CN110008396A - Object information method for pushing, device, equipment and computer readable storage medium - Google Patents
Object information method for pushing, device, equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN110008396A CN110008396A CN201811437429.1A CN201811437429A CN110008396A CN 110008396 A CN110008396 A CN 110008396A CN 201811437429 A CN201811437429 A CN 201811437429A CN 110008396 A CN110008396 A CN 110008396A
- Authority
- CN
- China
- Prior art keywords
- text
- active user
- similarity
- user
- information
- 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
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000004458 analytical method Methods 0.000 claims description 29
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000005516 engineering process Methods 0.000 description 18
- 238000010586 diagram Methods 0.000 description 12
- 238000004422 calculation algorithm Methods 0.000 description 11
- 239000011159 matrix material Substances 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 230000006854 communication Effects 0.000 description 6
- 230000006399 behavior Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 239000013598 vector Substances 0.000 description 4
- 239000000047 product Substances 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 239000013065 commercial product Substances 0.000 description 1
- 239000002537 cosmetic Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0255—Targeted advertisements based on user history
- G06Q30/0256—User search
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
The embodiment of the present disclosure provides object information method for pushing, device, equipment and computer readable storage medium.Object information method for pushing includes: that the similarity of search and the first text is greater than the second text of default similarity threshold in the text provided by the user other than active user according to the first text provided for active user；The object relevant to the second text actually obtained in the user's history of search the second text of offer；According to the similarity of the second text and the first text, first object object is selected from the object relevant to the second text searched out；The information of first object object is pushed to active user, it can carry out being associated with for text similarity and target object, the corresponding object type of text is searched to replace, so as to guarantee precision, multi-class, expert's grade target object recognition and information push when carrying out the push of object information in the text provided for user.
Description
Technical field
The embodiment of the present disclosure is related to field of computer technology more particularly to object information method for pushing, device, equipment and meter
Calculation machine readable storage medium storing program for executing.
Background technique
On network, the text (for example, user puts question to) provided for user, many platforms can lead in the database
It crosses similarity search and determines that user puts question to targeted object, then push the information of the object to user.For example, common
On network platform system, information push is carried out to user as follows:
1. the problem of couple user, in platform knowledge library similarity search, finds out the most similar problem in knowledge base.
2. the correspondence answer of Similar Problems the most in this knowledge base is found out, object belonging to the determination information to be pushed
Classification proposed algorithm corresponding with its.
3. exporting corresponding classification according to the browsing information of the user, acquisition information and popular object by proposed algorithm
Object.
4. result some objects in the top will be exported to select, the information of these objects is pushed to user.
In the example of the relevant technologies, on shopping at network platform, existing intelligent customer service can with user session mistake
Semantic analysis and part-of-speech tagging are carried out in journey, merchandise classification are determined by the similarity search with Intelligence repository, further according to visitor
The information (such as the commodity bought and browsed in the past) at family calls commercial product recommending system algorithm to carry out more fuzzy commodity letter
Breath push.
However, in the related technology, it is relatively difficult, example that the text provided according to user, which carries out accurately information push,
Such as, shopping platform progress cosmetics recommend the recommendation of this expert level with regard to relatively difficult.Especially, when platform does not have any use
When the information of family, it is even more extremely difficult to make accurately information push.
Summary of the invention
In view of this, disclosure first aspect provides a kind of object information method for pushing, comprising:
According to the first text provided for active user in the text provided by the user other than the active user
The similarity of search and first text is greater than the second text of default similarity threshold；
The object relevant to second text actually obtained in the user's history of search offer second text；
It is relevant to second text from what is searched out according to the similarity of second text and first text
First object object is selected in object；
The information of the first object object is pushed to the active user.
Disclosure second aspect provides a kind of object information driving means, comprising:
First search module, be configured as according to for active user provide the first text by the active user with
The similarity of search and first text is greater than the second text of default similarity threshold in the text that outer user provides；
Second search module, be configured to search for actually obtaining in the user's history that second text is provided with it is described
The relevant object of second text；
Selecting module is configured as the similarity according to second text and first text, from search out with
First object object is selected in the relevant object of second text；
Pushing module is configured as the information of the first object object being pushed to the active user.
The disclosure third aspect provides a kind of electronic equipment, including memory and processor；Wherein, the memory is used
In storing one or more computer instruction, wherein one or more computer instruction is executed by the processor with reality
Existing following steps:
According to the first text provided for active user in the text provided by the user other than the active user
The similarity of search and first text is greater than the second text of default similarity threshold；
The object relevant to second text actually obtained in the user's history of search offer second text；
It is relevant to second text from what is searched out according to the similarity of second text and first text
First object object is selected in object；
The information of the first object object is pushed to the active user.
Disclosure fourth aspect provides a kind of computer readable storage medium, is stored thereon with computer instruction, the meter
Method as described in relation to the first aspect is realized in the instruction of calculation machine when being executed by processor.
In disclosure embodiment, the first text of active user's offer is directed to by the active user by basis
The similarity of search and first text is greater than the second text of default similarity threshold in the text that user in addition provides；
The object relevant to second text actually obtained in the user's history of search offer second text；According to described
The similarity of two texts and first text selects first object from the object relevant to second text searched out
Object；The information of the first object object is pushed to the active user, text similarity and target object can be carried out
Association, the corresponding object type of text is searched to replace, so as to carry out object information in the text provided for user
Push when guarantee precision, the target object recognition of multi-class, expert grade and information push, especially in no any user
When information, still the information of the desired object of user accurately can be pushed to user.
These aspects or other aspects of the disclosure can more straightforwards in the following description.
Detailed description of the invention
Technical solution in order to illustrate more clearly of the embodiment of the present disclosure or in the related technology, below will be to exemplary implementation
Attached drawing needed in example or description of Related Art is briefly described, it should be apparent that, the accompanying drawings in the following description
It is some exemplary embodiments of the disclosure, for those of ordinary skill in the art, before not making the creative labor
It puts, is also possible to obtain other drawings based on these drawings.
