CN107748754A - A kind of knowledge mapping improving method and device - Google Patents
A kind of knowledge mapping improving method and device Download PDFInfo
- Publication number
- CN107748754A CN107748754A CN201710833203.2A CN201710833203A CN107748754A CN 107748754 A CN107748754 A CN 107748754A CN 201710833203 A CN201710833203 A CN 201710833203A CN 107748754 A CN107748754 A CN 107748754A
- Authority
- CN
- China
- Prior art keywords
- label
- commodity
- extracted
- knowledge
- illustrative plates
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- 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 invention discloses a kind of knowledge mapping improving method and device, belong to technical field of electronic commerce.Method includes：The label of commodity is extracted from the merchandise news including at least commodity picture；The label extracted is added to knowledge mapping, knowledge mapping includes knowledge of goods collection of illustrative plates and/or user knowledge collection of illustrative plates；The relation corresponding with label is set in knowledge mapping.The embodiment of the present invention can enrich one's knowledge the label in collection of illustrative plates, to realize that the accurate recommendation of commodity is bought in accurate search of the user to commodity and realizing to user.
Description
Technical field
The present invention relates to technical field of electronic commerce, more particularly to a kind of knowledge mapping improving method and device.
Background technology
Knowledge mapping is a kind of data structure based on figure, and its node represents entity (entity) or concept
(concept), can in a search engine by using knowledge mapping, while various semantic relations between representing entity/concept
The related information of complexity is preferably inquired about, user view is understood from semantic level, improves search quality.Such as in ecommerce
Field, it can aid in user using knowledge of goods collection of illustrative plates and search corresponding commodity, and can using user knowledge collection of illustrative plates
Help to carry out commercial product recommending to user.
In the prior art, the label of included commodity generally only identifies business in knowledge of goods collection of illustrative plates or user knowledge collection of illustrative plates
The fundamental characteristics such as the brands of product, color, material, size, base model, and carry when label is and uploads commodity according to seller user
The information of confession, directly mark is on commodity after manual examination and verification, but the angle that user is paid close attention to commodity at present is increasingly
More, if user wants the characteristic progress commercial articles searching beyond the fundamental characteristics for commodity, for example some of decorative pattern is specified in search
Commodity, it may result in because knowledge mapping can not realize the accurate search to commodity comprising corresponding label；In addition, also without
The angle that method is paid close attention to commodity according to user realizes the accurate recommendation that commodity are bought to user.
Therefore exist in the prior art knowledge mapping Commercial goods labelses it is fairly simple coarse the problem of, need badly to knowledge mapping
Perfect, the label in collection of illustrative plates of enriching one's knowledge is carried out, to realize that accurate search and realization of the user to commodity buy business to user
The accurate recommendation of product.
The content of the invention
In order to solve problem of the prior art, the embodiments of the invention provide a kind of knowledge mapping improving method and device,
The label enriched one's knowledge in collection of illustrative plates, precisely pushing away for commodity is bought to user to realize accurate search of the user to commodity and realize
Recommend.The technical scheme is as follows：
First aspect, there is provided a kind of knowledge mapping improving method, methods described include：
The label of commodity is extracted from the merchandise news including at least commodity picture；
The label extracted is added to knowledge mapping, the knowledge mapping includes knowledge of goods collection of illustrative plates and/or use
Family knowledge mapping；
The relation corresponding with the label extracted is set in the knowledge mapping.
With reference in a first aspect, in the first possible implementation, the merchandise news only includes the commodity picture,
The label that commodity are extracted from the merchandise news comprising commodity picture includes：
Characteristics of image is extracted from the commodity picture；
Obtain label corresponding with described image feature.
With reference in a first aspect, in second of possible implementation, the merchandise news include the commodity picture and
The text description information of the commodity, the label of commodity is extracted in the merchandise news from including at least commodity picture to be included：
Extract the characteristics of image in the commodity picture；And
Extract the keyword in the text description information；
According to described image feature and the keyword, the label of the commodity is determined；
The text description information includes the heading message of the commodity, the info web of the commodity and the commodity
At least one of comment information.
With reference to the first or second of possible implementation of first aspect, in the third possible implementation,
It is described to include the label extracted added to knowledge mapping：
It is determined that label classification corresponding to the label extracted；
According to the label classification, the label extracted is added to the knowledge mapping；
It is described to set the relation corresponding with the label extracted to include in the knowledge mapping：
Relation between the label extracted and the commodity is set in the knowledge of goods collection of illustrative plates；
The relation between the label and other labels extracted is matched in the user knowledge collection of illustrative plates；And
The relation between the label and other labels extracted is updated in the user knowledge collection of illustrative plates.
With reference to the first or second of possible implementation of first aspect, in the 4th kind of possible implementation,
It is described to include the label extracted added to knowledge mapping：
The label extracted is added to the tag library；
Define the label classification belonging to the label extracted；
According to the label classification, the label extracted is added to the knowledge mapping；
It is described to set the relation corresponding with the label extracted to include in the knowledge mapping：
Relation between the label extracted and the commodity is set in the knowledge of goods collection of illustrative plates；
The relation between the label and other labels extracted is updated in the user knowledge collection of illustrative plates.
Second aspect, there is provided a kind of knowledge mapping Perfected device, described device include：
Extraction module, for extracting the label of commodity from the merchandise news including at least commodity picture；
Add module, for the label extracted to be added into knowledge mapping, the knowledge mapping is known including commodity
Know collection of illustrative plates and/or user knowledge collection of illustrative plates；
Setup module, the corresponding relation of the label for setting with extracting in the knowledge mapping.
With reference to second aspect, in the first possible implementation, the merchandise news only includes the commodity picture,
The extraction module is specifically used for：
Characteristics of image is extracted from the commodity picture；
Obtain label corresponding with described image feature.
With reference to second aspect, in second of possible implementation, the merchandise news include the commodity picture and
The text description information of the commodity, the extraction module are specifically additionally operable to：
Extract the characteristics of image in the commodity picture；And
Extract the keyword in the text description information；
According to described image feature and the keyword, the label of the commodity is determined；
The text description information includes the heading message of the commodity, the info web of the commodity and the commodity
At least one of comment information.
With reference to the first or second of possible implementation of second aspect, in the third possible implementation,
The add module is specifically used for：
It is determined that label classification corresponding to the label extracted；
According to the label classification, the label extracted is added to the knowledge mapping；
The setup module is specifically used for：
Relation between the label extracted and the commodity is set in the knowledge of goods collection of illustrative plates；
The relation between the label and other labels extracted is matched in the user knowledge collection of illustrative plates；And
The relation between the label and other labels extracted is updated in the user knowledge collection of illustrative plates.
With reference to the first or second of possible implementation of second aspect, in the 4th kind of possible implementation,
The add module is specifically additionally operable to：
The label is added to the tag library；
Define the label classification belonging to the label；
According to the label classification, the label extracted is added to the knowledge mapping；
The setup module is specifically additionally operable to：
Relation between the label extracted and the commodity is set in the knowledge of goods collection of illustrative plates；
The relation between the label and other labels extracted is updated in the user knowledge collection of illustrative plates.
The beneficial effect that technical scheme provided in an embodiment of the present invention is brought is：
By extracting the label of commodity from the merchandise news including at least commodity picture, and label is added to knowledge graph
Spectrum, knowledge mapping include knowledge of goods collection of illustrative plates and/or user knowledge collection of illustrative plates, and the mark for setting and extracting in knowledge mapping
Corresponding relation is signed, the knowledge mapping of the prior art that compares only identifies the brand of commodity, color, material, size, basic
For the fundamental characteristics such as pattern, the label of the commodity extracted from the merchandise news including at least commodity picture can more reflect
The more information and characteristic of commodity, hence in so that the label in knowledge of goods collection of illustrative plates and/or user knowledge collection of illustrative plates is more abundant, more
Sample, so as to realize that the accurate recommendation of commodity is bought in accurate search and realization of the user to commodity to user.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, make required in being described below to embodiment
Accompanying drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for
For those of ordinary skill in the art, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings
Accompanying drawing.
Fig. 1 is a kind of knowledge mapping improving method flow chart provided in an embodiment of the present invention；
Fig. 2 is a kind of knowledge mapping improving method flow chart provided in an embodiment of the present invention；
Fig. 3 is a kind of knowledge mapping improving method flow chart provided in an embodiment of the present invention；
Fig. 4 is a kind of knowledge mapping Perfected device structural representation provided in an embodiment of the present invention.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached in the embodiment of the present invention
Figure, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only this
Invention part of the embodiment, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art exist
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
Embodiment one
The embodiments of the invention provide a kind of knowledge mapping improving method, shown in reference picture 1, this method includes：
101st, the label of commodity is extracted from the merchandise news including at least commodity picture.
Specifically, merchandise news only includes commodity picture, the mark of commodity is extracted from the merchandise news comprising commodity picture
Label, the process can include：
Characteristics of image is extracted from commodity picture；
Obtain label corresponding with characteristics of image.
Merchandise news includes commodity picture and the text description information of commodity, from the merchandise news including at least commodity picture
The label of middle extraction commodity, the process can include：
Extract the characteristics of image in commodity picture；And
Extract the keyword in text description information；
According to characteristics of image and keyword, the label of commodity is determined；
In heading message of the text description information including commodity, the info web of commodity and the comment information of commodity at least
It is a kind of.
102nd, the label extracted is added to knowledge mapping, knowledge mapping includes knowledge of goods collection of illustrative plates and/or user knows
Know collection of illustrative plates.
The 103rd, the relation corresponding with the label extracted is set in knowledge mapping.
The embodiments of the invention provide a kind of knowledge mapping improving method, by believing from the commodity including at least commodity picture
The label of commodity is extracted in breath, and the label extracted be added to knowledge mapping, knowledge mapping include knowledge of goods collection of illustrative plates with/
Or user knowledge collection of illustrative plates, and the relation corresponding with the label extracted is set in knowledge mapping, compare prior art
Knowledge mapping only identify the fundamental characteristics such as the brands of commodity, color, material, size, base model for, from including at least
The label of the commodity extracted in the merchandise news of commodity picture can more reflect the more information and characteristic of commodity, hence in so that
Label in knowledge of goods collection of illustrative plates and/or user knowledge collection of illustrative plates more enriches, is diversified, becoming more meticulous, so as to realize user to business
The accurate recommendation of commodity is bought in the accurate search and realization of product to user.
Embodiment two
The embodiments of the invention provide a kind of knowledge mapping improving method, shown in reference picture 2, this method includes：
201st, the label of commodity is extracted from the merchandise news including at least commodity picture.
Wherein, the commodity in commodity picture can be clothes, food, cosmetics, furniture, daily necessities or other commodity.
The label of commodity is used to describe commodity, and label can identify the names of an article of commodity, brand, size, color, style, version
Type, decorative pattern or for indicating to summarize in commodity or summing-up characteristic range defines, wherein, style can include " original
Wind ", " army's wind ", " trend of back-to-ancients ", " sweet and lovely " etc.；For indicating to summarize in commodity or summing-up characteristic range defines,
Such as " aobvious thin ", " light tone " etc..
In addition, label can also be other information that can be used in identifying commodity, and the embodiment of the present invention is to specific
Label is not limited.
Specifically, merchandise news only includes commodity picture, the mark of commodity is extracted from the merchandise news comprising commodity picture
Label, the process can include：
Obtain the merchandise news including at least commodity picture that buyer user uploads；
Characteristics of image is extracted from commodity picture；
Obtain label corresponding with characteristics of image.
Wherein, characteristics of image is extracted from commodity picture, the process can include：
Specifically, the process can include：
Commodity picture is pre-processed；
Extract the characteristics of image in pretreated commodity picture.
In the present embodiment, by being pre-processed to commodity picture, to eliminate information unrelated in commodity picture, for example filter
Except interference, noise, recover useful real information, the reliability thus, it is possible to ensure the characteristics of image in extraction commodity picture.
Wherein, label corresponding with characteristics of image is obtained, the process can include：
Obtain the identification model that trains, the input of the identification model trained is characteristics of image, export for the image
The corresponding label of feature, wherein, identification model can be depth convolutional neural networks；
When characteristics of image is input to the identification model trained, it is determined that the label that the identification model trained is exported is
Label corresponding with characteristics of image.
For example acquisition label corresponding with the characteristics of image of commodity can include " woman coat ", " shortage of money ", " institute
Wind ", " carreau "；Again for example, obtain corresponding with the characteristics of image of commodity label can including " satchel ", " light brown ",
" careful ", " ox-hide ", or " satchel ", " dark-brown ", " coarse grain ", " ox-hide " etc..
Due to from the merchandise news comprising commodity picture extract commodity label, can reflect commodity more information and
Characteristic, hence in so that the label of commodity is abundant, careful.
Specifically, merchandise news includes commodity picture and the text description information of commodity, from including at least commodity picture
The label of commodity is extracted in merchandise news, the process can include：
Extract the characteristics of image in commodity picture；And
Extract the keyword in text description information；
According to characteristics of image and keyword, the label of commodity is determined；
Wherein, the heading message of the text description information of commodity including commodity, commodity info web or commodity comment
At least one of information.
Wherein, the keyword in text description information is extracted, the process can include：
Effective text is drawn into from text description information；
Chinese word segmentation and part-of-speech tagging are carried out to text, utilize LDA (Latent Dirichlet Allocation) model
Or PLSA (Probabilistic Latent Semantic Analysis) model clusters to text；
Classification mark is carried out after arranging cluster result, carries out supervised learning, the text that do not mark in future is classified
Training；
Filter stop words, using TF-LDF (term frequency-inverse document frequency) or
TextRank algorithm carries out keyword abstraction to the text after cluster.
Wherein, according to characteristics of image and keyword, the label of commodity is determined, the process can include：
Multiple labels corresponding to characteristics of image are determined, and determine multiple labels corresponding to keyword；
Pair label corresponding with characteristics of image and label corresponding with keyword merge duplicate removal processing；
Label after duplicate removal is handled is defined as the label extracted from the merchandise news of user's input.
Exemplary, label corresponding with the characteristics of image of commodity includes " woman style wedding gauze kerchief ", " white ", " one-piece dress ", " yarn
Cloth ", " fringe margin decorative pattern ", label corresponding with the keyword that the text description information of commodity is included include " shoulder ", " white
These labels are merged duplicate removal processing by color ", " woman style wedding gauze kerchief ", the labels of commodity include " woman style wedding gauze kerchief ", " shoulder ",
" white ", " fringe margin decorative pattern ", " white ", " one-piece dress ".
In the embodiment of the present invention, due to extracting commodity from the merchandise news comprising commodity picture and text description information
Label, can more reflect the more information and characteristic of commodity, therefore further such that the label of commodity is more rich, diversified, more
Become more meticulous.
202nd, the label classification corresponding to the label that determination is extracted.
Wherein, multiple label classifications are preset with knowledge mapping, knowledge mapping includes knowledge of goods collection of illustrative plates and/or user knows
Know collection of illustrative plates, label classification is at least " name of an article ", " brand ", " color ", " style ", " version type " and one kind or more in " decorative pattern "
Kind.
Wherein, knowledge of goods collection of illustrative plates includes commodity, the relation letter corresponding to commodity between multiple labels and each label
Breath；User, the relation information corresponding to user between multiple labels and each label are contained in user knowledge collection of illustrative plates.
Specifically, when identifying that the label extracted is included in the tag library of knowledge mapping, it is determined that the label extracted
The corresponding relation being provided with corresponding label classification, wherein tag library between label and label classification.
Exemplary, the label extracted is " academic attitude ", " cream-coloured ", according to label and the other corresponding relation of tag class,
Label classification corresponding to " academic attitude " is " style ", and label classification corresponding to " cream-coloured " is " color ".
The present invention is not limited to specific determination process.
203rd, according to label classification, the label extracted is added to knowledge mapping.
Specifically, the present invention is not limited to specific adding procedure.
It is worth noting that, step 202 to step 203 is to realize the label extracted being added to knowledge mapping, knowledge
Collection of illustrative plates includes the process of knowledge of goods collection of illustrative plates and/or user knowledge collection of illustrative plates, in addition to the mode of above-mentioned steps, can also pass through
Other modes realize the process, and the embodiment of the present invention is not limited to specific mode.
It should be noted that for the label extracted is added into knowledge of goods collection of illustrative plates, the label is except can be from buying
Outside being extracted in the merchandise news including at least commodity picture that family user uploads, it can also be comprised at least from what seller user uploaded
Extracted in the merchandise news of commodity picture.
In the embodiment of the present invention, by the way that the label extracted is added into knowledge mapping, realize to the complete of knowledge mapping
It is kind so that label in knowledge mapping it is more rich, it is diversified, become more meticulous.
The 204th, relation between the label and commodity extracted is set in knowledge of goods collection of illustrative plates.
Specifically, the label classification according to corresponding to the label extracted, establishes what is extracted in knowledge of goods collection of illustrative plates
Incidence relation between label and corresponding commodity；And
According to default tie-in sale information, the pass established between the commodity corresponding to the label extracted and other commodity
System.
In the embodiment of the present invention, by the way that the label extracted is added into knowledge of goods collection of illustrative plates, and in knowledge of goods collection of illustrative plates
It is middle that the relation corresponding with label is set so that user can scan for corresponding business according to the label in knowledge of goods collection of illustrative plates
Product, it is achieved thereby that accurate search of the user to commodity.
205th, the relation between the label and other labels extracted is matched in user knowledge collection of illustrative plates.
Specifically, the process can include：
The matching relationship between the label pre-established is obtained, wherein, matching relationship is based on the relevance between label
Established；
According to the matching relationship between label, the label extracted and other marks of user are matched in user knowledge collection of illustrative plates
Relation between label.
Exemplary, for example the matching relationship between label " defending clothing " and label " outdoor activity " is pre-established, such as extract
The label gone out is " defending clothing ", label " outdoor activity " be present in user knowledge collection of illustrative plates, then matches label in user knowledge collection of illustrative plates
" defending clothing " and label " outdoor activity ".
206th, the relation between the label and other labels extracted is updated in user knowledge collection of illustrative plates.
Specifically, the process can include：
The history purchaser record of user is obtained, and the interest model of user, interest model are determined according to history purchaser record
At least it is used for instruction user at least one of consumption preferences, hobby description and collocation preference；
According to the interest model of user, label is updated in user knowledge collection of illustrative plates and is subordinated to the other label of other tag class
Between relation.
Relation between the label and other labels that are extracted by being updated in user knowledge collection of illustrative plates, is further described out
The hobby interest of user, so as to further realize the accurate recommendation that commodity are bought to user.
It is worth noting that, step 204 to step 206 is to realize to set the relation corresponding with label in knowledge mapping
Process, in addition to the mode of above-mentioned steps, the process can also be realized by other means, the embodiment of the present invention is to specific
Mode be not limited.
It should be noted that execution sequencing step of the embodiment of the present invention to step 204, step 205 and step 206
It is specific to limit, in actual applications, while step 204, step 205 and step 206 are performed, be preferred scheme, it is perfect to improve
The efficiency of knowledge mapping.
In the embodiment of the present invention, by the way that the label extracted is added into user knowledge collection of illustrative plates, and in user knowledge collection of illustrative plates
It is middle that the relation corresponding with label is set so that the label related to user is more rich, diversified, fine in user knowledge collection of illustrative plates
Change, it is achieved thereby that buying the accurate recommendation of commodity to user.
The embodiments of the invention provide a kind of knowledge mapping improving method, by believing from the commodity including at least commodity picture
The label of commodity is extracted in breath, and the label extracted be added to knowledge mapping, knowledge mapping include knowledge of goods collection of illustrative plates with/
Or user knowledge collection of illustrative plates, and the relation corresponding with the label extracted is set in knowledge mapping, compare prior art
Knowledge mapping only identify the fundamental characteristics such as the brands of commodity, color, material, size, base model for, from including at least
The label of the commodity extracted in the merchandise news of commodity picture can more reflect the more information and characteristic of commodity, hence in so that
Label in knowledge of goods collection of illustrative plates and/or user knowledge collection of illustrative plates is more abundant, diversified, so as to realize user to the accurate of commodity
The accurate recommendation of commodity is bought in search and realization to user.
Embodiment three
The embodiments of the invention provide a kind of knowledge mapping improving method, shown in reference picture 3, this method includes：
301st, the label of commodity is extracted from the merchandise news including at least commodity picture.
Specifically, the step is identical with step 201, it is not repeated here herein.
In the embodiment of the present invention, due to extracting commodity from the merchandise news comprising commodity picture and text description information
Label, can more reflect the more information and characteristic of commodity, therefore further such that the label of commodity is more rich, diversified, more
Become more meticulous.
302nd, the label extracted is added to tag library.
Specifically, when identifying that the label extracted is not included in the tag library of knowledge mapping, the label that will extract
Added to tag library.
The present invention is not limited to specific determination process.
Due to the label extracted is added in original tag library for not including the label, hence in so that tag library includes
Label it is more abundant.
303rd, the label classification belonging to the label that definition is extracted.
Specifically, judge in tag library with the presence or absence of the label with the label extracted with same alike result；
If in the presence of, it is determined that there is the label classification corresponding to the label of same alike result with the label extracted to be carried
Label classification corresponding to the label got；
Otherwise, then the label classification corresponding with the label being previously mentioned is created.
304th, according to label classification, the label extracted is added to knowledge mapping.
The present invention is not limited to specific definition procedure.
It is worth noting that, step 302 to step 304 is to realize the mistake that the label extracted is added to knowledge mapping
Journey, in addition to the mode of above-mentioned steps, the process can also be realized by other means, the embodiment of the present invention is to specific side
Formula is not limited.
It should be noted that for the label extracted is added into knowledge of goods collection of illustrative plates, the label is except can be from buying
Outside being extracted in the merchandise news including at least commodity picture that family user uploads, it can also be comprised at least from what seller user uploaded
Extracted in the merchandise news of commodity picture.
In the embodiment of the present invention, by the way that the label extracted is added into corresponding label classification, realize to knowledge graph
Spectrum it is perfect so that label in knowledge mapping it is more rich, it is diversified, become more meticulous.
The 305th, relation between the label and commodity extracted is set in knowledge of goods collection of illustrative plates.
Specifically, the step is identical with step 204, it is not repeated here herein.
In the embodiment of the present invention, by the way that the label extracted is added into knowledge of goods collection of illustrative plates, and in knowledge of goods collection of illustrative plates
It is middle that the relation corresponding with label is set so that user can scan for corresponding business according to the label in knowledge of goods collection of illustrative plates
Product, it is achieved thereby that accurate search of the user to commodity.
306th, the relation between the label extracted and other labels of user is updated in user knowledge collection of illustrative plates.
Specifically, the step is identical with step 206, it is not repeated here herein.
It is worth noting that, step 304 to step 305 be realize set in knowledge mapping it is relative with the label extracted
The process for the relation answered, in addition to the mode of above-mentioned steps, the process can also be realized by other means, the present invention is implemented
Example is not limited to specific mode.
It should be noted that the embodiment of the present invention is specifically limited the execution sequencing step of step 304 and step 305
It is fixed, in actual applications, while step 304 and step 305 are performed, be preferred scheme, to improve the efficiency for improving knowledge mapping.
In the embodiment of the present invention, by the way that the label extracted is added into user knowledge collection of illustrative plates, and in user knowledge collection of illustrative plates
It is middle that the relation corresponding with the label extracted is set so that the label related to user is more rich, more in user knowledge collection of illustrative plates
Sample, become more meticulous, it is achieved thereby that buying the accurate recommendation of commodity to user.
The embodiments of the invention provide a kind of knowledge mapping improving method, by believing from the commodity including at least commodity picture
The label of commodity is extracted in breath, and the label extracted be added to knowledge mapping, knowledge mapping include knowledge of goods collection of illustrative plates with/
Or user knowledge collection of illustrative plates, and the relation corresponding with the label extracted is set in knowledge mapping, compare prior art
Knowledge mapping only identify the fundamental characteristics such as the brands of commodity, color, material, size, base model for, from including at least
The label of the commodity extracted in the merchandise news of commodity picture can more reflect the more information and characteristic of commodity, hence in so that
Label in knowledge of goods collection of illustrative plates and/or user knowledge collection of illustrative plates is more abundant, diversified, so as to realize user to the accurate of commodity
The accurate recommendation of commodity is bought in search and realization to user.
Example IV
Shown in Figure 4 the embodiments of the invention provide a kind of knowledge mapping Perfected device, device 4 includes：
Extraction module 41, for extracting the label of commodity from the merchandise news including at least commodity picture；
Add module 42, for the label extracted to be added into knowledge mapping, knowledge mapping includes knowledge of goods collection of illustrative plates
And/or user knowledge collection of illustrative plates；
Setup module 43, the corresponding relation of label for setting with extracting in knowledge mapping.
Optionally, merchandise news only includes commodity picture, and extraction module 41 is specifically used for：
Characteristics of image is extracted from commodity picture；
Obtain label corresponding with characteristics of image.
Optionally, merchandise news includes commodity picture and the text description information of commodity, and extraction module 41 is specifically additionally operable to：
Extract the characteristics of image in commodity picture；And
Extract the keyword in text description information；
According to characteristics of image and keyword, the label of commodity is determined；
In heading message of the text description information including commodity, the info web of commodity and the comment information of commodity at least
It is a kind of.
Optionally, add module 42 is specifically used for：
It is determined that label classification corresponding to the label extracted；
According to label classification, the label extracted is added to knowledge mapping；
Setup module 43 is specifically used for：
Relation between the label and commodity extracted is set in knowledge of goods collection of illustrative plates；
The relation between the label and other labels extracted is matched in user knowledge collection of illustrative plates；And
The relation between the label and other labels extracted is updated in user knowledge collection of illustrative plates.
Optionally, add module 42 is specifically additionally operable to：
Label is added to tag library；
Define the label classification belonging to label；
According to label classification, the label extracted is added to knowledge mapping；
Setup module 43 is specifically additionally operable to：
Relation between the label and commodity extracted is set in knowledge of goods collection of illustrative plates；
The relation between the label and other labels extracted is updated in user knowledge collection of illustrative plates.
The embodiments of the invention provide a kind of knowledge mapping Perfected device, the device passes through from including at least commodity picture
The label of commodity is extracted in merchandise news, and the label extracted is added to knowledge mapping, knowledge mapping includes knowledge of goods
Collection of illustrative plates and/or user knowledge collection of illustrative plates, and the relation corresponding with the label extracted is set in knowledge mapping, compared to more existing
For the knowledge mapping for having technology only identifies the fundamental characteristics such as the brands of commodity, color, material, size, base model, to
The label of the commodity extracted less in the merchandise news comprising commodity picture can more reflect the more information and characteristic of commodity, because
This make it that the label in knowledge of goods collection of illustrative plates and/or user knowledge collection of illustrative plates is more abundant, diversified, so as to realize user to commodity
Accurate search and realize to user buy commodity accurate recommendation.
Above-mentioned all optional technical schemes, any combination can be used to form the alternative embodiment of the present invention, herein no longer
Repeat one by one.
It should be noted that：The knowledge mapping Perfected device that above-described embodiment provides is performing knowledge mapping improving method
When, only with the division progress of above-mentioned each functional module for example, in practical application, above-mentioned function can be divided as needed
With by different functional module completions, i.e., the internal structure of device is divided into different functional modules, to complete above description
All or part of function.In addition, the knowledge mapping Perfected device and knowledge mapping improving method that above-described embodiment provides are real
Apply example and belong to same design, its specific implementation process refers to embodiment of the method, repeats no more here.
One of ordinary skill in the art will appreciate that hardware can be passed through by realizing all or part of step of above-described embodiment
To complete, associated hardware can also be instructed to complete by program, described program can be stored in a kind of computer can
Read in storage medium, storage medium mentioned above can be read-only storage, disk or CD etc..
The foregoing is only presently preferred embodiments of the present invention, be not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent substitution and improvements made etc., it should be included in the scope of the protection.
Claims (10)
1. a kind of knowledge mapping improving method, it is characterised in that methods described includes：
The label of commodity is extracted from the merchandise news including at least commodity picture；
The label extracted is added to knowledge mapping, the knowledge mapping includes knowledge of goods collection of illustrative plates and/or user knows
Know collection of illustrative plates；
The relation corresponding with the label extracted is set in the knowledge mapping.
2. according to the method for claim 1, it is characterised in that the merchandise news only includes the commodity picture, described
The label of commodity is extracted from the merchandise news comprising commodity picture to be included：
Characteristics of image is extracted from the commodity picture；
Obtain label corresponding with described image feature.
3. according to the method for claim 1, it is characterised in that the merchandise news includes the commodity picture and the business
The text description information of product, the label of commodity is extracted in the merchandise news from including at least commodity picture to be included：
Extract the characteristics of image in the commodity picture；And
Extract the keyword in the text description information；
According to described image feature and the keyword, the label of the commodity is determined；
The comment of the heading message of the text description information including the commodity, the info web and the commodity of the commodity
At least one of information.
4. according to the method in claim 2 or 3, it is characterised in that
It is described to include the label extracted added to knowledge mapping：
It is determined that label classification corresponding to the label extracted；
According to the label classification, the label extracted is added to the knowledge mapping；
It is described to set the relation corresponding with the label extracted to include in the knowledge mapping：
Relation between the label extracted and the commodity is set in the knowledge of goods collection of illustrative plates；
The relation between the label and other labels extracted is matched in the user knowledge collection of illustrative plates；And
The relation between the label and other labels extracted is updated in the user knowledge collection of illustrative plates.
5. according to the method in claim 2 or 3, it is characterised in that
It is described to include the label extracted added to knowledge mapping：
The label extracted is added to the tag library；
Define the label classification belonging to the label extracted；
According to the label classification, the label extracted is added to the knowledge mapping；
It is described to set the relation corresponding with the label extracted to include in the knowledge mapping：
Relation between the label extracted and the commodity is set in the knowledge of goods collection of illustrative plates；
The relation between the label and other labels extracted is updated in the user knowledge collection of illustrative plates.
6. a kind of knowledge mapping Perfected device, it is characterised in that described device includes：
Extraction module, for extracting the label of commodity from the merchandise news including at least commodity picture；
Add module, for the label extracted to be added into knowledge mapping, the knowledge mapping includes knowledge of goods figure
Spectrum and/or user knowledge collection of illustrative plates；
Setup module, the corresponding relation of the label for setting with extracting in the knowledge mapping.
7. device according to claim 6, it is characterised in that the merchandise news only includes the commodity picture, described
Extraction module is specifically used for：
Characteristics of image is extracted from the commodity picture；
Obtain label corresponding with described image feature.
8. device according to claim 6, it is characterised in that the merchandise news includes the commodity picture and the business
The text description information of product, the extraction module are specifically additionally operable to：
Extract the characteristics of image in the commodity picture；And
Extract the keyword in the text description information；
According to described image feature and the keyword, the label of the commodity is determined；
The comment of the heading message of the text description information including the commodity, the info web and the commodity of the commodity
At least one of information.
9. the device according to claim 7 or 8, it is characterised in that
The add module is specifically used for：
It is determined that label classification corresponding to the label extracted；
According to the label classification, the label extracted is added to the knowledge mapping；
The setup module is specifically used for：
Relation between the label extracted and the commodity is set in the knowledge of goods collection of illustrative plates；
The relation between the label and other labels extracted is matched in the user knowledge collection of illustrative plates；And
The relation between the label and other labels extracted is updated in the user knowledge collection of illustrative plates.
10. the device according to claim 7 or 8, it is characterised in that
The add module is specifically additionally operable to：
The label is added to the tag library；
Define the label classification belonging to the label；
According to the label classification, the label extracted is added to the knowledge mapping；
The setup module is specifically additionally operable to：
Relation between the label extracted and the commodity is set in the knowledge of goods collection of illustrative plates；
The relation between the label and other labels extracted is updated in the user knowledge collection of illustrative plates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710833203.2A CN107748754B (en) | 2017-09-15 | 2017-09-15 | Knowledge graph perfecting method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710833203.2A CN107748754B (en) | 2017-09-15 | 2017-09-15 | Knowledge graph perfecting method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107748754A true CN107748754A (en) | 2018-03-02 |
| CN107748754B CN107748754B (en) | 2021-10-26 |
Family
ID=61255746
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710833203.2A Active CN107748754B (en) | 2017-09-15 | 2017-09-15 | Knowledge graph perfecting method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107748754B (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108427749A (en) * | 2018-03-12 | 2018-08-21 | 上海哔哩哔哩科技有限公司 | Commercial goods labels management method, server and Commercial goods labels manage system |
| CN108710695A (en) * | 2018-05-23 | 2018-10-26 | 掌阅科技股份有限公司 | Mind map generation method based on e-book and electronic equipment |
| CN109145098A (en) * | 2018-07-20 | 2019-01-04 | 西北大学 | The Chinese culture element information searching method of knowledge based map |
| CN109189938A (en) * | 2018-08-31 | 2019-01-11 | 北京字节跳动网络技术有限公司 | Method and apparatus for updating knowledge mapping |
| CN109189937A (en) * | 2018-08-22 | 2019-01-11 | 阿里巴巴集团控股有限公司 | A kind of characteristic relation recommended method and device, a kind of calculating equipment and storage medium |
| CN109299287A (en) * | 2018-10-24 | 2019-02-01 | 深圳素问智能信息技术有限公司 | A kind of querying method and device of drinks information |
| CN109710776A (en) * | 2018-12-29 | 2019-05-03 | 中国科学技术大学 | The construction method of the knowledge mapping of photograph album |
| CN109840288A (en) * | 2019-02-21 | 2019-06-04 | 北京一品智尚信息科技有限公司 | Picture retrieval method, equipment and storage medium |
| CN110609903A (en) * | 2019-08-01 | 2019-12-24 | 华为技术有限公司 | Information presentation method and device |
| CN110737779A (en) * | 2019-09-18 | 2020-01-31 | 北京三快在线科技有限公司 | Knowledge graph construction method and device, storage medium and electronic equipment |
| CN110990584A (en) * | 2019-11-26 | 2020-04-10 | 口口相传(北京)网络技术有限公司 | Knowledge graph generation method and device |
| CN111143575A (en) * | 2019-12-06 | 2020-05-12 | 苏宁云计算有限公司 | Optimization method and system for store commodity structure |
| CN111738802A (en) * | 2020-06-30 | 2020-10-02 | 广东奥园奥买家电子商务有限公司 | Introduction method, device and equipment of E-commerce commodities |
| CN112102029A (en) * | 2020-08-20 | 2020-12-18 | 浙江大学 | Knowledge graph-based long-tail recommendation calculation method |
| CN112784062A (en) * | 2019-03-15 | 2021-05-11 | 北京金山数字娱乐科技有限公司 | Idiom knowledge graph construction method and device |
| CN113505245A (en) * | 2021-09-10 | 2021-10-15 | 深圳平安综合金融服务有限公司 | Knowledge graph generation method, computer readable storage medium and computer device |
| CN113535985A (en) * | 2021-08-17 | 2021-10-22 | 北京爱奇艺科技有限公司 | Object label determination method and device |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5241671A (en) * | 1989-10-26 | 1993-08-31 | Encyclopaedia Britannica, Inc. | Multimedia search system using a plurality of entry path means which indicate interrelatedness of information |
| WO2009117149A2 (en) * | 2008-03-19 | 2009-09-24 | Delve Networks, Inc. | Method and subsystem for information acquisition and aggregation to facilitate ontology and language-model generation within a content-search-service system |
| CN102254043A (en) * | 2011-08-17 | 2011-11-23 | 电子科技大学 | Semantic mapping-based clothing image retrieving method |
| CN104750795A (en) * | 2015-03-12 | 2015-07-01 | 北京云知声信息技术有限公司 | Intelligent semantic searching system and method |
| CN105550190A (en) * | 2015-06-26 | 2016-05-04 | 许昌学院 | Knowledge graph-oriented cross-media retrieval system |
| CN105912656A (en) * | 2016-04-07 | 2016-08-31 | 桂林电子科技大学 | Construction method of commodity knowledge graph |
| CN106355627A (en) * | 2015-07-16 | 2017-01-25 | 中国石油化工股份有限公司 | Method and system used for generating knowledge graphs |
| CN106708940A (en) * | 2016-11-11 | 2017-05-24 | 百度在线网络技术（北京）有限公司 | Method and device used for processing pictures |
| CN106815307A (en) * | 2016-12-16 | 2017-06-09 | 中国科学院自动化研究所 | Public Culture knowledge mapping platform and its use method |
-
2017
- 2017-09-15 CN CN201710833203.2A patent/CN107748754B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5241671A (en) * | 1989-10-26 | 1993-08-31 | Encyclopaedia Britannica, Inc. | Multimedia search system using a plurality of entry path means which indicate interrelatedness of information |
| US5241671C1 (en) * | 1989-10-26 | 2002-07-02 | Encyclopaedia Britannica Educa | Multimedia search system using a plurality of entry path means which indicate interrelatedness of information |
| WO2009117149A2 (en) * | 2008-03-19 | 2009-09-24 | Delve Networks, Inc. | Method and subsystem for information acquisition and aggregation to facilitate ontology and language-model generation within a content-search-service system |
| CN102254043A (en) * | 2011-08-17 | 2011-11-23 | 电子科技大学 | Semantic mapping-based clothing image retrieving method |
| CN104750795A (en) * | 2015-03-12 | 2015-07-01 | 北京云知声信息技术有限公司 | Intelligent semantic searching system and method |
| CN105550190A (en) * | 2015-06-26 | 2016-05-04 | 许昌学院 | Knowledge graph-oriented cross-media retrieval system |
| CN106355627A (en) * | 2015-07-16 | 2017-01-25 | 中国石油化工股份有限公司 | Method and system used for generating knowledge graphs |
| CN105912656A (en) * | 2016-04-07 | 2016-08-31 | 桂林电子科技大学 | Construction method of commodity knowledge graph |
| CN106708940A (en) * | 2016-11-11 | 2017-05-24 | 百度在线网络技术（北京）有限公司 | Method and device used for processing pictures |
| CN106815307A (en) * | 2016-12-16 | 2017-06-09 | 中国科学院自动化研究所 | Public Culture knowledge mapping platform and its use method |
Non-Patent Citations (1)
| Title |
|---|
| 徐增林等: "知识图谱技术综述", 《电子科技大学学报》 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108427749A (en) * | 2018-03-12 | 2018-08-21 | 上海哔哩哔哩科技有限公司 | Commercial goods labels management method, server and Commercial goods labels manage system |
| CN108427749B (en) * | 2018-03-12 | 2021-06-18 | 上海哔哩哔哩科技有限公司 | Commodity label management method, server and commodity label management system |
| CN108710695A (en) * | 2018-05-23 | 2018-10-26 | 掌阅科技股份有限公司 | Mind map generation method based on e-book and electronic equipment |
| CN108710695B (en) * | 2018-05-23 | 2019-08-06 | 掌阅科技股份有限公司 | Mind map generation method and electronic equipment based on e-book |
| CN109145098A (en) * | 2018-07-20 | 2019-01-04 | 西北大学 | The Chinese culture element information searching method of knowledge based map |
| CN109145098B (en) * | 2018-07-20 | 2021-10-29 | 西北大学 | Chinese culture element information searching method based on knowledge graph |
| US11244232B2 (en) | 2018-08-22 | 2022-02-08 | Advanced New Technologies Co., Ltd. | Feature relationship recommendation method, apparatus, computing device, and storage medium |
| CN109189937A (en) * | 2018-08-22 | 2019-01-11 | 阿里巴巴集团控股有限公司 | A kind of characteristic relation recommended method and device, a kind of calculating equipment and storage medium |
| CN109189938A (en) * | 2018-08-31 | 2019-01-11 | 北京字节跳动网络技术有限公司 | Method and apparatus for updating knowledge mapping |
| CN109299287A (en) * | 2018-10-24 | 2019-02-01 | 深圳素问智能信息技术有限公司 | A kind of querying method and device of drinks information |
| CN109710776A (en) * | 2018-12-29 | 2019-05-03 | 中国科学技术大学 | The construction method of the knowledge mapping of photograph album |
| CN109710776B (en) * | 2018-12-29 | 2022-10-28 | 中国科学技术大学 | Construction method of knowledge graph of photo album |
| CN109840288A (en) * | 2019-02-21 | 2019-06-04 | 北京一品智尚信息科技有限公司 | Picture retrieval method, equipment and storage medium |
| CN112784063A (en) * | 2019-03-15 | 2021-05-11 | 北京金山数字娱乐科技有限公司 | Idiom knowledge graph construction method and device |
| CN112784062A (en) * | 2019-03-15 | 2021-05-11 | 北京金山数字娱乐科技有限公司 | Idiom knowledge graph construction method and device |
| CN110609903A (en) * | 2019-08-01 | 2019-12-24 | 华为技术有限公司 | Information presentation method and device |
| CN110737779B (en) * | 2019-09-18 | 2021-05-04 | 北京三快在线科技有限公司 | Knowledge graph construction method and device, storage medium and electronic equipment |
| CN110737779A (en) * | 2019-09-18 | 2020-01-31 | 北京三快在线科技有限公司 | Knowledge graph construction method and device, storage medium and electronic equipment |
| CN110990584B (en) * | 2019-11-26 | 2021-02-09 | 口口相传(北京)网络技术有限公司 | Knowledge graph generation method and device |
| CN110990584A (en) * | 2019-11-26 | 2020-04-10 | 口口相传(北京)网络技术有限公司 | Knowledge graph generation method and device |
| CN111143575A (en) * | 2019-12-06 | 2020-05-12 | 苏宁云计算有限公司 | Optimization method and system for store commodity structure |
| CN111143575B (en) * | 2019-12-06 | 2023-07-04 | 苏宁云计算有限公司 | Optimization method and system for store commodity structure |
| CN111738802A (en) * | 2020-06-30 | 2020-10-02 | 广东奥园奥买家电子商务有限公司 | Introduction method, device and equipment of E-commerce commodities |
| CN112102029A (en) * | 2020-08-20 | 2020-12-18 | 浙江大学 | Knowledge graph-based long-tail recommendation calculation method |
| CN113535985A (en) * | 2021-08-17 | 2021-10-22 | 北京爱奇艺科技有限公司 | Object label determination method and device |
| CN113505245A (en) * | 2021-09-10 | 2021-10-15 | 深圳平安综合金融服务有限公司 | Knowledge graph generation method, computer readable storage medium and computer device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107748754B (en) | 2021-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107748754A (en) | A kind of knowledge mapping improving method and device | |
| CN107833082A (en) | A kind of recommendation method and apparatus of commodity picture | |
| KR102075833B1 (en) | Curation method and system for recommending of art contents | |
| TWI631474B (en) | Method and device for product identification label and method for product navigation | |
| CN110263180B (en) | Intention knowledge graph generation method, intention identification method and device | |
| CN109815308A (en) | The determination of intention assessment model and retrieval intension recognizing method, device | |
| CN103927309B (en) | A kind of method and device to business object markup information label | |
| CN107861971A (en) | A kind of product search method and device | |
| Shi et al. | The exploration of artificial intelligence application in fashion trend forecasting | |
| Fuad et al. | Analysis and classification of mobile apps using topic modeling: A case study on Google Play Arabic apps | |
| Wang et al. | Leveraging image-processing techniques for empirical Research: Feasibility and Reliability in Online Shopping Context | |
| Shi et al. | Using artificial intelligence to analyze fashion trends | |
| Chen et al. | iLike: integrating visual and textual features for vertical search | |
| Bogdanov et al. | A content-based system for music recommendation and visualization of user preferences working on semantic notions | |
| Kang et al. | Recognising informative Web page blocks using visual segmentation for efficient information extraction. | |
| CN107730357A (en) | A kind of view-based access control model dictionary realizes the method and system of image quick-searching | |
| Li et al. | Sparse representation based visual element analysis | |
| CN113744019A (en) | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and storage medium | |
| Chishti et al. | Identify Website Personality by Using Unsupervised Learning Based on Quantitative Website Elements | |
| Zhang | Research on data mining security under the background of big data era | |
| CN111723566A (en) | Method and device for reconstructing product information | |
| CN113342920B (en) | Cataloging automation method and system for content resources | |
| Yang et al. | Zero-shot transfer learning based on visual and textual resemblance | |
| Chai et al. | A unified framework for micro-video background music automatic matching | |
| Dang-Nguyen et al. | Supervised models for multimodal image retrieval based on visual, semantic and geographic information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230106 Address after: 510123 building 6, No. 20, Huahai street, Fangcun, Liwan District, Guangzhou, Guangdong (office only) Patentee after: GUANGZHOU PINWEI SOFTWARE Co.,Ltd. Address before: 510000 room 01, No.314, Fangcun Avenue middle, Liwan District, Guangzhou City, Guangdong Province Patentee before: GUANGZHOU WEIPINHUI RESEARCH INSTITUTE CO.,LTD. |
|
| TR01 | Transfer of patent right |