) * | 2017-10-31 | 2018-06-05 | 深圳市博鑫创科科技有限公司 | A kind of shopping guide method and balance shopping cart based on balance shopping cart |
| CN109872211A (en) * | 2017-12-04 | 2019-06-11 | 北京京东尚科信息技术有限公司 | A kind of method and apparatus of object recommendation |
| CN109064261A (en) * | 2018-07-12 | 2018-12-21 | 北京京东金融科技控股有限公司 | The method, apparatus and computer readable storage medium of Recommendations |
| CN109447704A (en) * | 2018-10-26 | 2019-03-08 | 湖北大学 | Towards the client interests analysis method being newly sold |
| CN109961351A (en) * | 2019-02-13 | 2019-07-02 | 平安科技（深圳）有限公司 | Information recommendation method, device, storage medium and computer equipment |
| CN109961351B (en) * | 2019-02-13 | 2023-02-07 | 平安科技（深圳）有限公司 | Information recommendation method and device, storage medium and computer equipment |
| CN110009457A (en) * | 2019-04-09 | 2019-07-12 | 昆山古鳌电子机械有限公司 | A kind of commercial product recommending system based on big data |
| WO2020215751A1 (en) * | 2019-04-23 | 2020-10-29 | 北京三快在线科技有限公司 | Recommendation method, electronic device and readable storage medium |
| CN110458643A (en) * | 2019-05-15 | 2019-11-15 | 北京邮电大学 | Repeated commodity information recommendation method and electronic equipment based on Fusion Features |
| CN110458643B (en) * | 2019-05-15 | 2022-07-01 | 北京邮电大学 | Repetitive commodity information recommendation method based on feature fusion and electronic equipment |
| CN111027895A (en) * | 2019-05-16 | 2020-04-17 | 珠海随变科技有限公司 | Stock prediction and behavior data collection method, apparatus, device and medium for commodity |
| CN110298697A (en) * | 2019-06-24 | 2019-10-01 | 恩亿科(北京)数据科技有限公司 | The method and apparatus for obtaining the shopping need information of user |
| CN110610402A (en) * | 2019-08-30 | 2019-12-24 | 恩亿科(北京)数据科技有限公司 | Commodity pushing method and computer storage medium |
| CN112418969A (en) * | 2020-05-12 | 2021-02-26 | 上海哔哩哔哩科技有限公司 | Commodity matching method and device and computer equipment |
| CN112418969B (en) * | 2020-05-12 | 2023-07-04 | 上海哔哩哔哩科技有限公司 | Commodity matching method, commodity matching device and computer equipment |
| CN113301094A (en) * | 2020-09-10 | 2021-08-24 | 阿里巴巴集团控股有限公司 | Information pushing method and device |
| CN112381623A (en) * | 2020-12-04 | 2021-02-19 | 南京浪脆电子商务有限公司 | Electronic commerce platform commodity intelligent recommendation method based on big data analysis |
| CN112381623B (en) * | 2020-12-04 | 2021-07-06 | 深圳市快云科技有限公司 | Electronic commerce platform commodity intelligent recommendation method based on big data analysis |
| CN113538141A (en) * | 2021-07-14 | 2021-10-22 | 中数通信息有限公司 | Product recommendation method based on customer information |
| CN115204940A (en) * | 2022-06-22 | 2022-10-18 | 北京森科医药有限公司 | Medicine selling method, device, equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104599160B (en) | 2020-09-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104599160A (en) | Commodity recommendation method and commodity recommendation device | |
| CN104111941B (en) | The method and apparatus that information is shown | |
| CN103577432B (en) | A kind of Commodity Information Search method and system | |
| CN103729359B (en) | A kind of method and system recommending search word | |
| CN108665333B (en) | Commodity recommendation method and device, electronic equipment and storage medium | |
| JP5449628B2 (en) | Determining category information using multistage | |
| CN103473230B (en) | Service area determines that method, logistics service provider recommend method and related device | |
| CN106033415B (en) | Text content recommendation method and device | |
| CN103902545B (en) | A kind of classification path identification method and system | |
| CN102663064B (en) | A kind of disposal route of favorites data and device | |
| CN103440259A (en) | Network advertisement push method and device | |
| CN106126582A (en) | Recommend method and device | |
| CN105740268A (en) | Information pushing method and apparatus | |
| CN103365904B (en) | A kind of advertising message searching method and system | |
| CN103914484A (en) | Method and device for generating page contents | |
| CN105247507A (en) | Influence score of a brand | |
| CN106445963B (en) | Advertisement index keyword automatic generation method and device of APP platform | |
| CN105023178B (en) | A kind of electronic commerce recommending method based on ontology | |
| CN103778151A (en) | Method and device for identifying characteristic group and search method and device | |
| CN106062743A (en) | Systems and methods for keyword suggestion | |
| CN106446189A (en) | Message-recommending method and system | |
| CN103324631B (en) | The method and device of data search is provided | |
| TWI539306B (en) | Information delivery method, processing server and merge server | |
| CN110197404A (en) | The personalized long-tail Method of Commodity Recommendation and system of popularity deviation can be reduced | |
| CN103955480A (en) | Method and equipment for determining target object information corresponding to user |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 |