Fig. 1 shows the flow chart of the object information method for pushing according to one embodiment of the disclosure；
Fig. 2 shows the flow charts according to the object information method for pushing of another embodiment of the disclosure；
Fig. 3 shows the flow chart of the object information method for pushing according to the another embodiment of the disclosure；
Fig. 4 show the step S103 in the object information method for pushing according to one embodiment of the disclosure one is exemplary
Flow chart；
Fig. 5 shows the flow chart of the object information method for pushing according to the another embodiment of the disclosure；
Fig. 6 shows the flow chart of the object information method for pushing according to the another embodiment of the disclosure；
Fig. 7 shows the schematic diagram of the Application Scenarios-Example of the object information method for pushing according to one embodiment of the disclosure；
Fig. 8 shows the structural block diagram of the object information driving means according to another embodiment of the disclosure；
Fig. 9 shows the structural block diagram of the electronic equipment according to one embodiment of the disclosure；
Figure 10 is adapted for the computer system for realizing the object information method for pushing according to one embodiment of the disclosure
Structural schematic diagram.
Specific embodiment
In order to make those skilled in the art more fully understand disclosure scheme, below in conjunction with the exemplary implementation of the disclosure
Attached drawing in example, is clearly and completely described the technical solution in disclosure exemplary embodiment.
In some processes of the description in the specification and claims of the disclosure and above-mentioned attached drawing, contain according to
Multiple operations that particular order occurs, but it should be clearly understood that these operations can not be what appears in this article suitable according to its
Sequence is executed or is executed parallel, and serial number of operation such as 101,102 etc. is only used for distinguishing each different operation, serial number
It itself does not represent and any executes sequence.In addition, these processes may include more or fewer operations, and these operations can
To execute or execute parallel in order.It should be noted that the description such as " first " herein, " second ", is for distinguishing not
Same message, equipment, module etc., does not represent sequencing, does not also limit " first " and " second " and be different type.
Below in conjunction with the attached drawing in disclosure exemplary embodiment, to the technical solution in disclosure exemplary embodiment
It being clearly and completely described, it is clear that described exemplary embodiment is only disclosure a part of the embodiment, rather than
Whole embodiments.Based on the embodiment in the disclosure, those skilled in the art institute without creative efforts
The every other embodiment obtained belongs to the range of disclosure protection..
Fig. 1 shows the flow chart of the object information method for pushing according to one embodiment of the disclosure.This method may include
Step S101, S102 and S103 and S104.
In step s101, it is provided according to the first text provided for active user by the user other than active user
Text in search be greater than the second text of default similarity threshold with the similarity of the first text.
In step s 102, it is relevant to the second text right actually to obtain in the user's history of the second text of search offer
As.
In step s 103, related to the second text from what is searched out according to the similarity of the second text and the first text
Object in select first object object.
In step S104, the information of first object object is pushed to active user.
In one embodiment of the present disclosure, the first text that active user provides can be active user in a certain platform
The problem of proposition.The text that user other than active user provides can be the user other than active user and propose on the platform
The problem of.In embodiment of the disclosure, by finding second text most like with the first text (the problem of active user)
After (the problem of other users), lookup be second text associated user, rather than the second text (problem) is targeted answers
Case.First object object relevant to the second text is obtained by the behavior of associated user (for example, associated user's actual purchase
The product crossed).Therefore, the information of first object object relevant to the second text can be pushed to active user.
In one embodiment of the present disclosure, second text (its most like with the first text (the problem of active user)
The problem of his user) may have it is multiple, it is thus possible to can exist it is multiple the user of second text is provided, and then may also exist
Multiple first objects relevant to the second text are to object.Therefore, can by from multiple second from different other users
The relevant multiple first objects of text are pushed to active user to the information of object.
It in one embodiment of the present disclosure, can be with latent semantic analysis (Latent Semantics
Indexing) ", " doc2vec " also " cosine similarity (Cosine Similarity) " three kinds of basic technologies determine second
The similarity of text and the first text.
In one embodiment of the present disclosure, each word can be calculated by latent semantic analysis to tie up in text/dictionary
TF-IDF (the inverse literary frequency of word frequency-, Term Frequency-Inverse Document Frequency) value of degree, and pass through
Singular value decomposition carries out dimensionality reduction, word-document matrix after obtaining dimensionality reduction.By the cosine phase for calculating new text and each column of matrix
Like degree, most like text is found.
In one embodiment of the present disclosure, in doc2vec technology, text can be generated by a shallow-layer neural network
This algorithm.Its network generated contains the matrix for having text information, can be used for calculating text similarity.
In one embodiment of the present disclosure, cosine similarity (Cosine Similarity) is referred to by calculating two
The cosine angle of a text vector obtains the similarity of two text vectors.It is higher to be worth bigger similarity.At one of the disclosure
In embodiment, the similarity of the second text and first text is cosine similarity.
In one embodiment of the present disclosure, it determines that the similarity of long text can use latent semantic analysis technology, determines short
The similarity of text can use word2vec technology, and the text for finally finding out active user is similar to the cosine of other users text
Degree.For example, the text of preceding 1% other users most like with the first text can be found out as the second text.Distinguish text
Length is primarily due to the connection that short text relys more on context, and long text relies more on the context near text.Herein,
Long text/short text judges that this text size analysis model can by one based on the text size analysis model of accuracy rate
To be trained by related art method, details are not described herein.
Above is only to latent semantic analysis (Latent Semantics Indexing) ", " doc2vec " also " cosine
The brief introduction of three kinds of technologies of similarity (Cosine Similarity) ", it will be understood by those skilled in the art that can be from related skill
The details of three of the above technology is obtained in art, details are not described herein for particular content.
In disclosure embodiment, the first text of active user's offer is directed to other than by active user by basis
The text that provides of user in search be greater than the second text of default similarity threshold with the similarity of the first text；Search provides
The object relevant to the second text actually obtained in the user's history of second text；According to the phase of the second text and the first text
Like degree, first object object is selected from the object relevant to the second text searched out；The information of first object object is pushed away
Active user is given, being associated with for text similarity and target object can be carried out, searches the corresponding object type of text to replace,
So as to guarantee precision, multi-class, expert's grade mesh when carrying out the push of object information in the text provided for user
Object Selection and information push are marked, especially when no any user information, still can accurately it is expected user
The information of object be pushed to user.
It is described further referring to Fig. 2 to according to the object information method for pushing of another embodiment of the disclosure.
Fig. 2 shows the flow charts according to the object information method for pushing of another embodiment of the disclosure.As shown in Fig. 2, with
Embodiment shown in FIG. 1 the difference is that, further include step S201 and S202 before step S101.
In step s 201, determine that the first text that active user provides is short text of the length less than pre-set length threshold
Or length is more than or equal to the long text of pre-set length threshold.
It is the determination of long text or short text as a result, by different similar according to the first text in step S202
Spend the similarity that calculation calculates the text that the user other than the first text and active user provides.
In one embodiment according to the disclosure, the text size analysis model above-mentioned based on accuracy rate can be passed through
To judge that the first text is long text or short text.For example, it is then short text, text that text size, which is less than pre-set length threshold,
It is then long text that length, which is more than or equal to pre-set length threshold,.
In one embodiment according to the disclosure, when determining the first text is long text, aforementioned latent language can be passed through
Adopted analysis method calculates the similarity of the text that the user other than the first text and active user provides.When determining the first text
When being short text, the text that the user other than the first text and active user provides can be calculated by aforementioned doc2vec method
This similarity.By using different similarity calculation modes for long text and short text, the can be more accurately calculated
The similarity for the text that one text is provided with the user other than active user.
In one embodiment according to the disclosure, step S201 includes: true according to preset text size analysis model
The first text for determining active user's offer is that length is less than the short text of pre-set length threshold or length is more than or equal to default length
Spend the long text of threshold value.As previously mentioned, long text/short text can pass through a text size analysis model based on accuracy rate
Judge, this text size analysis model can be trained by related art method, details are not described herein.Word2vec is calculated
Method and the matrix generating algorithm of latent semantic analysis can be based on the automatic off-line training of accuracy.
It is described further referring to Fig. 3 to according to the object information method for pushing of the another embodiment of the disclosure.
Fig. 3 shows the flow chart of the object information method for pushing according to the another embodiment of the disclosure.Implementation shown in Fig. 3
The difference of mode and embodiment shown in FIG. 1 is to further include step S301.
In step S301, multiple texts that active user is provided synthesize the first text.
In one embodiment according to the disclosure, if active user provides multiple texts, for example, proposing multiple
Problem, each problem may relate to different target objects, then multiple texts can be synthesized the first text.This is because
Embodiment of the present disclosure is not necessarily to the corresponding answer of each problem search for active user, and it is similar to can be based on text
Spend find other users proposition similar second text (Similar Problems), thus come from other it is practical it is obtaining with second text
Selection target object in this relevant object.
As previously mentioned, after the multiple texts for providing active user synthesize the first text, it is (current to use with the first text
The problem of family) most like the second text (the problem of other users) may have it is multiple, it is thus possible to can exist and multiple provide the
The user of two texts, and then may also have multiple first objects relevant to the second text to object.It therefore, can will be with
The relevant multiple first objects of multiple the second texts from different other users are pushed to active user to the information of object.
The step S103 in the object information method for pushing according to the another embodiment of the disclosure is carried out referring to Fig. 4
It further describes.
Fig. 4 show the step S103 in the object information method for pushing according to one embodiment of the disclosure one is exemplary
Flow chart.As shown in figure 4, step S103 includes step S401 and S402.
In step S401, according to the size of the similarity of the second text and the first text, to searching out and second text
This relevant object is ranked up.
In step S402, the first mesh is selected from the object relevant to the second text searched out according to the result of sequence
Mark object.
In one embodiment according to the disclosure, can find out with the first text similarity it is highest it is preceding it is several (for example,
Preceding 1%, preceding 10% etc.) text of other users is as the second text.According to sequence as a result, from searching out and second text
First object object is selected in relevant object, that is, which and the second text the user of the second text of offer specifically obtain
Relevant object selects first object object from these objects.In the information for the first object object that will be selected in this way to working as
When preceding user pushes, precision, multi-class, expert's grade object information push may be implemented.
It is described further referring to Fig. 5 to according to the object information method for pushing of the another embodiment of the disclosure.
Fig. 5 shows the flow chart of the object information method for pushing according to the another embodiment of the disclosure.As shown in figure 5, with
The difference of embodiment shown in FIG. 1 is to further include step S501, and step S101 includes step S502.
In step S501, the high degree of association is filtered out from the user other than active user according to preset degree of association condition
User.
In step S502, according to the first text provided for active user in the text provided by high degree of association user
The similarity of middle search and the first text is greater than the second text of default similarity threshold.
In one embodiment according to the disclosure, high degree of association user also refers to that there are special passes with active user
The user of system, for example, good friend of the active user on platform can be used as high degree of association user, active user is good in reality
Friend can also be used as high degree of association user.Degree of association condition can be set in various ways, for example, active user and other users
Connection number, active user to other users be provided with specific markers, active user and other users participated in jointly it is a certain
Behavior etc..In one embodiment according to the disclosure, search and the first text in the text provided by high degree of association user
The second text that this similarity is greater than default similarity threshold can promote the target pair for searching and meeting active user's requirement
As a possibility that.
It is described further referring to Fig. 6 to according to the object information method for pushing of the another embodiment of the disclosure.
Fig. 6 shows the flow chart of the object information method for pushing according to the another embodiment of the disclosure.As shown in fig. 6, with
The difference of embodiment shown in FIG. 1 is to further include step S601, and step S104 includes step S602.
In step s 601, object corresponding with the first text is selected from the object that active user actually obtains in history
As the second target object.
In step S602, the information of the information of first object object and the second target object is pushed to the current use
Family.
In one embodiment according to the disclosure, the selection and first from the object that active user actually obtains in history
The corresponding object of text can more accurately search for the target object for meeting existing object requirement as the second target object.
