CN108614832A - A kind of user individual commercial articles searching implementation method and device - Google Patents
A kind of user individual commercial articles searching implementation method and device Download PDFInfo
- Publication number
- CN108614832A CN108614832A CN201611142235.XA CN201611142235A CN108614832A CN 108614832 A CN108614832 A CN 108614832A CN 201611142235 A CN201611142235 A CN 201611142235A CN 108614832 A CN108614832 A CN 108614832A
- Authority
- CN
- China
- Prior art keywords
- user
- item property
- score
- merchandise news
- search
- 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
Abstract
The present invention relates to Internet technical field, it is related to a kind of user individual commercial articles searching implementation method and device so that search result varies with each individual, to improve the matching degree between the search intention of user and the commodity result of displaying.Described search method includes：According to search term input by user, commercial articles searching is carried out, the merchandise news of initial search is obtained；According to search term input by user, the product features in user individual characteristic are filtered；According to filtered user individual characteristic, the merchandise news for meeting user individual characteristic is extracted from the merchandise news of the initial search, as merchandise news to be pushed；The merchandise news for treating push is ranked up, and the merchandise news after sequence is pushed to user.
Description
Technical field
The present invention relates to Internet technical field, more particularly to a kind of user individual commercial articles searching implementation method and dress
It sets.
Background technology
Traditional search engine always shows identical search result to same search term.However in reality, due to search
Skill is different, even if to being different if demand of the different types of user of same search term to the same kind of goods, i.e., in many feelings
It will appear that there are the feelings that the user of different demands (such as demand similar or similar commodity) uses identical search term under condition
Condition.Therefore, the search model of this " general-purpose type " (one-size-fits-all) reduces search result to a certain extent
Correlation.A new generation search engine need consider user preference and current background, the search intention of deep understanding user,
Personalized search service is provided, user is allowed to be more convenient, more quickly search and admire commodity.How deep understanding different user
Search intention, and meet the commodity of its search intention to user's displaying, be the technical barrier that those skilled in the art face.
Invention content
The object of the present invention is to provide a kind of user individual commercial articles searching implementation method and devices so that search knot
Fruit varies with each individual, to improve the matching degree between the search intention of user and the commodity result of displaying.
To achieve the above object, technical solution used in the embodiment of the present invention is：
In a first aspect, the embodiment of the present invention provides a kind of user individual commercial articles searching implementation method, this method includes：
According to search term input by user, commercial articles searching is carried out, the merchandise news of initial search is obtained；
According to search term input by user, the product features in user individual characteristic are filtered；
According to filtered user individual characteristic, is extracted from the merchandise news of the initial search and meet user
The merchandise news of individualized feature data, as merchandise news to be pushed；
The merchandise news for treating push is ranked up, and the merchandise news after sequence is pushed to user.
With reference to first aspect, described according to search term input by user as the first mode in the cards, to
Product features in the individualized feature data of family are filtered, including：
Search term input by user is obtained, described search word includes product features data；
The individualized feature data of calling and obtaining user, and by product features data and the user individual characteristic in search term
According to being compared, if it is different, then retaining the product features data in search term, corresponding commodity in individualized feature data are filtered
Characteristic.
With reference to first aspect or the first mode in the cards, as second of mode in the cards, the use
The individualized feature data at family generate by the following method：
According to user in the user behaviors log of e-commerce website, item property is extracted；
According to behavior weight, item property feature scores are calculated；
According to user information, and the item property feature scores of calculating, the individualized feature data of user are generated.
Second of mode in the cards with reference to first aspect, as the third mode in the cards, the use
The individualized feature data at family are indicated using the vector of item property feature scores composition.
With reference to first aspect, as the 4th kind of mode in the cards, the merchandise news for treating push is arranged
Sequence specifically includes：
The item property of each commodity in merchandise news to be pushed is extracted, and extracts the corresponding non-individual character of the item property
Change ranking score；
Obtain the personalized score of item property；
According to the personalized score of the impersonal theory ranking score of the item property of extraction and the item property of acquisition, calculate
The personalized ordering score of commodity；
According to the personalized ordering score of commodity, commodity are ranked up.
The 4th kind of mode in the cards with reference to first aspect, as the 5th kind of mode in the cards, the meter
The personalized ordering score for calculating commodity, specifically includes：
By the impersonal theory ranking score weight phase of the impersonal theory ranking score of the item property of extraction and item property
Multiply, obtains the first parameter；
By the personalized score multiplied by weight of the personalized score of item property and item property, the second parameter is obtained；
First parameter is added with the second parameter, obtains the personalized ordering score of item property；
The personalized ordering score of each item property is done into normalized, obtains the personalized ordering score of commodity.
Second aspect, the embodiment of the present invention also provide a kind of user individual commercial articles searching realization device, which includes：
Search module：For according to search term input by user, carrying out commercial articles searching, the commodity letter of initial search is obtained
Breath；
Filtering module：For according to search term input by user, to the product features in user individual characteristic into
Row filtering；
Extraction module：For the individualized feature data according to filtered user, believe from the commodity of the initial search
Extraction meets the merchandise news of the individualized feature data of user in breath, as merchandise news to be pushed；
Sorting module：Merchandise news for treating push is ranked up；
Pushing module：For the merchandise news after sequence to be pushed to user.
In conjunction with second aspect, as the first mode in the cards, the filtering module includes：
Acquisition submodule：For obtaining search term input by user, described search word includes product features data；
Comparison sub-module：For the individualized feature data of calling and obtaining user, and by search term product features data with
User individual characteristic is compared, if it is different, then retaining the product features data in search term, filters individualized feature
Corresponding product features data in data.
In conjunction with the first of second aspect or second aspect mode in the cards, as second of side in the cards
Formula, the individualized feature data of the user with lower unit by being generated：
Extraction unit：For, in the user behaviors log of e-commerce website, extracting item property according to user；
Computing unit：For according to behavior weight, calculating the item property feature scores of extraction unit extraction；
Generation unit：For the item property feature scores according to user information and computing unit calculating, user is generated
Individualized feature data.
In conjunction with second of mode in the cards of second aspect, as the third mode in the cards, the use
The individualized feature data at family are indicated using the vector of item property feature scores composition.
In conjunction with second aspect, as the 4th kind of mode in the cards, the sorting module specifically includes：
Extracting sub-module：Item property for extracting each commodity in merchandise news to be pushed, and extract the commodity
The corresponding impersonal theory ranking score of attribute；
Acquisition submodule：Personalized score for obtaining item property；
Computational submodule：For according to the item property of the impersonal theory ranking score and acquisition of the item property of extraction
Personalized score calculates the personalized ordering score of commodity；
Sorting sub-module：According to the personalized ordering score of commodity, commodity are ranked up.
In conjunction with the 4th kind of mode in the cards of second aspect, as the 5th kind of mode in the cards, the meter
Operator module specifically includes：
First acquisition unit：The impersonal theory ranking score of item property for that will extract and the non-individual character of item property
Change ranking score multiplied by weight, obtains the first parameter；
Second acquisition unit：By the personalized score multiplied by weight of the personalized score of item property and item property, obtain
Obtain the second parameter；
Third acquiring unit：For the first parameter to be added with the second parameter, the personalized ordering point of item property is obtained
Number；
4th acquiring unit：For the personalized ordering score of each item property to be done normalized, commodity are obtained
Personalized ordering score.
Compared with prior art, the user individual commercial articles searching implementation method and device of the embodiment of the present invention, Ke Yiti
Matching degree between the search intention of high user and the commodity result of displaying.Existing search engine is to the total exhibitions of same search term
Existing identical search result.And the implementation method of the present embodiment, by the individualized feature data of user, from the quotient of initial search
Then the merchandise news that extraction meets the individualized feature data of user in product information is treated and is pushed away as merchandise news to be pushed
The merchandise news sent is ranked up, and the merchandise news after sequence is pushed to user.In this method, the individualized feature of user
Data are different because of user's difference.Therefore, identical initial search is obtained even from identical search term input by user
Merchandise news, also can be because of the individualized feature data of user, and generate different merchandise news to be pushed so that wait pushing
Merchandise news meet the search intention of each user.
Description of the drawings
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to needed in the embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for ability
For the those of ordinary skill of domain, without creative efforts, it can also be obtained according to these attached drawings other attached
Figure.
Fig. 1 is the method flow block diagram of the embodiment of the present invention；
Fig. 2 is the information exchange flow graph of the embodiment of the present invention；
Fig. 3 is the system structure diagram of the embodiment of the present invention.
Specific implementation mode
Below in conjunction with the accompanying drawings, the technical solution of the embodiment of the present invention is described in detail.
To make those skilled in the art more fully understand technical scheme of the present invention, below in conjunction with the accompanying drawings and specific embodiment party
Present invention is further described in detail for formula.Embodiments of the present invention are described in more detail below, the embodiment is shown
Example is shown in the accompanying drawings, and in which the same or similar labels are throughly indicated same or similar element or has identical or class
Like the element of function.It is exemplary below with reference to the embodiment of attached drawing description, is only used for explaining the present invention, and cannot
It is construed to limitation of the present invention.Those skilled in the art of the present technique are appreciated that unless expressly stated, odd number shape used herein
Formula " one ", "one", " described " and "the" may also comprise plural form.It is to be further understood that the specification of the present invention
The middle wording " comprising " used refers to that there are the feature, integer, step, operation, element and/or component, but it is not excluded that
Other one or more features of presence or addition, integer, step, operation, element, component and/or their group.It should be understood that
When we say that an element is " connected " or " coupled " to another element, it can be directly connected or coupled to other elements, or
There may also be intermediary elements.In addition, " connection " used herein or " coupling " may include being wirelessly connected or coupling.Here make
Wording "and/or" includes any cell of one or more associated list items and all combines.The art
Technical staff is appreciated that unless otherwise defined all terms (including technical terms and scientific terms) used herein have
Meaning identical with the general understanding of the those of ordinary skill in fields of the present invention.It should also be understood that such as general
Term, which should be understood that, those of defined in dictionary has a meaning that is consistent with the meaning in the context of the prior art, and
Unless being defined as here, will not be explained with the meaning of idealization or too formal.
As shown in Figure 1, the embodiment of the present invention provides a kind of user individual commercial articles searching implementation method, including：
S10 carries out commercial articles searching, obtains the merchandise news of initial search according to search term input by user；
S20 is filtered the product features in user individual characteristic according to search term input by user；
S30 is extracted from the merchandise news of the initial search and is met according to filtered user individual characteristic
The merchandise news of user individual characteristic, as merchandise news to be pushed；
The merchandise news that S40 treats push is ranked up, and the merchandise news after sequence is pushed to user.
Using the method for above-described embodiment, different search results can be shown to different user.Even if different user is defeated
Enter identical search term, can also show different search results.The process employs the individualized feature data of user.Difference is used
The individualized feature data at family are different.Extraction meets the individualized feature data of user from the merchandise news of initial search
Merchandise news as merchandise news to be pushed.Even if the merchandise news of initial search is identical, because of the individualized feature of user
Data are different, so the merchandise news to be pushed generated is different.This realizes search result and varies with each individual, that is, in fact
Personalized search is showed.
In above-described embodiment, in step S20, according to search term input by user, in user individual characteristic
Product features are filtered, including：
Search term input by user is obtained, described search word includes product features data；
The individualized feature data of calling and obtaining user, and by product features data and the user individual characteristic in search term
According to being compared, if it is different, then retaining the product features data in search term, corresponding commodity in individualized feature data are filtered
Characteristic.
When user clearly limits product features in search term, if in the individuation data of the user including such spy
Sign, then filter out such characteristic in the individuation data to currently extracting, i.e., do not do personalisation process to this feature, and
Only retain the individualized feature data not occurred in search term.That is, when user clearly limits commodity spy in search term
When sign, then personalisation process is not done to this feature.For example, user searches for " black iphone6s ".Although the past history of user
Show that the user likes the mobile phone of white, personalization system still shields the personalization of mobile phone color, only shows black
Mobile phone is as search result.
In above-described embodiment, the individualized feature data of user are the product attribute informations based on user preference.User's
Individualized feature data are established by the following method：
S201, in the user behaviors log of e-commerce website, extracts item property according to user；
In this step, by analyzing user in the user behaviors log of e-commerce website, such as purchase is clicked, search history
Deng the extraction interested item property of user.Item property includes commodity color, model, brand, style etc..Due to different use
The interest at family is different, and behavior is also different, therefore, by analyzing user in the user behaviors log of e-commerce website, extracts quotient
Product attribute is also inevitable different.In this way, being directed to different users, different item properties is extracted.The user is established to each user
To color, function, the preference of the item properties such as size.In this step, user can be analyzed recently for a period of time in e-commerce
The user behaviors log of website, such as the user behaviors log in three months.
S202 calculates item property feature scores according to behavior weight；
Different behaviors generates item property different influences.Such as check behavior relative to click, buying behavior is more
The search intention of user can be embodied.In this way, the weight of buying behavior is more than the weight of click behavior.In another example in search history
In, search is primary repeatedly to be compared with search, and search repeatedly can more embody the buying intention of user.In this way, searching for multiple behavior
The weight of behavior of weight great-than search.Usually, the weight of buying behavior is more than the weight of click behavior, clicks behavior
Weight great-than search behavior weight.Weight concrete numerical value can be arranged according to concrete condition.According to behavior weight, calculate
Item property feature scores.Item property feature scores are higher, and it is bigger to illustrate that user searches for the intention bought.It, can in this step
To analyze in user's nearest a period of time, such as three months, in the user behaviors log of e-commerce website, each behavior is divided
Analysis obtains behavior weight.
The item property feature scores that S203 is calculated according to user information and S202, generate the individualized feature of user
Data.
User information refers to the information that user registers on e-commerce website, including the age, gender, residence etc. basic letter
Breath.The individualized feature data of user are generated according to user information and S202 the item property feature scores calculated.Each user
Individualized feature data by the feature scores of item property form vector indicate.Vector dimension include user information and partially
Good product features, such as age, color.The value of each dimension storage is personalized score of the user in the dimension.
The individualized feature data of user, are arranged according to user individual.Each user generates respective personalized special
Levy data.User is different, and individualized feature data are also different.The individualized feature data for the user that the step generates, Ke Yicun
It stores up in the database, such as REDIS databases.The individualized feature data of user can be generated by off-line module, can also be by
It is generated in wire module.The individualized feature data of user are called for front end.
As preference, in step S40, the merchandise news for treating push is ranked up, and is specifically included：
S401 extracts the item property of each commodity in merchandise news to be pushed, and it is corresponding non-to extract the item property
Personalized ordering score.The impersonal theory ranking score of the item property refers to row of the general search system to each commodity
Sequence score.The score comes from upstream ordering system, includes the degree of correlation to search term, the considerations such as stock position.
S402 obtains the personalized score of item property；
Wherein, if item property is matched with the item property in the individualized feature data of user, by the item property
Personalized score of the corresponding score as item property in individualized feature data.If the individual character of item property and user
The item property changed in characteristic mismatches, then the personalized score of the item property is considered as 0.
S403 according to the personalized score of the impersonal theory ranking score of the item property of extraction and the item property of acquisition,
Calculate the personalized ordering score of commodity；
Wherein, as a preference, the personalized ordering score of the calculating commodity specifically includes：
S4031 weighs the impersonal theory ranking score of the item property of extraction and the impersonal theory ranking score of item property
Heavy phase multiplies, and obtains the first parameter；
The personalized score multiplied by weight of the personalized score of item property and item property is obtained the second ginseng by S4032
Number；
First parameter is added by S4033 with the second parameter, obtains the personalized ordering score of item property；
The personalized ordering score of each item property is done normalized by S4034, obtains the personalized ordering of commodity
Score.
In addition to the above method, other methods can also be used to calculate the personalized ordering score of commodity, for example, by extraction
The impersonal theory ranking score of item property is added with the personalized score of item property.Using above-mentioned preferred method, synthesis is examined
The weight of different scores is considered, and result has been normalized, the personalized ordering score for obtaining commodity is more accurate.
S404 is ranked up commodity according to the personalized ordering score of the step S403 commodity calculated.Preferably, pressing
According to the personalized ordering score of commodity, sort from high to low.
In the sort method, personalized ordering is based on the search word correlation in common sequence.If original common sequence
As a result uncorrelated, personalized ordering can not help user to find the product that oneself is satisfied with.For example, user searches for " black
When iPhone7 ", if the result that original common ordering system is shown all is " iPhone6 ".Even if personalized search can be by user
The black iPhone results liked shift to an earlier date, and user can not find oneself desired product.
In the sort method, according to the personalized ordering score of commodity, commodity are ranked up.The personalized ordering of commodity
This parameter of personalized score containing item property in score.The personalized score of item property is the personalization from user
It is obtained in characteristic.In this way, the individualized feature data of different user are different, the personalized score of corresponding item property
It is different.This makes the personalized ordering score of finally obtained commodity, embodies the hobby of user.The personalized ordering of commodity point
Number is higher, and the buying intention of user is also bigger.
The correlation of the method for the present embodiment, deep understanding user demand, search result and user preference improves, and can help
User is further clear or reduces search target and range, shortens user's search time.When the merchandise news symbol pushed to user
When closing its search intention, page clicking rate can be improved, purchase conversion rate is clicked to improve.
The method of the present embodiment also improves the probability that new commodity information is pushed to user.In this method, to new commodity
It can carry out individualized fit.The method of the present embodiment establishes being associated with for user and product features by the historical record of user.
After extracting the feature of new commodity, as long as the hobby with user is consistent, the sequence of the new commodity can be improved, new commodity is shown
To user.And the prior art is usually using commodity number of clicks as important parameters sortnig so that new commodity is because being clicked number
It is less, and user can not be pushed to.
The method of above-described embodiment is the individual searching engine based on e-commerce, is not based on the individual character of web document
Change search engine.The currently used search engine based on web document, such as Google, Baidu ordinary search engine, mainly
It is to provide other web documents " answer " to user's search term.And the method for the present embodiment mainly solves user in e-commerce
Buy the problem of which kind of commodity in website.The method personalization of the present embodiment establishes the connection of user and merchandise news, makes user
The commodity oneself liked can be quickly and easily bought on e-commerce website.And the personalization of ordinary search engine is intended to understand
The search terms of user provide user's web results wanted.
In the method for above-described embodiment, by learning the historical record of user, user characteristic data is extracted, it is inclined to find user
It is good.Ranking results are adjusted so that be located further forward with the maximally related model sequencing of user.The system is better understood on user
Behavioral trait, new resource is provided for search service, to provide the user with suitable search result, improve search result with
The matching degree of the search intention of user promotes user's search experience.
In the method for above-described embodiment, the information pushed to user is commodity data.And these commodity datas are according to user
The search term of input and the individualized feature data acquisition of user.The method of the present embodiment is the process in user's search commercial articles
In, a kind of optimization to search result, what it is to user feedback is the specific commodity purchasing page (or link).The side of the present embodiment
Method, by the extraction that is recorded to user's history, cluster, working process, then the matching to product information of reaching the standard grade, to finally searching for knot
The webpage of fruit reorders so that user can find oneself desired product in the first few items of webpage.The side of the present embodiment
Method, by personalized search, the merchandise news merchandise news that necessarily user searches for pushed to user, for example, search " TV
The search result of machine " is limited to television product, but for the brand or function of television set, can be right according to liking for user
Searching results optimize.
As shown in Fig. 2, the method for the embodiment of the present invention can be realized by client, foreground and backstage.Wherein client
End inputs search term for user and final search result is showed user by foreground.Foreground according to user for inputting
Search term, search commercial articles information.Individualized feature data of the backstage for generating user, specifically include the behavior according to user
Item property is extracted in daily record；According to behavior weight, item property feature scores are calculated；According to user information, and the quotient of calculating
Product attributive character score generates the individualized feature data of user.Foreground needs to transfer backstage generation in search commercial articles information
User individualized feature data, and extraction meets the individualized feature data of user from the merchandise news of initial search
As merchandise news to be pushed, the merchandise news for then treating push is ranked up merchandise news, finally by the quotient after sequence
Product information is pushed to client, is shown to user.
As shown in figure 3, the embodiment of the present invention additionally provides a kind of user individual commercial articles searching realization device, including：
Search module：For according to search term input by user, carrying out commercial articles searching, the commodity letter of initial search is obtained
Breath；
Filtering module：For according to search term input by user, to the product features in user individual characteristic into
Row filtering；
Extraction module：For the individualized feature data according to filtered user, believe from the commodity of the initial search
Extraction meets the merchandise news of the individualized feature data of user in breath, as merchandise news to be pushed；
Sorting module：Merchandise news for treating push is ranked up；
Pushing module：For the merchandise news after sequence to be pushed to user.
In above-mentioned apparatus, as preference, filtering module includes：
Acquisition submodule：For obtaining search term input by user, described search word includes product features data；
Comparison sub-module：For the individualized feature data of calling and obtaining user, and by search term product features data with
User individual characteristic is compared, if it is different, then retaining the product features data in search term, filters individualized feature
Corresponding product features data in data.
In above-mentioned apparatus, user individual characteristic can be generated by off-line module, be specifically included with lower unit：
Extraction unit：For, in the user behaviors log of e-commerce website, extracting item property according to user；
Computing unit：For according to behavior weight, calculating the item property feature scores of extraction unit extraction；
Generation unit：For the item property feature scores according to user information and computing unit calculating, user is generated
Individualized feature data.
By extraction unit, computing unit and generation unit, the individualized feature data of user are generated.The personalization of user
The vector that item property feature scores composition may be used in characteristic indicates.
In above-mentioned apparatus, preferably, the sorting module, specifically includes：
Extracting sub-module：Item property for extracting each commodity in merchandise news to be pushed, and extract the commodity
The corresponding impersonal theory ranking score of attribute；
Acquisition submodule：Personalized score for obtaining item property；
Computational submodule：For according to the item property of the impersonal theory ranking score and acquisition of the item property of extraction
Personalized score calculates the personalized ordering score of commodity；
Sorting sub-module：According to the personalized ordering score of commodity, commodity are ranked up.
Wherein, the computational submodule, specifically includes：
First acquisition unit：The impersonal theory ranking score of item property for that will extract and the non-individual character of item property
Change ranking score multiplied by weight, obtains the first parameter；
Second acquisition unit：By the personalized score multiplied by weight of the personalized score of item property and item property, obtain
Obtain the second parameter；
Third acquiring unit：For the first parameter to be added with the second parameter, the personalized ordering point of item property is obtained
Number；
4th acquiring unit：For the personalized ordering score of each item property to be done normalized, commodity are obtained
Personalized ordering score.
Using the device of above-described embodiment, different search results can be shown to different user.Even if different user is defeated
Enter identical search term, can also show different search results.The device uses the individualized feature data of user.Difference is used
The individualized feature data at family are different.Extraction meets the individualized feature data of user from the merchandise news of initial search
Merchandise news as merchandise news to be pushed.Even if the merchandise news of initial search is identical, because of the individualized feature of user
Data are different, so the merchandise news to be pushed generated is different.This realizes search result and varies with each individual, that is, in fact
Personalized search is showed.
Each embodiment in this specification is described in a progressive manner, identical similar portion between each embodiment
Point just to refer each other, and each embodiment focuses on the differences from other embodiments.Especially for equipment reality
For applying example, since it is substantially similar to the method embodiment, so describing fairly simple, related place is referring to embodiment of the method
Part explanation.
One of ordinary skill in the art will appreciate that realizing all or part of flow in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the program can be stored in a computer read/write memory medium
In, the program is when being executed, it may include such as the flow of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic
Dish, CD, read-only memory (Read-Only Memory, ROM) or random access memory (Random Access
Memory, RAM) etc..
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, all answer by the change or replacement that can be readily occurred in
It is included within the scope of the present invention.Therefore, protection scope of the present invention should be subject to the protection scope in claims.
Claims (12)
1. a kind of user individual commercial articles searching implementation method, which is characterized in that this method includes：
According to search term input by user, commercial articles searching is carried out, the merchandise news of initial search is obtained；
According to search term input by user, the product features in user individual characteristic are filtered；
According to filtered user individual characteristic, is extracted from the merchandise news of the initial search and meet user personality
The merchandise news for changing characteristic, as merchandise news to be pushed；
The merchandise news for treating push is ranked up, and the merchandise news after sequence is pushed to user.
2. according to the method described in claim 1, it is characterized in that, described according to search term input by user, to user
Product features in property characteristic are filtered, including：
Search term input by user is obtained, described search word includes product features data；
The individualized feature data of calling and obtaining user, and by product features data and the user individual characteristic in search term into
Row compares, if it is different, then retaining the product features data in search term, filters corresponding product features in individualized feature data
Data.
3. method according to claim 1 or 2, which is characterized in that the individualized feature data of the user by with
Lower method generates：
According to user in the user behaviors log of e-commerce website, item property is extracted；
According to behavior weight, item property feature scores are calculated；
According to user information, and the item property feature scores of calculating, the individualized feature data of user are generated.
4. according to the method described in claim 3, it is characterized in that, the individualized feature data of the user use commodity category
Property feature scores composition vector indicate.
5. according to the method described in claim 1, it is characterized in that, the merchandise news for treating push is ranked up, have
Body includes：
The item property of each commodity in merchandise news to be pushed is extracted, and extracts the corresponding impersonal theory row of the item property
Sequence score；
Obtain the personalized score of item property；
According to the personalized score of the impersonal theory ranking score of the item property of extraction and the item property of acquisition, commodity are calculated
Personalized ordering score；
According to the personalized ordering score of commodity, commodity are ranked up.
6. according to the method described in claim 5, it is characterized in that, the personalized ordering score of the described calculating commodity, specifically
Including：
By the impersonal theory ranking score multiplied by weight of the impersonal theory ranking score of the item property of extraction and item property, obtain
Obtain the first parameter；
By the personalized score multiplied by weight of the personalized score of item property and item property, the second parameter is obtained；
First parameter is added with the second parameter, obtains the personalized ordering score of item property；
The personalized ordering score of each item property is done into normalized, obtains the personalized ordering score of commodity.
7. a kind of user individual commercial articles searching realization device, which is characterized in that the device includes：
Search module：For according to search term input by user, carrying out commercial articles searching, obtaining the merchandise news of initial search；
Filtering module：For according to search term input by user, being carried out to the product features in user individual characteristic
Filter；
Extraction module：For the individualized feature data according to filtered user, from the merchandise news of the initial search
Extraction meets the merchandise news of the individualized feature data of user, as merchandise news to be pushed；
Sorting module：Merchandise news for treating push is ranked up；
Pushing module：For the merchandise news after sequence to be pushed to user.
8. device according to claim 7, which is characterized in that the filtering module includes：
Acquisition submodule：For obtaining search term input by user, described search word includes product features data；
Comparison sub-module：For the individualized feature data of calling and obtaining user, and by search term product features data and user
Individualized feature data are compared, if it is different, then retaining the product features data in search term, filter individualized feature data
In corresponding product features data.
9. device according to claim 7 or 8, which is characterized in that the individualized feature data of the user by with
Lower unit generates：
Extraction unit：For, in the user behaviors log of e-commerce website, extracting item property according to user；
Computing unit：For according to behavior weight, calculating the item property feature scores of extraction unit extraction；
Generation unit：For the item property feature scores according to user information and computing unit calculating, of user is generated
Property characteristic.
10. device according to claim 9, which is characterized in that the individualized feature data of the user use commodity
The vector of attributive character score composition indicates.
11. device according to claim 7, which is characterized in that the sorting module specifically includes：
Extracting sub-module：Item property for extracting each commodity in merchandise news to be pushed, and extract the item property
Corresponding impersonal theory ranking score；
Acquisition submodule：Personalized score for obtaining item property；
Computational submodule：For the individual character according to the item property of the impersonal theory ranking score and acquisition of the item property of extraction
Change score, calculates the personalized ordering score of commodity；
Sorting sub-module：According to the personalized ordering score of commodity, commodity are ranked up.
12. according to the devices described in claim 11, which is characterized in that the computational submodule specifically includes：
First acquisition unit：The impersonal theory ranking score of item property for that will extract and the impersonal theory of item property are arranged
Sequence fractional weight is multiplied, and obtains the first parameter；
Second acquisition unit：By the personalized score multiplied by weight of the personalized score of item property and item property, the is obtained
Two parameters；
Third acquiring unit：For the first parameter to be added with the second parameter, the personalized ordering score of item property is obtained；
4th acquiring unit：For the personalized ordering score of each item property to be done normalized, of commodity is obtained
Property ranking score.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611142235.XA CN108614832B (en) | 2016-12-12 | 2016-12-12 | Method and device for realizing user personalized commodity search |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611142235.XA CN108614832B (en) | 2016-12-12 | 2016-12-12 | Method and device for realizing user personalized commodity search |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108614832A true CN108614832A (en) | 2018-10-02 |
| CN108614832B CN108614832B (en) | 2022-01-28 |
Family
ID=63657781
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611142235.XA Active CN108614832B (en) | 2016-12-12 | 2016-12-12 | Method and device for realizing user personalized commodity search |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108614832B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110083677A (en) * | 2019-05-07 | 2019-08-02 | 北京字节跳动网络技术有限公司 | Searching method, device, equipment and the storage medium of contact person |
| CN110850997A (en) * | 2019-09-29 | 2020-02-28 | 上海萌家网络科技有限公司 | Input method and system thereof |
| CN111028040A (en) * | 2019-06-17 | 2020-04-17 | 拉扎斯网络科技（上海）有限公司 | Object sorting method and device, readable storage medium and electronic equipment |
| CN111737418A (en) * | 2020-07-20 | 2020-10-02 | 北京每日优鲜电子商务有限公司 | Method, apparatus and storage medium for predicting relevance of search term and commodity |
| CN112102037A (en) * | 2020-09-16 | 2020-12-18 | 汤涛 | Live E-commerce platform commodity content intelligent pushing management system based on big data |
| CN112651839A (en) * | 2021-01-07 | 2021-04-13 | 中国农业银行股份有限公司 | Product optimization method and system |
| CN113297511A (en) * | 2021-05-24 | 2021-08-24 | 北京三快在线科技有限公司 | Method and device for constructing merchant ranking model, server and storage medium |
| WO2022220738A1 (en) * | 2021-04-15 | 2022-10-20 | Grabtaxi Holdings Pte. Ltd. | Communications server apparatus, method and communications system for providing a recommendation of offerings by merchants to a user |
| TWI826881B (en) * | 2021-12-22 | 2023-12-21 | 阿物科技股份有限公司 | System and method for awakening non-shopping consumers |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102446180A (en) * | 2010-10-09 | 2012-05-09 | 腾讯科技（深圳）有限公司 | Commodity searching method and device adopting same |
| CN104933172A (en) * | 2015-06-30 | 2015-09-23 | 百度在线网络技术（北京）有限公司 | Information pushing method and device based on user searching behavior |
| CN105446972A (en) * | 2014-06-17 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Search method, device and system based on and fusing with user relation data |
-
2016
- 2016-12-12 CN CN201611142235.XA patent/CN108614832B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102446180A (en) * | 2010-10-09 | 2012-05-09 | 腾讯科技（深圳）有限公司 | Commodity searching method and device adopting same |
| CN105446972A (en) * | 2014-06-17 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Search method, device and system based on and fusing with user relation data |
| CN104933172A (en) * | 2015-06-30 | 2015-09-23 | 百度在线网络技术（北京）有限公司 | Information pushing method and device based on user searching behavior |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110083677A (en) * | 2019-05-07 | 2019-08-02 | 北京字节跳动网络技术有限公司 | Searching method, device, equipment and the storage medium of contact person |
| CN111028040A (en) * | 2019-06-17 | 2020-04-17 | 拉扎斯网络科技（上海）有限公司 | Object sorting method and device, readable storage medium and electronic equipment |
| CN110850997A (en) * | 2019-09-29 | 2020-02-28 | 上海萌家网络科技有限公司 | Input method and system thereof |
| CN111737418A (en) * | 2020-07-20 | 2020-10-02 | 北京每日优鲜电子商务有限公司 | Method, apparatus and storage medium for predicting relevance of search term and commodity |
| CN111737418B (en) * | 2020-07-20 | 2021-05-14 | 北京每日优鲜电子商务有限公司 | Method, apparatus and storage medium for predicting relevance of search term and commodity |
| CN112102037A (en) * | 2020-09-16 | 2020-12-18 | 汤涛 | Live E-commerce platform commodity content intelligent pushing management system based on big data |
| CN112102037B (en) * | 2020-09-16 | 2021-05-07 | 广州伊的家网络科技有限公司 | Live E-commerce platform commodity content intelligent pushing management system based on big data |
| CN112651839A (en) * | 2021-01-07 | 2021-04-13 | 中国农业银行股份有限公司 | Product optimization method and system |
| WO2022220738A1 (en) * | 2021-04-15 | 2022-10-20 | Grabtaxi Holdings Pte. Ltd. | Communications server apparatus, method and communications system for providing a recommendation of offerings by merchants to a user |
| CN113297511A (en) * | 2021-05-24 | 2021-08-24 | 北京三快在线科技有限公司 | Method and device for constructing merchant ranking model, server and storage medium |
| TWI826881B (en) * | 2021-12-22 | 2023-12-21 | 阿物科技股份有限公司 | System and method for awakening non-shopping consumers |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108614832B (en) | 2022-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108614832A (en) | A kind of user individual commercial articles searching implementation method and device | |
| Lu et al. | BizSeeker: a hybrid semantic recommendation system for personalized government‐to‐business e‐services | |
| Sohail et al. | Feature extraction and analysis of online reviews for the recommendation of books using opinion mining technique | |
| Shinde et al. | Hybrid personalized recommender system using centering-bunching based clustering algorithm | |
| CN106127525A (en) | A kind of TV shopping Method of Commodity Recommendation based on sorting algorithm | |
| US20150379609A1 (en) | Generating recommendations for unfamiliar users by utilizing social side information | |
| CN110175895B (en) | Article recommendation method and device | |
| CN103914492A (en) | Method for query term fusion, method for commodity information publish and method and system for searching | |
| WO2015135110A1 (en) | Systems and methods for keyword suggestion | |
| CN110503497A (en) | A kind of electric business management platform Method of Commodity Recommendation of the consumption habit based on client | |
| CN105023178B (en) | A kind of electronic commerce recommending method based on ontology | |
| Dhillon et al. | Modeling dynamic user interests: A neural matrix factorization approach | |
| Małecki et al. | The classification of internet shop customers based on the cluster analysis and graph cellular automata | |
| CN102831543A (en) | E-commerce recommendation method | |
| CN109190027A (en) | Multi-source recommended method, terminal, server, computer equipment, readable medium | |
| Almaghrabi et al. | A deep learning based collaborative neural network framework for recommender system | |
| CN105205046A (en) | System and method for on-line user recommendation based on semantic analysis | |
| CN112801803B (en) | Financial product recommendation method and device | |
| CN108572981B (en) | Information recommendation method and device | |
| Kim et al. | Recommender systems using SVD with social network information | |
| KR20100070661A (en) | Apparatus and method for offering item recommendation | |
| CN104050174B (en) | A kind of personal page generation method and device | |
| KR101985603B1 (en) | Recommendation method based on tripartite graph | |
| Bar-Gill et al. | Online Exploration, Content Choice & Echo Chambers: An Experiment | |
| Somya et al. | A novel approach to collect and analyze market customer behavior data on online shop |
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 | ||
| CB02 | Change of applicant information |
Address after: 210000, 1-5 story, Jinshan building, 8 Shanxi Road, Nanjing, Jiangsu. Applicant after: SUNING.COM Co.,Ltd. Address before: 210042 Suning Headquarters, No. 1 Suning Avenue, Xuanwu District, Nanjing City, Jiangsu Province Applicant before: SUNING COMMERCE GROUP Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |