CN106327227A - Information recommendation system and information recommendation method - Google Patents
Information recommendation system and information recommendation method Download PDFInfo
- Publication number
- CN106327227A CN106327227A CN201510345416.1A CN201510345416A CN106327227A CN 106327227 A CN106327227 A CN 106327227A CN 201510345416 A CN201510345416 A CN 201510345416A CN 106327227 A CN106327227 A CN 106327227A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- information recommendation
- commodity
- recommendation
- 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.)
- Pending
Links
Abstract
The invention relates to the technical field of the internet, in particular to a recommendation system and a recommendation method for a website provision object. An information recommendation system comprises a data module used for collecting and storing attribute data of users and extracting and analyzing keywords of orders, a triggering module used for generating recommended candidate sets by utilizing a triggering policy, a screening module used for fusing the recommended candidate sets and filtering commodities which do not meet recommendation conditions, a sorting module used for performing priority re-sorting on commodities to be recommended to the users, and a recommendation module used for recommending the sorted commodities to the users. According to the system and the method, commodities related to commodities browsed or purchased by the users can be recommended to the users, so that the diversity of recommendation is greatly enhanced, the commodities purchased by the users are not repeatedly recommended, a recommendation function is optimized, the conversion from recommendation to sale is promoted, and the conversion rate of sale and the usage rate of the users are increased.
Description
[technical field]
The present invention relates to Internet technical field, being specifically related to a kind of website provides commending system and the recommendation method of object.
[background technology]
Information overabundance Internet era, information recommendation technology is self-evident for the importance of Internet firm's operation, current various website all can use commending system on backstage, reach the standard grade " recommendation today " and " guessing that you like " two functions, the most all kinds of shopping websites, music site, video website and all kinds of site for services etc. downloaded of paying, they are the accessing characteristics according to user, calculate the candidate products being best suitable for recommending this user, then these products are presented to user, select for it.Owing to product quantity and number of users are the hugest, commending system uses data mining based on cloud computing in a large number, by digging user behavior and merchandise news, uses various proposed algorithm to calculate recommendation list, realize showing to different user the effect of different content, improve the conversion ratio of the page.
In existing commending system, the most representational one is that like product is recommended, and records and order record, the similar-type products in user's recommendation record or like products according to browsing of user.This recommendation is personalized not enough, it is recommended that the multiformity of product is inadequate, it is impossible to find user's potential demand to other products, because the product interest bought often can be declined by user.It addition, the commending system of major part website is all based on Baidu's plug-in unit in the Internet, it is impossible to be really achieved the purpose that recommendation is turned to operation.
[summary of the invention]
In view of the above problems, it is proposed that the present invention, in order to a kind of a kind of information recommendation system and recommendation method thereof overcoming the problems referred to above or solving the problems referred to above at least in part is provided.
According to one aspect of the present invention, it is provided that a kind of information recommendation system, this system includes:
Data module, for collecting and store the attribute data of user, keyword extraction and the analysis to order；
Trigger module, the Candidate Set recommended for utilizing trigger policy to produce；
Screening module, for merging the candidate collection of recommendation, and filters out the commodity not meeting recommendation condition；
Order module, for carrying out priority rearrangement to the commodity intended to user recommends；
Recommending module, for recommending to user the commodity after sequence.
Further, described data module includes data generating unit and data storage cell, wherein:
Described data generating unit, is to utilize various data processing tools, is carried out original log, removes noise data, be processed into the data of formatting；
Described data storage cell, is to the data formatted, is divided into different types and stores.
Further, described candidate collection from historical behavior based on user, preference based on user and can obtain based on three, the region residing for user dimension, it is also possible to from online relevant, the most similar, off-line is correlated with, and off-line is similar, and each dimension of recent hot-sale products personal style and attitude class obtains.
Further, the historical behavior dimension of described user, relevant or similar commodity can be recommended according to the purchasing behavior before user or navigation patterns.
Further, the preference dimension of described user, can include that user draws a portrait and multi-screen intercommunication, wherein: described user portrait refer to combine Brand, be suitable for crowd, price index and user to the click of commodity, buy, pay close attention to and the behavior such as collection, determine the attribute of user；Described multi-screen intercommunication refers to the synchronization that can realize recommending on various intelligent terminal.
Further, the region dimension residing for described user, is that whole map is divided into multiple grid, and the result of maintenance data statistics realizes recommending the user being in a certain region.
Further, for not having user behavior before or the fewer new user of user behavior before, the region dimension residing for the user described in employing triggers.
Further, described triggering algorithm includes collaborative filtering, based on position, based on inquiry, based on legend, active user behavior and substitute strategy.
Further, described order module, can be from the interactive log of user, by training characteristics weight, then by sequence study (Learning
To Rank, L2R) algorithm improvement realization.
According to a further aspect in the invention, the present invention also provides for a kind of information recommendation method, including:
Collect and store the attribute data of user, keyword extraction and the analysis to order；
Trigger policy is utilized to produce the Candidate Set recommended；
Filter out the commodity not meeting recommendation condition；
The commodity intended to user recommends are carried out priority rearrangement；
Commodity after sequence are recommended to user.
The present invention can recommend, to user, the commodity that commodity that are browsed to it or that bought are relevant, greatly strengthen the multiformity of recommendation, and, the commodity will not bought client repeat to recommend, optimize recommendation function, promote the conversion recommended to selling, improve the conversion ratio of sale and the utilization rate of user.
[accompanying drawing explanation]
Fig. 1 is the information recommendation system schematic diagram of one embodiment of the invention.
Fig. 2 is the information recommendation method schematic diagram of the present invention.
[detailed description of the invention]
Describe in more detail the exemplary embodiment of the present invention below with reference to accompanying drawings.Although accompanying drawing shows the exemplary embodiment of the present invention, it being understood, however, that may be realized in various forms the present invention, and should not limited by embodiments set forth here.On the contrary, it is provided that these embodiments are to enable the more thorough explanation present invention, and complete for the scope of the present invention can be conveyed to those skilled in the art.
Fig. 1 shows a kind of according to an embodiment of the invention information recommendation system.As it is shown in figure 1, this system includes:
Data module 100, for collecting and store the attribute data of user, keyword extraction and the analysis to order.Data are the basic of follow-up modules.As a transaction platform, there is the customer volume of quickly growth simultaneously, therefore the user behavior data that magnanimity is abundant is created, different types of data, it is worth and the power of reflection user view is the most different, accordingly, it would be desirable to data are carried out different extractions and analysis, to facilitate calling of subsequent module.
Data module 100 includes data generating unit and data storage cell, wherein:
Data generating unit is mainly by various data processing toolses, original log is carried out, it is processed into the data of formatting, such as cheating, brush is single, buy on behalf etc. is had a strong impact on to the noise data of subsequent module algorithm effect, first have to reject in data cleansing.Data generating unit includes the extraction to data and analysis.For registration user, the attribute information of registration user, such as sex, age, educational background etc. can be extracted；For registration or nonregistered user, according to the computer caching that the geographical position residing for user and user are recent, by to UGC(User Generated Content, user's original content) excavation of data, some key words can be extracted, then carry out key word analysis, use the key word after analyzing to label to order (deal), generate the data formatted, show personalized order.
Data storage cell is to the data formatted, and is divided into different types and stores, and such as hive data base, hbase data base, mysql data base, redis data base etc. use for trigger module 200.
Trigger module 200, utilizes trigger policy to produce the Candidate Set recommended.The acquisition of Candidate Set, can mainly realize from three dimensions: be historical behavior based on user, preference based on user and based on the region residing for user respectively.Certainly, the acquisition of candidate collection can also consider from other dimensions, is correlated with the most online, the most similar, and off-line is correlated with, and off-line is similar, recent hot-sale products personal style and attitude class etc., may be added in the middle of the dimension that candidate collection obtains.Evaluate certain dimension from recommend to buy transformation efficiency height, mainly see the clicking rate of user under this dimension, conversion ratio and year gross turnover.After candidate collection obtains, candidate collection is carried out efficiency analysis, use different trigger policy to produce the candidate collection recommended according to different dimensions.
For historical behavior dimension based on user, relevant or similar commodity can be recommended according to the purchasing behavior before user or navigation patterns.Further; purchasing behavior before user can be considered as an important demarcation line; when user has bought certain commodity; select to recommend associated therewith or similar commodity with purposes according to type of merchandize; rather than ground repeats to recommend, such as recommend iphone6 plus protection set rather than iphone6 plus for iphone6 plus buyer.In order to improve the transformation efficiency recommending to buy, the commercial product recommending that user can be browsed recently or pay close attention to is to homepage, or shopping cart will be put into but non-purchased goods recommends user, especially these commodity are done activity by businessman when, it is recommended that to associated user.
For preference dimension based on user, it is recommended that can include that user draws a portrait and multi-screen intercommunication.Wherein, user's portrait refer to combine Brand, be suitable for crowd, price index and user to the click of commodity, buy, pay close attention to and the behavior such as collection, user is drawn a portrait, determines the attribute of user, so that it is determined that the category that can recommend for a long time.Such as, according to concern user behavior under the conditions of above-mentioned, user for a frequent purchase imported and famous brand infant product, can substantially determine the sex of user be women, purchasing power consumption level very strong, commodity merchandise classification higher, that pay close attention to be infant product, and then according to the time locus of infant growth, can determine that the follow-up commodity recommended for a long time to it, such as the clothes of age bracket, toy, early education books etc. successively.Multi-screen intercommunication refers to recommend to should apply to different clients, can realize the synchronization recommended on various intelligent terminal (such as notebook, mobile phone, panel computer etc.).
For based on the region dimension residing for user, being that whole map is divided into multiple grid, the result of maintenance data statistics realizes recommending the user being in a certain region.As a example by Beijing, the commodity that the user in East 3rd Ring Road CBD area is interested are mobile phone, fashionable dress, cosmetics.Fangshan long sun area (this district is new city, and major part is newly-built community) can fit up according to youngster, married, raw son, being concentrated mainly on recommendation marry photography, fit up, mother and baby's articles for use etc..Recommendation based on region is mainly used in not having user behavior before or the fewer new user of user behavior before.
For above three dimension, the triggering algorithm related to includes:
(1) collaborative filtering
Collaborative filtering almost can be used in each commending system.Rudimentary algorithm is very simple, but is intended to obtain more preferable effect, generally requires the process doing some differentiation according to concrete business, such as removes the noise datas such as cheating, brush is single, buy on behalf；Rational choice training data, the time window of the training data chosen is unsuitable long, certainly can not be too short, and concrete window phase numerical value needs to determine according to experiment；It is also conceivable to the decay of introducing time, because recent user behavior more can reflect the ensuing behavior act of user.
In data analysis, some algorithm needs to utilize existing data construct model, such as Bayes classifier, decision tree, linear regression etc., and this kind of algorithm is referred to as supervised learning (Supervisied
Learning) algorithm.Typically give a forecast analysis time, can split data into two large divisions:
A part is training data (Train Data), is used for building model.But, the building process of model is sometimes also required to testing model, submodel builds, so training data can be further divided into two parts: 1) training data；2) checking data (Validation
Data), optional, build for submodel, can reuse.Typical example is with K-Fold Cross Validation cutting decision tree, obtains optimum leaf segment and counts, prevents transition matching (Overfitting).
Another part is test data (Test Data), for the structure of testing model.These data only use when model testing, for the accuracy rate of assessment models.Absolutely not it is allowed for model construction process, otherwise can cause transition matching.
(2) location-based (based on position)
For mobile device, one of difference maximum with PC end is that the position of mobile device often changes.Different geographical position reflects different user's scenes, can make full use of the geographical position residing for user in concrete business.In the Candidate Set recommended triggers, also can trigger corresponding strategy according to geographical position such as the real-time geographical locations of user, place of working, residences.Such as according to the history consumption of user, historical viewings etc., excavate the region consumption boom list in the region (such as commercial circle) of a certain granularity and heat list is bought in region, when on new line, user asks to arrive, region consumption boom list and the region in corresponding geographical position are bought heat and are singly weighted by the several geographical position according to user, finally give a recommendation list.
Further, it is also possible to the geographical position occurred according to user, use the mode of collaborative filtering to calculate the similarity of user simultaneously.
(3) query-based(is based on inquiry)
Search is a kind of strong user view, the clearest and the most definite wish having reacted user, but under many circumstances, because various reason, is formed without final conversion.While it is true, this sight still represents certain user intention, can be used.Specific practice is as follows: the search to user's the past period is excavated without conversion behavior, calculates the weight that difference is inquired about by each user.
(4) graph-based(is based on legend)
For collaborative filtering, the map distance between user (user) or between order (deal) is double bounce, then can not take into account for more remote relation.And nomography can break this restriction, the relation of user Yu order regarding as a bigraph (bipartite graph), mutual relation can be propagated on figure.Simrank [2] is a kind of nomography weighing peer-entities similarity.Its basic thought is, if two entities have dependency relation with other similar entities, they are also similar, i.e. similarity can be propagated.
(5) active user behavior
At present the business on the Internet can produce include search for, screen, collect, browse, the user behavior enriched such as place an order, and is the important foundation that can carry out effect optimization.It is of course desirable that each user behavior stream can arrive the link of conversion, but in fact it is far from so.When user creates some behavior of the behavior upstream that places an order, have quite a few because a variety of causes makes behavior stream be formed without converting.But, these upstream behaviors of user are the most important prioris.In the case of a lot, user does not the most convert and does not represent user and lose interest in current project.When user arrives again at recommendation exhibition position, real intention according to the priori behavior understanding user produced before user, the relevant order meeting user view is presented to again user, guides user to flow to downstream along behavior and advance, be finally reached this ultimate aim that places an order.Active user behavior comprises the steps that displaying live view, collects in real time.
(6) substitute strategy
Although there being a series of Candidate Set based on user's historical behavior triggering algorithm, but the user that user new for part or historical behavior less enrich, the Candidate Set of above-mentioned algorithm triggers is the least, it is therefore desirable to use some substitute strategies to be filled with.Including:
Fast-selling single: the weight that sales volume is most within a certain period of time, it may be considered that the impact etc. of time decay.
Favorable comment list: in the evaluation that user produces, higher weight of marking.
City is single: meet basic qualifications, in the incity, request city of user.
Substrategy merges: in order to combine the different advantage triggering algorithm, improves multiformity and the coverage rate of Candidate Set simultaneously, needs by different triggering algorithm fusions together.Adoptable fusion method has following several:
Weighting type: simplest fusion method is exactly that algorithms of different is assigned to different weights based on experience value, and the Candidate Set producing each algorithm is weighted according to given weight, then according still further to weight sequencing.
Classification-type: the algorithm that preferential employing is effective, when the Candidate Set size produced is insufficient for desired value, re-uses the next best algorithm of effect, and the rest may be inferred.
Modulation type: different algorithms produces a certain amount of Candidate Set according to different ratios, and then superposition produces the most total Candidate Set.
Filter-type: the Candidate Set that previous stage algorithm is produced by current algorithm filters, and the rest may be inferred, and Candidate Set is filtered step by step, one small but excellent candidate collection of final generation.
Screening module 300, for merging the candidate collection of recommendation, and filters out the commodity not meeting recommendation condition.Screening module 300 includes integrated unit and filter element, and integrated unit is that the different Candidate Sets producing trigger module 200 merge, to improve coverage and the precision of Generalization bounds；Filter element undertakes certain filtering responsibilities, from the angle-determining of product, operation, some are the most regular, filter out ineligible items, such as, product and the word country forbidden in plain text by the key word system set up effectively are filtered, commodity and identical commodity for not having cooperation carry out screening or manual intervention, do not recommend user.
The commodity intended to user recommends are carried out priority rearrangement by order module 400.The Candidate Set triggering out according to algorithms of different due to trigger module, simply determine according to the historic effect of algorithm that the position of the weight of algorithm generation seems that some is simple and crude, simultaneously, inside at each algorithm, the order of different weights is also simply simply determined by one or several factors, and the method for these sequences is only used for the primary election process of the first step, and final ranking results needs the method by machine learning, use relevant order models, comprehensive many because usually determining.In the present invention, it is recommended that the problem of sequence is converted into the problem of classification and realizes, the screening Candidate Set that screens of module 300 is reordered by the model mainly by machine learning.
Existing order module uses relevancy ranking or importance ranking, and relevancy ranking is to be ranked up document according to the similarity between inquiry and document；Importance ranking does not consider inquiry, and judges the authoritative degree of document only according to the graph structure between webpage.Owing to existing order module the most only considers some aspect (degree of association or importance), so its conversion ratio is the highest.In order to make conversion ratio obtain lifting again, and invent from the interactive log of user, by training characteristics weight, then realized by sequence study (Learning to Rank, L2R) algorithm improvement.Concrete, the L2R algorithm that the present invention uses is different from text classification, it is considered that the sequence of the collection of document of given inquiry, therefore, the feature that L2R algorithm is used not only comprises some features etc. of document itself, also includes the degree of association between document and given inquiry, and the importance (such as PageRank value etc.) that document is over the entire network, therefore, the present invention uses the output of relevance ranking and importance ranking to be used as the feature weight of L2R algorithm.
The training data of L2R algorithm can be obtained by two kinds of methods: artificial mark and excavation from journal file.
Artificial mark: first randomly draw some inquiries from the search record of search engine, multiple different search engine is submitted in these inquiries, then choose each search engine and return the front several of result, finally by professional, these documents are labeled according to the degree of association with inquiry.
Excavate from daily record: search engine has the behavior of substantial amounts of log recording user, can therefrom extract the training data of L2R.A given inquiry, the results list that search engine returns is L, and the collection of the document that user clicks on is combined into C, if a clicked mistake of document " di ", before another one document " dj " does not has a clicked mistake, and " dj " comes di in the results list, then di > dj is exactly a training record.
The concrete L2R algorithm of the present invention uses PointWise, PairWise and ListWise tri-class, wherein:
1). PointWise L2R
PointWise method only considers under given inquiry, the absolute degree of association of single document, and does not consider other documents and the degree of association of given inquiry.That is give a genuine document sequence of inquiry q, it is only necessary to consider degree of correlation ci of single document di and this inquiry,
2). Pairwise L2R
Pairwise method considers under given inquiry, the relative relevance between two documents.That is give a genuine document sequence of inquiry q, it is only necessary to the relative relevance between the document that consideration any two degree of association is different: di>dj, or di<dj.
3). Listwise L2R
Different from Pointwise and Pairwise method, Listwise method directly considers the overall sequence of the collection of document under given inquiry, the document sequence of direct Optimized model output so that it is as close possible to genuine document sequence.
Commodity after sequence are recommended by recommending module 500 to user.
For example, for the data of user's active behavior in data module, have recorded the various actions of user's link different on website platform, the search such as carried out on website, consulting and the record browsed, combine the product of periphery for these contents and carry out the analysis of data, on the one hand these behaviors can be used for the calculated off line in trigger module algorithm, still further aspect, the strong and weak difference of the intention that these behaviors represent, therefore different regressive object values can be set for different behaviors, more carefully to portray the behavior degree of strength of user in order module.Additionally, user is also used as the cross feature of order module to these behaviors of transaction, for off-line training and on-line prediction.
Negative factor evidence in data module, reflect current result and may can not meet the demand of user in some aspects, therefore need to consider specific factor is filtered or dropped power in follow-up Candidate Set trigger process, reduce the probability that negative factor occurs again, improve Consumer's Experience；Simultaneously in the model training of order module, negative factor is according to participating in model training as rare negative example, and these negative examples are more than the sample not clicking on after those displayings, not placing an order.
User's portrait is the basic data portraying user property, some of which is the initial data directly obtained, some is the secondary operations data through excavating, on the one hand these attributes may be used for being weighted order or dropping in Candidate Set trigger process power, and still further aspect can be as the user's dimensional characteristics in order module.
The such scheme of the present invention has the advantage that
1, the present invention can recommend, to user, the commodity that commodity that are browsed to it or that bought are relevant, greatly strengthen the multiformity of Recommendations, and, the commodity will not bought client repeat to recommend, and optimize recommendation function.
2, present invention facilitates the conversion recommended to selling, improve the conversion ratio of sale and the utilization rate of user, can be that network selling provides marketing platform more accurately, improve the income of website operator.
Fig. 2 shows the recommendation method corresponding with Fig. 1 commending system, and due to the systems compliant that its principle is corresponding with Fig. 1, the most too much repeats.A kind of information recommendation method, including:
Collect and store the attribute data of user, keyword extraction and the analysis to order.Concrete, available various data processing toolses, original log is carried out, cancelling noise data, is processed into the data of formatting；Different types of storage is carried out to formatting data.
Trigger policy is utilized to produce the Candidate Set recommended.Can be the most similar from historical behavior based on user, preference based on user, based on the region residing for user and be correlated with online, off-line is correlated with, and off-line is similar, and recent hot-sale products personal style and attitude class etc. obtains candidate collection.Relevant or similar commodity can be recommended according to the purchasing behavior before user or navigation patterns.Further, the purchasing behavior before user can be considered as an important demarcation line, when user has bought certain commodity, select to recommend associated therewith or similar commodity rather than ground to repeat to recommend with purposes according to type of merchandize.User can be drawn a portrait, determine the attribute of user, so that it is determined that the category that can recommend for a long time.Can be divided into multiple grid by whole map, the result of maintenance data statistics realizes recommending the user being in a certain region.
Candidate collection is merged, and filters out the commodity not meeting recommendation condition.Different Candidate Sets are merged, to improve coverage and the precision of Generalization bounds；From the angle-determining of product, operation, some are the most regular, filter out ineligible items.
The commodity intended to user recommends are carried out priority rearrangement.The Candidate Set screened is reordered by the model utilizing machine learning.
Commodity after sequence are recommended to user.
It should be understood that
Algorithm and display are not intrinsic to any certain computer, virtual system or miscellaneous equipment relevant provided herein.Various general-purpose systems can also be used together with based on teaching in this.As described above, construct the structure required by this kind of system to be apparent from.Additionally, the present invention is also not for any specific programming language.It is understood that, it is possible to use various programming languages realize invention described herein content.
Those skilled in the art are appreciated that module each in embodiment can carry out adaptivity changes and they are arranged in one or more equipment different from this embodiment.Unless otherwise being expressly recited, each feature disclosed in this specification can be replaced by the alternative features providing identical, equivalent or similar purpose.
The all parts embodiment of the present invention can realize with hardware, or realizes with the software module run on one or more processor, or realizes with combinations thereof.
The foregoing is only the preferred embodiments of the invention, be not limited to the claims of the present invention.The most described above, for those skilled in the technology concerned it would be appreciated that and implement, therefore other equivalents completed based on disclosed content change, and should be included in the covering scope of the claims.
Claims (10)
1. an information recommendation system, it is characterised in that including:
Data module, for collecting and store the attribute data of user, keyword extraction and the analysis to order；
Trigger module, the Candidate Set recommended for utilizing trigger policy to produce；
Screening module, for merging the candidate collection of recommendation, and filters out the commodity not meeting recommendation condition；
Order module, for carrying out priority rearrangement to the commodity intended to user recommends；
Recommending module, for recommending to user the commodity after sequence.
Information recommendation system the most according to claim 1, it is characterised in that: described data module includes data generating unit and data storage cell, wherein:
Described data generating unit, is to utilize various data processing tools, is carried out original log, removes noise data, be processed into the data of formatting；
Described data storage cell, is to the data formatted, is divided into different types and stores.
Information recommendation system the most according to claim 1, it is characterized in that: described candidate collection from historical behavior based on user, preference based on user and can obtain based on three, the region residing for user dimension, can also be from online relevant, the most similar, off-line is correlated with, off-line is similar, and each dimension of recent hot-sale products personal style and attitude class obtains.
Information recommendation system the most according to claim 3, it is characterised in that: the historical behavior dimension of described user, relevant or similar commodity can be recommended according to the purchasing behavior before user or navigation patterns.
Information recommendation system the most according to claim 3, it is characterized in that: the preference dimension of described user, can include that user draws a portrait and multi-screen intercommunication, wherein: described user portrait refer to combine Brand, be suitable for crowd, price index and user to the click of commodity, buy, pay close attention to and the behavior such as collection, determine the attribute of user；Described multi-screen intercommunication refers to the synchronization that can realize recommending on various intelligent terminal.
Information recommendation system the most according to claim 3, it is characterised in that: the region dimension residing for described user, is that whole map is divided into multiple grid, and the result of maintenance data statistics realizes recommending the user being in a certain region.
Information recommendation system the most according to claim 3, it is characterised in that: for not having user behavior before or the fewer new user of user behavior before, the region dimension residing for the user described in employing triggers.
Information recommendation system the most according to claim 1, it is characterised in that: described triggering algorithm includes collaborative filtering, based on position, based on inquiry, based on legend, active user behavior and substitute strategy.
Information recommendation system the most according to claim 1, it is characterised in that: described order module, can be from the interactive log of user, by training characteristics weight, then realized by sequence study (Learning to Rank, L2R) algorithm improvement.
10. information recommendation method based on information recommendation system described in claim 1, it is characterised in that described information recommendation method includes:
Collect and store the attribute data of user, keyword extraction and the analysis to order；
Trigger policy is utilized to produce the Candidate Set recommended；
Filter out the commodity not meeting recommendation condition；
The commodity intended to user recommends are carried out priority rearrangement；
Commodity after sequence are recommended to user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510345416.1A CN106327227A (en) | 2015-06-19 | 2015-06-19 | Information recommendation system and information recommendation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510345416.1A CN106327227A (en) | 2015-06-19 | 2015-06-19 | Information recommendation system and information recommendation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106327227A true CN106327227A (en) | 2017-01-11 |
Family
ID=57727528
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510345416.1A Pending CN106327227A (en) | 2015-06-19 | 2015-06-19 | Information recommendation system and information recommendation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106327227A (en) |
Cited By (72)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106709076A (en) * | 2017-02-27 | 2017-05-24 | 华南理工大学 | Social network recommendation device and method based on collaborative filtering |
| CN106897899A (en) * | 2017-01-24 | 2017-06-27 | 武汉奇米网络科技有限公司 | A kind of method and system by personalized recommendation commodity after customer grouping |
| CN106991592A (en) * | 2017-03-22 | 2017-07-28 | 南京财经大学 | A kind of personalized recommendation method based on purchase user behavior analysis |
| CN107133836A (en) * | 2017-03-22 | 2017-09-05 | 无锡中科富农物联科技有限公司 | A kind of adaptive weighting combined recommendation algorithm |
| CN107169586A (en) * | 2017-03-29 | 2017-09-15 | 北京百度网讯科技有限公司 | Resource optimization method, device and storage medium based on artificial intelligence |
| CN107194718A (en) * | 2017-05-09 | 2017-09-22 | 上海邮电设计咨询研究院有限公司 | Commercial circle information-pushing method based on location of mobile users and user behavior |
| CN107196999A (en) * | 2017-05-03 | 2017-09-22 | 网易传媒科技（北京）有限公司 | Method and apparatus for issuing information flow propelling data |
| CN107256512A (en) * | 2017-06-08 | 2017-10-17 | 贵州优联博睿科技有限公司 | One kind house-purchase personalized recommendation method and system |
| CN107403359A (en) * | 2017-07-20 | 2017-11-28 | 义乌洞开网络科技有限公司 | A kind of accurate commending system of electric business platform commodity and its method |
| CN107424043A (en) * | 2017-06-15 | 2017-12-01 | 北京三快在线科技有限公司 | A kind of Products Show method and device, electronic equipment |
| CN107481086A (en) * | 2017-06-28 | 2017-12-15 | 北京小度信息科技有限公司 | Information processing method and device |
| CN107562818A (en) * | 2017-08-16 | 2018-01-09 | 中国工商银行股份有限公司 | Information recommendation system and method |
| CN107657506A (en) * | 2017-09-06 | 2018-02-02 | 北京五八到家信息技术有限公司 | The intelligent recommendation method and commending system of a kind of service commodity |
| CN107730372A (en) * | 2017-11-29 | 2018-02-23 | 济南浪潮高新科技投资发展有限公司 | The method and apparatus that a kind of Bank Hall personalized product based on recognition of face is recommended |
| CN107886390A (en) * | 2017-09-30 | 2018-04-06 | 北京小蓦机器人技术有限公司 | Method, equipment, system and the storage medium of the actual demand resource of user are provided |
| CN107908735A (en) * | 2017-11-15 | 2018-04-13 | 北京三快在线科技有限公司 | Information displaying method and device and computing device |
| CN108038647A (en) * | 2017-12-22 | 2018-05-15 | 江苏满运软件科技有限公司 | A kind of source of goods of arterial highway logistics recommends method |
| CN108198007A (en) * | 2018-02-08 | 2018-06-22 | 王四春 | A kind of anti-cross-border electric business commercial affairs big data decision and analysis system divulged a secret |
| CN108256910A (en) * | 2018-01-10 | 2018-07-06 | 京东方科技集团股份有限公司 | Content recommendation method, device and electronic equipment |
| CN108305099A (en) * | 2018-01-18 | 2018-07-20 | 阿里巴巴集团控股有限公司 | Determine the method and device for buying user on behalf |
| CN108376339A (en) * | 2017-09-11 | 2018-08-07 | 加拿大辉莱广告公司 | Advertising specialty recommends method and device |
| CN108429865A (en) * | 2017-02-13 | 2018-08-21 | 中国移动通信集团广东有限公司 | A kind of Products Show processing method and processing device |
| CN108427683A (en) * | 2017-02-14 | 2018-08-21 | 阿里巴巴集团控股有限公司 | A kind of resource put-on method and device |
| CN108573408A (en) * | 2018-04-13 | 2018-09-25 | 浙江工业大学 | A kind of much-sought-after item list formulating method maximizing interests |
| CN108596701A (en) * | 2017-12-29 | 2018-09-28 | 深圳正品创想科技有限公司 | A kind of Method of Commodity Recommendation and device |
| CN108734460A (en) * | 2018-04-02 | 2018-11-02 | 阿里巴巴集团控股有限公司 | A kind of means of payment recommends method, apparatus and equipment |
| CN108804450A (en) * | 2017-04-28 | 2018-11-13 | 百度在线网络技术（北京）有限公司 | The method and apparatus of information push |
| CN108920673A (en) * | 2018-07-09 | 2018-11-30 | 北京兴农丰华科技有限公司 | A kind of recommendation analysis method for corn and wheat breed |
| CN108932643A (en) * | 2017-05-25 | 2018-12-04 | 苏宁云商集团股份有限公司 | A kind of personalized recommendation method and device |
| CN108960945A (en) * | 2017-05-18 | 2018-12-07 | 北京京东尚科信息技术有限公司 | Method of Commodity Recommendation and device |
| CN109257398A (en) * | 2017-07-12 | 2019-01-22 | 阿里巴巴集团控股有限公司 | A kind of method for pushing and equipment of business object |
| CN109409516A (en) * | 2017-08-11 | 2019-03-01 | 微软技术许可有限责任公司 | Machine learning model for the tool depth and width that position is recommended |
| CN109492159A (en) * | 2018-10-31 | 2019-03-19 | 深圳市轱辘汽车维修技术有限公司 | A kind of content delivery method, content push device and electronic equipment |
| CN109658206A (en) * | 2019-01-14 | 2019-04-19 | 京东方科技集团股份有限公司 | Information recommendation method and device |
| CN109688178A (en) * | 2017-10-19 | 2019-04-26 | 阿里巴巴集团控股有限公司 | Recommended method, device and equipment |
| CN109739972A (en) * | 2018-12-27 | 2019-05-10 | 上海连尚网络科技有限公司 | A kind of novel recommended method and equipment |
| CN109886778A (en) * | 2019-01-29 | 2019-06-14 | 上海华程西南国际旅行社有限公司 | The recommended method and system of the tie-in sale product of air ticket |
| CN110019851A (en) * | 2017-09-19 | 2019-07-16 | 香港智能金融科技有限公司 | A kind of the deal maker's teaching auto recommending method and system of personalization |
| CN110060127A (en) * | 2019-04-18 | 2019-07-26 | 南京德通信息科技有限公司 | It is a kind of for medicine ecommerce information management system, method and apparatus |
| CN110060057A (en) * | 2019-04-18 | 2019-07-26 | 南京德通信息科技有限公司 | A kind of health care product based on big data chooses integration platform, method and apparatus |
| CN110163693A (en) * | 2018-01-30 | 2019-08-23 | 哈尔滨学院 | A kind of electric business Method of Commodity Recommendation based on big data |
| CN110209919A (en) * | 2018-02-23 | 2019-09-06 | 北京国双科技有限公司 | The various dimensions method for decomposing and device of attribution |
| CN110223102A (en) * | 2019-05-08 | 2019-09-10 | 五八有限公司 | Lead referral method, apparatus, electronic equipment and storage medium |
| CN110223166A (en) * | 2019-06-14 | 2019-09-10 | 哈尔滨哈银消费金融有限责任公司 | The prediction technique and equipment of consumer finance user's overdue loan based on big data |
| CN110264301A (en) * | 2019-05-10 | 2019-09-20 | 拉扎斯网络科技（上海）有限公司 | Recommended method, device, electronic equipment and non-volatile memory medium |
| CN110727813A (en) * | 2019-08-27 | 2020-01-24 | 达而观信息科技（上海）有限公司 | Self-adaptive hot index sorting method for commodity pictures |
| CN110781397A (en) * | 2019-10-29 | 2020-02-11 | 上海连尚网络科技有限公司 | Method and equipment for providing novel information |
| CN110956511A (en) * | 2019-12-30 | 2020-04-03 | 深圳前海微众银行股份有限公司 | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and computer readable storage medium |
| CN110998619A (en) * | 2017-07-28 | 2020-04-10 | 纽诺有限公司 | System and mechanism for marketing products on autonomous vehicles |
| CN111026966A (en) * | 2019-12-06 | 2020-04-17 | 创新奇智(成都)科技有限公司 | Search recommendation ranking method based on user, product portrait and correlation degree of user and product portrait |
| CN111028065A (en) * | 2019-12-17 | 2020-04-17 | 北京每日优鲜电子商务有限公司 | Information pushing method and device, storage medium and equipment |
| CN111047412A (en) * | 2019-12-16 | 2020-04-21 | 武汉智领云科技有限公司 | Big data electricity merchant operation platform |
| CN111125272A (en) * | 2018-10-31 | 2020-05-08 | 百度在线网络技术（北京）有限公司 | Regional feature acquisition method and device, computer equipment and medium |
| CN111210290A (en) * | 2018-11-21 | 2020-05-29 | 北京京东尚科信息技术有限公司 | Sorting method, sorting device and computer-readable storage medium |
| CN111292170A (en) * | 2020-02-18 | 2020-06-16 | 重庆锐云科技有限公司 | Method, device and storage medium for recommending intention customers for appointed building |
| CN111723273A (en) * | 2019-03-18 | 2020-09-29 | 北京中电翔云信息技术有限公司 | Smart cloud retrieval system and method |
| CN111754300A (en) * | 2020-06-22 | 2020-10-09 | 深圳市分期乐网络科技有限公司 | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and storage medium |
| CN111768218A (en) * | 2019-04-15 | 2020-10-13 | 北京沃东天骏信息技术有限公司 | Method and device for processing user interaction information |
| CN111984837A (en) * | 2019-05-23 | 2020-11-24 | 浙江口碑网络技术有限公司 | Commodity data processing method, device and equipment |
| CN112070552A (en) * | 2020-09-15 | 2020-12-11 | 苏宇航 | Electronic commerce platform based on big data |
| CN112116427A (en) * | 2020-09-22 | 2020-12-22 | 深圳市分期乐网络科技有限公司 | Commodity recommendation method and device, electronic equipment and storage medium |
| CN112232880A (en) * | 2019-12-23 | 2021-01-15 | 北京来也网络科技有限公司 | Commodity recommendation method and device combining RPA and AI |
| CN112288517A (en) * | 2019-12-23 | 2021-01-29 | 北京来也网络科技有限公司 | Commodity recommendation method and device combining RPA and AI |
| CN112381608A (en) * | 2020-11-12 | 2021-02-19 | 盐城工业职业技术学院 | E-commerce commodity recommendation method based on big data multi-label |
| CN112633321A (en) * | 2020-11-26 | 2021-04-09 | 北京瑞友科技股份有限公司 | Artificial intelligence recommendation system and method |
| CN112801761A (en) * | 2021-04-13 | 2021-05-14 | 中智关爱通(南京)信息科技有限公司 | Commodity recommendation method, computing device and computer-readable storage medium |
| CN112989276A (en) * | 2019-12-12 | 2021-06-18 | 北京沃东天骏信息技术有限公司 | Evaluation method and device of information push system |
| CN113379503A (en) * | 2021-06-24 | 2021-09-10 | 北京沃东天骏信息技术有限公司 | Recommendation information display method and device, electronic equipment and computer readable medium |
| CN113706260A (en) * | 2021-09-01 | 2021-11-26 | 镇江纵陌阡横信息科技有限公司 | E-commerce platform commodity recommendation method and device based on search content |
| CN114282106A (en) * | 2021-12-22 | 2022-04-05 | 北京网聘咨询有限公司 | Method for quickly delivering position information |
| TWI784218B (en) * | 2019-12-11 | 2022-11-21 | 中華電信股份有限公司 | Product ranking device and product ranking method |
| CN116934531A (en) * | 2023-07-28 | 2023-10-24 | 重庆安特布鲁精酿啤酒有限公司 | Wine information intelligent management method and system based on data analysis |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102411754A (en) * | 2011-11-29 | 2012-04-11 | 南京大学 | Personalized recommendation method based on commodity property entropy |
| CN102968506A (en) * | 2012-12-14 | 2013-03-13 | 北京理工大学 | Personalized collaborative filtering recommendation method based on extension characteristic vectors |
| CN103914550A (en) * | 2014-04-11 | 2014-07-09 | 百度在线网络技术（北京）有限公司 | Recommended content displaying method and recommended content displaying device |
| CN104572889A (en) * | 2014-12-24 | 2015-04-29 | 深圳市腾讯计算机系统有限公司 | Method, device and system for recommending search terms |
-
2015
- 2015-06-19 CN CN201510345416.1A patent/CN106327227A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102411754A (en) * | 2011-11-29 | 2012-04-11 | 南京大学 | Personalized recommendation method based on commodity property entropy |
| CN102968506A (en) * | 2012-12-14 | 2013-03-13 | 北京理工大学 | Personalized collaborative filtering recommendation method based on extension characteristic vectors |
| CN103914550A (en) * | 2014-04-11 | 2014-07-09 | 百度在线网络技术（北京）有限公司 | Recommended content displaying method and recommended content displaying device |
| CN104572889A (en) * | 2014-12-24 | 2015-04-29 | 深圳市腾讯计算机系统有限公司 | Method, device and system for recommending search terms |
Cited By (90)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106897899A (en) * | 2017-01-24 | 2017-06-27 | 武汉奇米网络科技有限公司 | A kind of method and system by personalized recommendation commodity after customer grouping |
| CN108429865A (en) * | 2017-02-13 | 2018-08-21 | 中国移动通信集团广东有限公司 | A kind of Products Show processing method and processing device |
| CN108429865B (en) * | 2017-02-13 | 2020-10-16 | 中国移动通信集团广东有限公司 | Product recommendation processing method and device |
| CN108427683A (en) * | 2017-02-14 | 2018-08-21 | 阿里巴巴集团控股有限公司 | A kind of resource put-on method and device |
| CN106709076A (en) * | 2017-02-27 | 2017-05-24 | 华南理工大学 | Social network recommendation device and method based on collaborative filtering |
| CN106709076B (en) * | 2017-02-27 | 2023-09-29 | 华南理工大学 | Social network recommendation device and method based on collaborative filtering |
| CN107133836A (en) * | 2017-03-22 | 2017-09-05 | 无锡中科富农物联科技有限公司 | A kind of adaptive weighting combined recommendation algorithm |
| CN106991592A (en) * | 2017-03-22 | 2017-07-28 | 南京财经大学 | A kind of personalized recommendation method based on purchase user behavior analysis |
| CN107169586A (en) * | 2017-03-29 | 2017-09-15 | 北京百度网讯科技有限公司 | Resource optimization method, device and storage medium based on artificial intelligence |
| CN108804450A (en) * | 2017-04-28 | 2018-11-13 | 百度在线网络技术（北京）有限公司 | The method and apparatus of information push |
| CN108804450B (en) * | 2017-04-28 | 2022-02-01 | 百度在线网络技术（北京）有限公司 | Information pushing method and device |
| CN107196999A (en) * | 2017-05-03 | 2017-09-22 | 网易传媒科技（北京）有限公司 | Method and apparatus for issuing information flow propelling data |
| CN107196999B (en) * | 2017-05-03 | 2020-01-24 | 网易传媒科技（北京）有限公司 | Method and equipment for transmitting information flow push data |
| CN107194718A (en) * | 2017-05-09 | 2017-09-22 | 上海邮电设计咨询研究院有限公司 | Commercial circle information-pushing method based on location of mobile users and user behavior |
| CN108960945A (en) * | 2017-05-18 | 2018-12-07 | 北京京东尚科信息技术有限公司 | Method of Commodity Recommendation and device |
| CN108932643A (en) * | 2017-05-25 | 2018-12-04 | 苏宁云商集团股份有限公司 | A kind of personalized recommendation method and device |
| CN107256512A (en) * | 2017-06-08 | 2017-10-17 | 贵州优联博睿科技有限公司 | One kind house-purchase personalized recommendation method and system |
| CN107424043B (en) * | 2017-06-15 | 2021-06-25 | 北京三快在线科技有限公司 | Product recommendation method and device and electronic equipment |
| CN107424043A (en) * | 2017-06-15 | 2017-12-01 | 北京三快在线科技有限公司 | A kind of Products Show method and device, electronic equipment |
| CN107481086A (en) * | 2017-06-28 | 2017-12-15 | 北京小度信息科技有限公司 | Information processing method and device |
| CN109257398A (en) * | 2017-07-12 | 2019-01-22 | 阿里巴巴集团控股有限公司 | A kind of method for pushing and equipment of business object |
| CN107403359A (en) * | 2017-07-20 | 2017-11-28 | 义乌洞开网络科技有限公司 | A kind of accurate commending system of electric business platform commodity and its method |
| US11830058B2 (en) | 2017-07-28 | 2023-11-28 | Nuro, Inc. | Automated retail store on autonomous or semi-autonomous vehicle |
| CN110998619A (en) * | 2017-07-28 | 2020-04-10 | 纽诺有限公司 | System and mechanism for marketing products on autonomous vehicles |
| CN110998619B (en) * | 2017-07-28 | 2023-10-24 | 纽诺有限公司 | System and mechanism for adding and selling products on autonomous carriers |
| CN109409516A (en) * | 2017-08-11 | 2019-03-01 | 微软技术许可有限责任公司 | Machine learning model for the tool depth and width that position is recommended |
| CN107562818A (en) * | 2017-08-16 | 2018-01-09 | 中国工商银行股份有限公司 | Information recommendation system and method |
| CN107657506A (en) * | 2017-09-06 | 2018-02-02 | 北京五八到家信息技术有限公司 | The intelligent recommendation method and commending system of a kind of service commodity |
| CN108376339A (en) * | 2017-09-11 | 2018-08-07 | 加拿大辉莱广告公司 | Advertising specialty recommends method and device |
| CN110019851A (en) * | 2017-09-19 | 2019-07-16 | 香港智能金融科技有限公司 | A kind of the deal maker's teaching auto recommending method and system of personalization |
| CN107886390A (en) * | 2017-09-30 | 2018-04-06 | 北京小蓦机器人技术有限公司 | Method, equipment, system and the storage medium of the actual demand resource of user are provided |
| CN107886390B (en) * | 2017-09-30 | 2019-09-06 | 北京小蓦机器人技术有限公司 | Method, equipment, system and the storage medium of the actual demand resource of user are provided |
| CN109688178A (en) * | 2017-10-19 | 2019-04-26 | 阿里巴巴集团控股有限公司 | Recommended method, device and equipment |
| CN107908735A (en) * | 2017-11-15 | 2018-04-13 | 北京三快在线科技有限公司 | Information displaying method and device and computing device |
| CN107730372A (en) * | 2017-11-29 | 2018-02-23 | 济南浪潮高新科技投资发展有限公司 | The method and apparatus that a kind of Bank Hall personalized product based on recognition of face is recommended |
| CN108038647A (en) * | 2017-12-22 | 2018-05-15 | 江苏满运软件科技有限公司 | A kind of source of goods of arterial highway logistics recommends method |
| CN108596701A (en) * | 2017-12-29 | 2018-09-28 | 深圳正品创想科技有限公司 | A kind of Method of Commodity Recommendation and device |
| CN108256910A (en) * | 2018-01-10 | 2018-07-06 | 京东方科技集团股份有限公司 | Content recommendation method, device and electronic equipment |
| CN108305099B (en) * | 2018-01-18 | 2021-11-19 | 创新先进技术有限公司 | Method and device for determining purchasing user |
| CN108305099A (en) * | 2018-01-18 | 2018-07-20 | 阿里巴巴集团控股有限公司 | Determine the method and device for buying user on behalf |
| CN110163693A (en) * | 2018-01-30 | 2019-08-23 | 哈尔滨学院 | A kind of electric business Method of Commodity Recommendation based on big data |
| CN108198007A (en) * | 2018-02-08 | 2018-06-22 | 王四春 | A kind of anti-cross-border electric business commercial affairs big data decision and analysis system divulged a secret |
| CN110209919A (en) * | 2018-02-23 | 2019-09-06 | 北京国双科技有限公司 | The various dimensions method for decomposing and device of attribution |
| CN108734460A (en) * | 2018-04-02 | 2018-11-02 | 阿里巴巴集团控股有限公司 | A kind of means of payment recommends method, apparatus and equipment |
| WO2019192261A1 (en) * | 2018-04-02 | 2019-10-10 | 阿里巴巴集团控股有限公司 | Payment mode recommendation method and device and equipment |
| CN108573408B (en) * | 2018-04-13 | 2021-07-27 | 浙江工业大学 | Popular commodity list making method for maximizing benefits |
| CN108573408A (en) * | 2018-04-13 | 2018-09-25 | 浙江工业大学 | A kind of much-sought-after item list formulating method maximizing interests |
| CN108920673B (en) * | 2018-07-09 | 2023-11-17 | 北京兴农丰华科技有限公司 | Recommendation analysis method for corn and wheat varieties |
| CN108920673A (en) * | 2018-07-09 | 2018-11-30 | 北京兴农丰华科技有限公司 | A kind of recommendation analysis method for corn and wheat breed |
| CN111125272A (en) * | 2018-10-31 | 2020-05-08 | 百度在线网络技术（北京）有限公司 | Regional feature acquisition method and device, computer equipment and medium |
| CN109492159A (en) * | 2018-10-31 | 2019-03-19 | 深圳市轱辘汽车维修技术有限公司 | A kind of content delivery method, content push device and electronic equipment |
| CN111125272B (en) * | 2018-10-31 | 2023-07-25 | 百度在线网络技术（北京）有限公司 | Regional characteristic acquisition method, regional characteristic acquisition device, computer equipment and medium |
| CN111210290A (en) * | 2018-11-21 | 2020-05-29 | 北京京东尚科信息技术有限公司 | Sorting method, sorting device and computer-readable storage medium |
| CN109739972A (en) * | 2018-12-27 | 2019-05-10 | 上海连尚网络科技有限公司 | A kind of novel recommended method and equipment |
| CN109658206A (en) * | 2019-01-14 | 2019-04-19 | 京东方科技集团股份有限公司 | Information recommendation method and device |
| CN109886778A (en) * | 2019-01-29 | 2019-06-14 | 上海华程西南国际旅行社有限公司 | The recommended method and system of the tie-in sale product of air ticket |
| CN111723273A (en) * | 2019-03-18 | 2020-09-29 | 北京中电翔云信息技术有限公司 | Smart cloud retrieval system and method |
| CN111768218A (en) * | 2019-04-15 | 2020-10-13 | 北京沃东天骏信息技术有限公司 | Method and device for processing user interaction information |
| CN110060127A (en) * | 2019-04-18 | 2019-07-26 | 南京德通信息科技有限公司 | It is a kind of for medicine ecommerce information management system, method and apparatus |
| CN110060057A (en) * | 2019-04-18 | 2019-07-26 | 南京德通信息科技有限公司 | A kind of health care product based on big data chooses integration platform, method and apparatus |
| CN110223102A (en) * | 2019-05-08 | 2019-09-10 | 五八有限公司 | Lead referral method, apparatus, electronic equipment and storage medium |
| CN110264301A (en) * | 2019-05-10 | 2019-09-20 | 拉扎斯网络科技（上海）有限公司 | Recommended method, device, electronic equipment and non-volatile memory medium |
| CN111984837A (en) * | 2019-05-23 | 2020-11-24 | 浙江口碑网络技术有限公司 | Commodity data processing method, device and equipment |
| CN111984837B (en) * | 2019-05-23 | 2021-04-23 | 浙江口碑网络技术有限公司 | Commodity data processing method, device and equipment |
| CN110223166A (en) * | 2019-06-14 | 2019-09-10 | 哈尔滨哈银消费金融有限责任公司 | The prediction technique and equipment of consumer finance user's overdue loan based on big data |
| CN110727813B (en) * | 2019-08-27 | 2023-06-09 | 达而观信息科技（上海）有限公司 | Self-adaptive trending index ordering method for commodity pictures |
| CN110727813A (en) * | 2019-08-27 | 2020-01-24 | 达而观信息科技（上海）有限公司 | Self-adaptive hot index sorting method for commodity pictures |
| CN110781397A (en) * | 2019-10-29 | 2020-02-11 | 上海连尚网络科技有限公司 | Method and equipment for providing novel information |
| CN110781397B (en) * | 2019-10-29 | 2023-04-07 | 上海连尚网络科技有限公司 | Method and equipment for providing novel information |
| CN111026966B (en) * | 2019-12-06 | 2023-12-22 | 创新奇智(成都)科技有限公司 | Search recommendation ordering method based on user and product portrait and association degree of user and product portrait |
| CN111026966A (en) * | 2019-12-06 | 2020-04-17 | 创新奇智(成都)科技有限公司 | Search recommendation ranking method based on user, product portrait and correlation degree of user and product portrait |
| TWI784218B (en) * | 2019-12-11 | 2022-11-21 | 中華電信股份有限公司 | Product ranking device and product ranking method |
| CN112989276A (en) * | 2019-12-12 | 2021-06-18 | 北京沃东天骏信息技术有限公司 | Evaluation method and device of information push system |
| CN111047412A (en) * | 2019-12-16 | 2020-04-21 | 武汉智领云科技有限公司 | Big data electricity merchant operation platform |
| CN111028065A (en) * | 2019-12-17 | 2020-04-17 | 北京每日优鲜电子商务有限公司 | Information pushing method and device, storage medium and equipment |
| CN112288517A (en) * | 2019-12-23 | 2021-01-29 | 北京来也网络科技有限公司 | Commodity recommendation method and device combining RPA and AI |
| CN112232880A (en) * | 2019-12-23 | 2021-01-15 | 北京来也网络科技有限公司 | Commodity recommendation method and device combining RPA and AI |
| CN112232880B (en) * | 2019-12-23 | 2024-04-09 | 北京来也网络科技有限公司 | Goods recommendation method and device combining RPA and AI |
| CN110956511A (en) * | 2019-12-30 | 2020-04-03 | 深圳前海微众银行股份有限公司 | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and computer readable storage medium |
| CN111292170A (en) * | 2020-02-18 | 2020-06-16 | 重庆锐云科技有限公司 | Method, device and storage medium for recommending intention customers for appointed building |
| CN111754300A (en) * | 2020-06-22 | 2020-10-09 | 深圳市分期乐网络科技有限公司 | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and storage medium |
| CN112070552A (en) * | 2020-09-15 | 2020-12-11 | 苏宇航 | Electronic commerce platform based on big data |
| CN112116427A (en) * | 2020-09-22 | 2020-12-22 | 深圳市分期乐网络科技有限公司 | Commodity recommendation method and device, electronic equipment and storage medium |
| CN112381608A (en) * | 2020-11-12 | 2021-02-19 | 盐城工业职业技术学院 | E-commerce commodity recommendation method based on big data multi-label |
| CN112633321A (en) * | 2020-11-26 | 2021-04-09 | 北京瑞友科技股份有限公司 | Artificial intelligence recommendation system and method |
| CN112801761A (en) * | 2021-04-13 | 2021-05-14 | 中智关爱通(南京)信息科技有限公司 | Commodity recommendation method, computing device and computer-readable storage medium |
| CN113379503A (en) * | 2021-06-24 | 2021-09-10 | 北京沃东天骏信息技术有限公司 | Recommendation information display method and device, electronic equipment and computer readable medium |
| CN113706260A (en) * | 2021-09-01 | 2021-11-26 | 镇江纵陌阡横信息科技有限公司 | E-commerce platform commodity recommendation method and device based on search content |
| CN114282106A (en) * | 2021-12-22 | 2022-04-05 | 北京网聘咨询有限公司 | Method for quickly delivering position information |
| CN116934531A (en) * | 2023-07-28 | 2023-10-24 | 重庆安特布鲁精酿啤酒有限公司 | Wine information intelligent management method and system based on data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106327227A (en) | Information recommendation system and information recommendation method | |
| US11354584B2 (en) | Systems and methods for trend aware self-correcting entity relationship extraction | |
| US8346782B2 (en) | Method and system of information matching in electronic commerce website | |
| CN102236663B (en) | Query method, query system and query device based on vertical search | |
| CN103164804B (en) | The information-pushing method of a kind of personalization and device | |
| KR101419504B1 (en) | System and method providing a suited shopping information by analyzing the propensity of an user | |
| CN102629360B (en) | A kind of effective dynamic commodity recommend method and commercial product recommending system | |
| CN108885624B (en) | Information recommendation system and method | |
| CN108205766A (en) | Information-pushing method, apparatus and system | |
| CN103970850B (en) | Site information recommends method and system | |
| CN105718184A (en) | Data processing method and apparatus | |
| CN103914492A (en) | Method for query term fusion, method for commodity information publish and method and system for searching | |
| CN104239338A (en) | Information recommendation method and information recommendation device | |
| CN111709810A (en) | Object recommendation method and device based on recommendation model | |
| CN1985255A (en) | Target advertising method and system using secondary keywords having relation to first internet searching keywords, and method and system for providing a list of the secondary keywords | |
| CN104679771A (en) | Individual data searching method and device | |
| Lee et al. | Fundamentals of big data network analysis for research and industry | |
| JP2014209376A (en) | Application of advertisement source and key word set in online commercial transaction platform | |
| KR102547491B1 (en) | System for providing artificial intelligence based price decision service for business-to-business market | |
| CN104239335B (en) | User-specific information acquisition methods and device | |
| CN112150227A (en) | Commodity recommendation method, system, device and medium | |
| CN106326318A (en) | Search method and device | |
| Dias et al. | Automating the extraction of static content and dynamic behaviour from e-commerce websites | |
| CN113157752A (en) | Scientific and technological resource recommendation method and system based on user portrait and situation | |
| CN109190027A (en) | Multi-source recommended method, terminal, server, computer equipment, readable medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20170704 Address after: 100195 110, building 1, building 18, apricot Road, Haidian District, Beijing Applicant after: Aisino Corporation Inc. Address before: 100195, Haidian District, Beijing, Yan Shi Road, No. 18, 1, 2 Applicant before: BEIJING AEROSPACE ONLINE NETWORK TECHNOLOGY CO., LTD. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170111 |