Referring to Fig. 7 to an Application Scenarios-Example according to the object information method for pushing of one embodiment of the disclosure
It is described.
Fig. 7 shows the schematic diagram of the Application Scenarios-Example of the object information method for pushing according to one embodiment of the disclosure.
As shown in fig. 7, though it is shown that two user images, however, it is to be understood that the two user images indicate same
User.Active user sends word dialog (proposing problem) to intelligent customer service, and intelligent customer service is analyzed based on preset text size
Model judges text size.When judging text size < preset threshold n, determines that text is short text, calculated by word2vec
Method calculates the problem of highest other users of cosine similarity propose.When judging text size >=preset threshold n, determine
Text is long text, calculates the problem of highest other users of cosine similarity propose by latent semantic analysis algorithm.In this public affairs
In the embodiment opened, global text training can be carried out to word2vec algorithm and latent semantic analysis algorithm.
For example, phase can be found out by finding out and the problem of active user the problem of similarity highest 1% other users
Like the corresponding goods for spending highest 1% problem, that is, the other users for getting Similar Problems finally have purchased any commodity.It can be with
Is ranked up to the cosine similarity of problem corresponding to these commodity (the second text) similar the problem of selecting to active user
First 5 for spending highest second text, and commodity corresponding to first 5 the second texts are found out, and its information is pushed to
Active user.
In addition, if active user is old user, it can reach preset condition level-one filtering out associated therewith spend and use
Similar Problems (the second text) is looked for behind family again.For example, primary user is friend of the active user on platform.It can choose one
Commodity corresponding to the problem of the problem of grade user proposes and active user similarity highest (for example, first 5), and believed
Breath is pushed to active user.
Furthermore it is also possible to will based on the pervious buying behavior recommendation of active user with active user the problem of corresponding quotient
The information that the information of product is dissolved into the aforementioned commodity based on the second text selecting pushes (displaying) to active user together.
Fig. 8 shows the structural block diagram of the object information driving means according to another embodiment of the disclosure.The device can be with
Including the first search module 801, the second search module 802, selecting module 803 and pushing module 804.
First search module 801 is configured as according to the first text provided for active user other than by active user
The text that provides of user in search be greater than the second text of default similarity threshold with the similarity of the first text.
Second search module 802 be configured to search for provide the second text user's history on actually obtain with second text
This relevant object.
Selecting module 803 is configured as the similarity according to the second text and the first text, from searching out and the second text
First object object is selected in this relevant object.
Pushing module 804 is configured as the information of first object object being pushed to active user.
The foregoing describe the built-in function of object information supplying system and structures, in a possible design, the object
The structure of information transmission system can realize that for object information pushing equipment, as shown in Figure 9, which may include
Processor 901 and memory 902.
The memory 902 executes object information in any of the above-described embodiment for storing support target information transmission system
The program of method for pushing, the processor 901 are configurable for executing the program stored in the memory 902.
The memory 902 is for storing one or more computer instruction, wherein one or more computer refers to
It enables and being executed by the processor 901 to perform the steps of
According to the first text provided for active user in the text provided by the user other than the active user
The similarity of search and first text is greater than the second text of default similarity threshold；
The object relevant to second text actually obtained in the user's history of search offer second text；
It is relevant to second text from what is searched out according to the similarity of second text and first text
First object object is selected in object；
The information of the first object object is pushed to the active user.
In one embodiment of the present disclosure, the first text that active user provides can be active user in a certain platform
The problem of proposition.The text that user other than active user provides can be the user other than active user and propose on the platform
The problem of.In embodiment of the disclosure, by finding second text most like with the first text (the problem of active user)
After (the problem of other users), lookup be second text associated user, rather than the second text (problem) is targeted answers
Case.First object object relevant to the second text is obtained by the behavior of associated user (for example, associated user's actual purchase
The product crossed).Therefore, the information of first object object relevant to the second text can be pushed to active user.
In one embodiment of the present disclosure, second text (its most like with the first text (the problem of active user)
The problem of his user) may have it is multiple, it is thus possible to can exist it is multiple the user of second text is provided, and then may also exist
Multiple first objects relevant to the second text are to object.Therefore, can by from multiple second from different other users
The relevant multiple first objects of text are pushed to active user to the information of object.
It in one embodiment of the present disclosure, can be with latent semantic analysis (Latent Semantics
Indexing) ", " doc2vec " also " cosine similarity (Cosine Similarity) " three kinds of basic technologies determine second
The similarity of text and the first text.
In one embodiment of the present disclosure, each word can be calculated by latent semantic analysis to tie up in text/dictionary
TF-IDF (the inverse literary frequency of word frequency-, Term Frequency-Inverse Document Frequency) value of degree, and pass through
Singular value decomposition carries out dimensionality reduction, word-document matrix after obtaining dimensionality reduction.By the cosine phase for calculating new text and each column of matrix
Like degree, most like text is found.
In one embodiment of the present disclosure, in doc2vec technology, text can be generated by a shallow-layer neural network
This algorithm.Its network generated contains the matrix for having text information, can be used for calculating text similarity.
In one embodiment of the present disclosure, cosine similarity (Cosine Similarity) is referred to by calculating two
The cosine angle of a text vector obtains the similarity of two text vectors.It is higher to be worth bigger similarity.At one of the disclosure
In embodiment, the similarity of the second text and first text is cosine similarity.
In one embodiment of the present disclosure, it determines that the similarity of long text can use latent semantic analysis technology, determines short
The similarity of text can use word2vec technology, and the text for finally finding out active user is similar to the cosine of other users text
Degree.For example, the text of preceding 1% other users most like with the first text can be found out as the second text.Distinguish text
Length is primarily due to the connection that short text relys more on context, and long text relies more on the context near text.Herein,
Long text/short text judges that this text size analysis model can by one based on the text size analysis model of accuracy rate
To be trained by related art method, details are not described herein.
Above is only to latent semantic analysis (Latent Semantics Indexing) ", " doc2vec " also " cosine
The brief introduction of three kinds of technologies of similarity (Cosine Similarity) ", it will be understood by those skilled in the art that can be from related skill
The details of three of the above technology is obtained in art, details are not described herein for particular content.
In disclosure embodiment, the first text of active user's offer is directed to other than by active user by basis
The text that provides of user in search be greater than the second text of default similarity threshold with the similarity of the first text；Search provides
The object relevant to the second text actually obtained in the user's history of second text；According to the phase of the second text and the first text
Like degree, first object object is selected from the object relevant to the second text searched out；The information of first object object is pushed away
Active user is given, being associated with for text similarity and target object can be carried out, searches the corresponding object type of text to replace,
So as to guarantee precision, multi-class, expert's grade mesh when carrying out the push of object information in the text provided for user
Object Selection and information push are marked, especially when no any user information, still can accurately it is expected user
The information of object be pushed to user.
In one embodiment according to the disclosure, the first text of active user's offer is directed to by currently using in basis
In the text that user other than family provides the similarity of search and the first text be greater than default similarity threshold the second text it
Before, one or more computer instruction is also executed by processor 901 to perform the steps of the first of determining active user's offer
Text is the long text that length is more than or equal to pre-set length threshold less than the short text or length of pre-set length threshold.
It is the determination of long text or short text as a result, being calculated by different similarity calculation modes according to the first text
The similarity for the text that first text is provided with the user other than active user.
In one embodiment according to the disclosure, the text size analysis model above-mentioned based on accuracy rate can be passed through
To judge that the first text is long text or short text.For example, it is then short text, text that text size, which is less than pre-set length threshold,
It is then long text that length, which is more than or equal to pre-set length threshold,.
In one embodiment according to the disclosure, when determining the first text is long text, aforementioned latent language can be passed through
Adopted analysis method calculates the similarity of the text that the user other than the first text and active user provides.When determining the first text
When being short text, the text that the user other than the first text and active user provides can be calculated by aforementioned doc2vec method
This similarity.By using different similarity calculation modes for long text and short text, the can be more accurately calculated
The similarity for the text that one text is provided with the user other than active user.
In one embodiment according to the disclosure, determine that the first text that active user provides is that length is less than default length
The short text or length of spending threshold value are more than or equal to the long text of the pre-set length threshold, comprising: long according to preset text
Degree analysis model determines that the first text that active user provides is that length is big less than the short text of pre-set length threshold or length
In the long text for being equal to pre-set length threshold.As previously mentioned, long text/short text can pass through a text based on accuracy rate
Length analysis model judges that this text size analysis model can be trained by related art method, no longer superfluous herein
It states.Word2vec algorithm and the matrix generating algorithm of latent semantic analysis can be based on the automatic off-line training of accuracy.
In one embodiment according to the disclosure, one or more computer instruction is also executed by processor 901
The first text is synthesized to perform the steps of the multiple texts for providing active user.
In one embodiment according to the disclosure, if active user provides multiple texts, for example, proposing multiple
Problem, each problem may relate to different target objects, then multiple texts can be synthesized the first text.This is because
Embodiment of the present disclosure is not necessarily to the corresponding answer of each problem search for active user, and it is similar to can be based on text
Spend find other users proposition similar second text (Similar Problems), thus come from other it is practical it is obtaining with second text
Selection target object in this relevant object.
As previously mentioned, after the multiple texts for providing active user synthesize the first text, it is (current to use with the first text
The problem of family) most like the second text (the problem of other users) may have it is multiple, it is thus possible to can exist and multiple provide the
The user of two texts, and then may also have multiple first objects relevant to the second text to object.It therefore, can will be with
The relevant multiple first objects of multiple the second texts from different other users are pushed to active user to the information of object.
In one embodiment according to the disclosure, according to the similarity of the second text and the first text, from what is searched out
First object object is selected in object relevant to the second text, comprising: according to the similarity of the second text and the first text
Size is ranked up the object relevant to the second text searched out；According to the result of sequence from searching out and the second text
First object object is selected in this relevant object.
In one embodiment according to the disclosure, can find out with the first text similarity it is highest it is preceding it is several (for example,
Preceding 1%, preceding 10% etc.) text of other users is as the second text.According to sequence as a result, from searching out and second text
First object object is selected in relevant object, that is, which and the second text the user of the second text of offer specifically obtain
Relevant object selects first object object from these objects.In the information for the first object object that will be selected in this way to working as
When preceding user pushes, precision, multi-class, expert's grade object information push may be implemented.
In one embodiment according to the disclosure, one or more computer instruction is also executed by processor 901
High degree of association user is filtered out from the user other than active user according to preset degree of association condition to perform the steps of；
Wherein, the search and the in the text provided by the user other than active user according to the first text provided for active user
The similarity of one text is greater than the second text of default similarity threshold, comprising: according to the first text provided for active user
This is searched in the text provided by high degree of association user is greater than the second of default similarity threshold with the similarity of the first text
Text.
In one embodiment according to the disclosure, high degree of association user also refers to that there are special passes with active user
The user of system, for example, good friend of the active user on platform can be used as high degree of association user, active user is good in reality
Friend can also be used as high degree of association user.Degree of association condition can be set in various ways, for example, active user and other users
Connection number, active user to other users be provided with specific markers, active user and other users participated in jointly it is a certain
Behavior etc..In one embodiment according to the disclosure, search and the first text in the text provided by high degree of association user
The second text that this similarity is greater than default similarity threshold can promote the target pair for searching and meeting active user's requirement
As a possibility that.
In one embodiment according to the disclosure, one or more computer instruction is also executed by processor 901
Select object corresponding with the first text as from the object that active user actually obtains in history to perform the steps of
Two target objects；Wherein, the information of first object object is pushed to active user, comprising: by the information of first object object
Active user is pushed to the information of the second target object.
In one embodiment according to the disclosure, the selection and first from the object that active user actually obtains in history
The corresponding object of text can more accurately search for the target object for meeting existing object requirement as the second target object.
The processor 901 is used to execute all or part of the steps in aforementioned approaches method step.
Wherein, it can also include communication interface in the structure of the object information pushing equipment, be pushed for object information
Equipment and other equipment or communication.
Disclosure exemplary embodiment additionally provides a kind of computer storage medium, for storing the object information push
Computer software instructions used in system, it includes for executing in any of the above-described embodiment involved by object information method for pushing
Program.
Figure 10 is adapted for the computer system for realizing the object information method for pushing according to one embodiment of the disclosure
Structural schematic diagram.
As shown in Figure 10, computer system 1000 include central processing unit (CPU) 1001, can according to be stored in only
It reads the program in memory (ROM) 1002 or is loaded into random access storage device (RAM) 1003 from storage section 1008
Program and execute the various processing in above-mentioned embodiment shown in FIG. 1.In RAM1003, it is also stored with the operation of system 1000
Required various programs and data.CPU1001, ROM1002 and RAM1003 are connected with each other by bus 1004.Input/output
(I/O) interface 1005 is also connected to bus 1004.
I/O interface 1005 is connected to lower component: the importation 1006 including keyboard, mouse etc.；Including such as cathode
The output par, c 1007 of ray tube (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section including hard disk etc.
1008；And the communications portion 1009 of the network interface card including LAN card, modem etc..Communications portion 1009 passes through
Communication process is executed by the network of such as internet.Driver 1010 is also connected to I/O interface 1005 as needed.It is detachable to be situated between
Matter 1011, such as disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 1010, so as to
In being mounted into storage section 1008 as needed from the computer program read thereon.
Particularly, according to embodiment of the present disclosure, it is soft to may be implemented as computer above with reference to Fig. 1 method described
Part program.For example, embodiment of the present disclosure includes a kind of computer program product comprising be tangibly embodied in and its readable
Computer program on medium, the computer program include the program code for executing the data processing method of Fig. 1.At this
In the embodiment of sample, which can be downloaded and installed from network by communications portion 1009, and/or from can
Medium 1011 is dismantled to be mounted.
Flow chart and block diagram in attached drawing illustrate system, method and computer according to the various embodiments of the disclosure
The architecture, function and operation in the cards of program product.In this regard, each box in course diagram or block diagram can be with
A part of a module, section or code is represented, a part of the module, section or code includes one or more
Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box
The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical
On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants
It is noted that the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart, Ke Yiyong
The dedicated hardware based system of defined functions or operations is executed to realize, or can be referred to specialized hardware and computer
The combination of order is realized.
Being described in unit or module involved in disclosure embodiment can be realized by way of software, can also
It is realized in a manner of through hardware.Described unit or module also can be set in the processor, these units or module
Title do not constitute the restriction to the unit or module itself under certain conditions.
As on the other hand, the disclosure additionally provides a kind of computer readable storage medium, the computer-readable storage medium
Matter can be computer readable storage medium included in system described in above embodiment；It is also possible to individualism,
Without the computer readable storage medium in supplying equipment.Computer-readable recording medium storage has one or more than one journey
Sequence, described program is used to execute by one or more than one processor is described in disclosed method.
Above description is only the preferred embodiment of the disclosure and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that invention scope involved in the disclosure, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from the inventive concept, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed in the disclosure
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (18)
1. a kind of object information method for pushing characterized by comprising
It is searched in the text provided by the user other than the active user according to the first text provided for active user
It is greater than the second text of default similarity threshold with the similarity of first text；
The object relevant to second text actually obtained in the user's history of search offer second text；
According to the similarity of second text and first text, from the object relevant to second text searched out
Middle selection first object object；
The information of the first object object is pushed to the active user.
2. the method according to claim 1, wherein being directed to the first text that active user provides in the basis
Search is similar greater than presetting to the similarity of first text in the text provided by the user other than the active user
It spends before the second text of threshold value, further includes:
Determine that the first text that the active user provides is that length is less than the short text of pre-set length threshold or length is greater than
Equal to the long text of the pre-set length threshold；
It is the determination of long text or short text as a result, being calculated by different similarity calculation modes according to first text
The similarity for the text that first text is provided with the user other than the active user.
3. method according to claim 1 or 2, which is characterized in that further include:
Multiple texts that the active user is provided synthesize first text.
4. the method according to claim 1, wherein described according to second text and first text
Similarity selects first object object from the object relevant to second text searched out, comprising:
It is related to second text to what is searched out according to the size of second text and the similarity of first text
Object be ranked up；
First object object is selected from the object relevant to second text searched out according to the result of the sequence.
5. the method according to claim 1, wherein further include:
High degree of association user is filtered out from the user other than the active user according to preset degree of association condition；
Wherein, the basis is for the first text of active user's offer in the text provided by the user other than the active user
Similarity of search and first text is greater than the second text of default similarity threshold in this, comprising:
According to the first text provided for active user in the text provided by the high degree of association user search with it is described
The similarity of first text is greater than the second text of default similarity threshold.
6. the method according to claim 1, wherein further include:
Select object corresponding with first text as second from the object that the active user actually obtains in history
Target object,
Wherein, the information of the first object object is pushed to the active user, comprising:
The information of the information of the first object object and second target object is pushed to the active user.
7. according to the method described in claim 2, it is characterized in that, the first text that the determination active user provides is
Length is more than or equal to the long text of the pre-set length threshold less than the short text or length of pre-set length threshold, comprising:
Determine that the first text that the active user provides is that length is less than default length according to preset text size analysis model
The short text or length of spending threshold value are more than or equal to the long text of the pre-set length threshold.
8. the method according to claim 1, wherein the similarity of second text and first text is
Cosine similarity.
9. a kind of object information driving means characterized by comprising
First search module is configured as according to the first text provided for active user other than by the active user
The similarity of search and first text is greater than the second text of default similarity threshold in the text that user provides；
Second search module, be configured to search for actually obtaining in the user's history that second text is provided with described second
The relevant object of text；
Selecting module is configured as the similarity according to second text and first text, from search out with it is described
First object object is selected in the relevant object of second text；
Pushing module is configured as the information of the first object object being pushed to the active user.
10. a kind of electronic equipment, which is characterized in that including memory and processor；Wherein, the memory is for storing one
Or a plurality of computer instruction, wherein one or more computer instruction is executed by the processor to perform the steps of
It is searched in the text provided by the user other than the active user according to the first text provided for active user
It is greater than the second text of default similarity threshold with the similarity of first text；
The object relevant to second text actually obtained in the user's history of search offer second text；
According to the similarity of second text and first text, from the object relevant to second text searched out
Middle selection first object object；
The information of the first object object is pushed to the active user.
11. electronic equipment according to claim 10, which is characterized in that the provided in the basis for active user
One text similarity of search and first text in the text provided by the user other than the active user is greater than pre-
If before the second text of similarity threshold, one or more computer instruction also by the processor execute with realize with
Lower step:
Determine that the first text that the active user provides is that length is less than the short text of pre-set length threshold or length is greater than
Equal to the long text of the pre-set length threshold；
It is the determination of long text or short text as a result, being calculated by different similarity calculation modes according to first text
The similarity for the text that first text is provided with the user other than the active user.
12. electronic equipment described in 0 or 11 according to claim 1, which is characterized in that one or more computer instruction is also
It is executed by the processor to perform the steps of
Multiple texts that the active user is provided synthesize first text.
13. electronic equipment according to claim 10, which is characterized in that described according to second text and described first
The similarity of text selects first object object from the object relevant to second text searched out, comprising:
It is related to second text to what is searched out according to the size of second text and the similarity of first text
Object be ranked up；
First object object is selected from the object relevant to second text searched out according to the result of the sequence.
14. electronic equipment according to claim 10, which is characterized in that one or more computer instruction is also by institute
Processor is stated to execute to perform the steps of
High degree of association user is filtered out from the user other than the active user according to preset degree of association condition；
Wherein, the basis is for the first text of active user's offer in the text provided by the user other than the active user
Similarity of search and first text is greater than the second text of default similarity threshold in this, comprising:
According to the first text provided for active user in the text provided by the high degree of association user search with it is described
The similarity of first text is greater than the second text of default similarity threshold.
15. electronic equipment according to claim 10, which is characterized in that one or more computer instruction is also by institute
Processor is stated to execute to perform the steps of
Select object corresponding with first text as second from the object that the active user actually obtains in history
Target object,
Wherein, the information of the first object object is pushed to the active user, comprising:
The information of the information of the first object object and second target object is pushed to the active user.
16. electronic equipment according to claim 11, which is characterized in that the first of determination active user's offer
Text is the long text of short text or length more than or equal to the pre-set length threshold that length is less than pre-set length threshold, packet
It includes:
Determine that the first text that the active user provides is that length is less than default length according to preset text size analysis model
The short text or length of spending threshold value are more than or equal to the long text of the pre-set length threshold.
17. electronic equipment according to claim 10, which is characterized in that the phase of second text and first text
It is cosine similarity like degree.
18. a kind of computer readable storage medium, is stored thereon with computer instruction, which is characterized in that the computer instruction quilt
Processor realizes the method according to claim 1 when executing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811437429.1A CN110008396B (en) | 2018-11-28 | 2018-11-28 | Object information pushing method, device, equipment and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811437429.1A CN110008396B (en) | 2018-11-28 | 2018-11-28 | Object information pushing method, device, equipment and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110008396A true CN110008396A (en) | 2019-07-12 |
| CN110008396B CN110008396B (en) | 2023-11-24 |
Family
ID=67165001
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811437429.1A Active CN110008396B (en) | 2018-11-28 | 2018-11-28 | Object information pushing method, device, equipment and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110008396B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111708863A (en) * | 2020-06-02 | 2020-09-25 | 上海硬通网络科技有限公司 | Method and device for text matching based on doc2vec and electronic equipment |
| CN112784569A (en) * | 2021-02-04 | 2021-05-11 | 北京秒针人工智能科技有限公司 | Method, system, equipment and storage medium for realizing similar text aggregation |
| WO2021092803A1 (en) * | 2019-11-13 | 2021-05-20 | 深圳市欢太科技有限公司 | Push user determination method and apparatus, server, and storage medium |
| CN112784569B (en) * | 2021-02-04 | 2024-04-19 | 北京秒针人工智能科技有限公司 | Method, system, equipment and storage medium for realizing similar text aggregation |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101206752A (en) * | 2007-12-25 | 2008-06-25 | 北京科文书业信息技术有限公司 | Electric commerce website related products recommendation system and method |
| JP2011070422A (en) * | 2009-09-25 | 2011-04-07 | Dainippon Printing Co Ltd | Goods recommending device |
| CN103942279A (en) * | 2014-04-01 | 2014-07-23 | 百度(中国)有限公司 | Method and device for showing search result |
| CN104809154A (en) * | 2015-03-19 | 2015-07-29 | 百度在线网络技术（北京）有限公司 | Method and device for recommending information |
| CN104933081A (en) * | 2014-03-21 | 2015-09-23 | 阿里巴巴集团控股有限公司 | Search suggestion providing method and apparatus |
| CN105354300A (en) * | 2015-11-05 | 2016-02-24 | 上海智臻智能网络科技股份有限公司 | Information recommendation method and apparatus |
| CN107562761A (en) * | 2016-06-30 | 2018-01-09 | 阿里巴巴集团控股有限公司 | A kind of information-pushing method and device |
| CN107644364A (en) * | 2017-09-18 | 2018-01-30 | 北京京东尚科信息技术有限公司 | Object filter method and system |
| CN107767172A (en) * | 2017-10-12 | 2018-03-06 | 百度在线网络技术（北京）有限公司 | Information-pushing method, device, server and medium |
| CN108241667A (en) * | 2016-12-26 | 2018-07-03 | 百度在线网络技术（北京）有限公司 | For the method and apparatus of pushed information |
| CN108388630A (en) * | 2018-02-13 | 2018-08-10 | 北京奇虎科技有限公司 | A kind of shopping information method for pushing, device and electronic equipment |
| CN108829764A (en) * | 2018-05-28 | 2018-11-16 | 腾讯科技（深圳）有限公司 | Recommendation information acquisition methods, device, system, server and storage medium |
-
2018
- 2018-11-28 CN CN201811437429.1A patent/CN110008396B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101206752A (en) * | 2007-12-25 | 2008-06-25 | 北京科文书业信息技术有限公司 | Electric commerce website related products recommendation system and method |
| JP2011070422A (en) * | 2009-09-25 | 2011-04-07 | Dainippon Printing Co Ltd | Goods recommending device |
| CN104933081A (en) * | 2014-03-21 | 2015-09-23 | 阿里巴巴集团控股有限公司 | Search suggestion providing method and apparatus |
| CN103942279A (en) * | 2014-04-01 | 2014-07-23 | 百度(中国)有限公司 | Method and device for showing search result |
| CN104809154A (en) * | 2015-03-19 | 2015-07-29 | 百度在线网络技术（北京）有限公司 | Method and device for recommending information |
| CN105354300A (en) * | 2015-11-05 | 2016-02-24 | 上海智臻智能网络科技股份有限公司 | Information recommendation method and apparatus |
| CN107562761A (en) * | 2016-06-30 | 2018-01-09 | 阿里巴巴集团控股有限公司 | A kind of information-pushing method and device |
| CN108241667A (en) * | 2016-12-26 | 2018-07-03 | 百度在线网络技术（北京）有限公司 | For the method and apparatus of pushed information |
| CN107644364A (en) * | 2017-09-18 | 2018-01-30 | 北京京东尚科信息技术有限公司 | Object filter method and system |
| CN107767172A (en) * | 2017-10-12 | 2018-03-06 | 百度在线网络技术（北京）有限公司 | Information-pushing method, device, server and medium |
| CN108388630A (en) * | 2018-02-13 | 2018-08-10 | 北京奇虎科技有限公司 | A kind of shopping information method for pushing, device and electronic equipment |
| CN108829764A (en) * | 2018-05-28 | 2018-11-16 | 腾讯科技（深圳）有限公司 | Recommendation information acquisition methods, device, system, server and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| SHREYA BASU: "Information search in the internet markets: Experience versus search goods", 《 ELECTRONIC COMMERCE RESEARCH AND APPLICATIONS》, pages 25 - 37 * |
| SLADESAL: "应用：深度学习下的电商商品推荐1.常见算法套路2.item2vec的工程引入3.python代码实现", pages 1, Retrieved from the Internet <URL:https://cloud.tencent.com/developer/article/1195496> * |
| 张恒 等: "考虑图书推荐列表相似性的图书推荐研究", 《图书与情报》, pages 128 - 134 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021092803A1 (en) * | 2019-11-13 | 2021-05-20 | 深圳市欢太科技有限公司 | Push user determination method and apparatus, server, and storage medium |
| CN111708863A (en) * | 2020-06-02 | 2020-09-25 | 上海硬通网络科技有限公司 | Method and device for text matching based on doc2vec and electronic equipment |
| CN111708863B (en) * | 2020-06-02 | 2024-03-15 | 上海硬通网络科技有限公司 | Text matching method and device based on doc2vec and electronic equipment |
| CN112784569A (en) * | 2021-02-04 | 2021-05-11 | 北京秒针人工智能科技有限公司 | Method, system, equipment and storage medium for realizing similar text aggregation |
| CN112784569B (en) * | 2021-02-04 | 2024-04-19 | 北京秒针人工智能科技有限公司 | Method, system, equipment and storage medium for realizing similar text aggregation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110008396B (en) | 2023-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bauman et al. | Aspect based recommendations: Recommending items with the most valuable aspects based on user reviews | |
| US7805431B2 (en) | System and method for generating a display of tags | |
| Liu et al. | Analyzing changes in hotel customers’ expectations by trip mode | |
| US8909626B2 (en) | Determining user preference of items based on user ratings and user features | |
| JP5603337B2 (en) | System and method for supporting search request by vertical proposal | |
| US7283997B1 (en) | System and method for ranking the relevance of documents retrieved by a query | |
| US20190318407A1 (en) | Method for product search using the user-weighted, attribute-based, sort-ordering and system thereof | |
| EP3579125A1 (en) | System, computer-implemented method and computer program product for information retrieval | |
| US20120143789A1 (en) | Click model that accounts for a user's intent when placing a quiery in a search engine | |
| CN107835994A (en) | Pass through the task focused search of image | |
| CN111061946A (en) | Scenario content recommendation method and device, electronic equipment and storage medium | |
| US20150081714A1 (en) | Active Knowledge Guidance Based on Deep Document Analysis | |
| CN109359247A (en) | Content delivery method and storage medium, computer equipment | |
| CN110008396A (en) | Object information method for pushing, device, equipment and computer readable storage medium | |
| Ma et al. | A visual analytical approach for transfer learning in classification | |
| Poussevin et al. | Extended recommendation framework: Generating the text of a user review as a personalized summary | |
| US20090077093A1 (en) | Feature Discretization and Cardinality Reduction Using Collaborative Filtering Techniques | |
| JP5138621B2 (en) | Information processing apparatus, dissatisfied product discovery method and program | |
| Hssina et al. | Recommendation system using the k-nearest neighbors and singular value decomposition algorithms | |
| KR20230025459A (en) | Methods and apparatus for analyzing information | |
| Javari et al. | Recommender systems for social networks analysis and mining: precision versus diversity | |
| CN110020195B (en) | Article recommendation method and device, storage medium and electronic equipment | |
| Zhang et al. | Less is more: Rejecting unreliable reviews for product question answering | |
| Lim et al. | No. 3. Hybrid-based Recommender System for Online Shopping: A Review: Manuscript Received: 8 February 2023, Accepted: 21 February 2023, Published: 15 March 2023, ORCiD: 0000-0002-7190-0837 | |
| Banouar et al. | Enriching SPARQL queries by user preferences for results adaptation |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200918 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20200918 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |