CN110232141A - Resource acquiring method, resource acquisition device, storage medium and electronic equipment - Google Patents
Resource acquiring method, resource acquisition device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN110232141A CN110232141A CN201910472052.1A CN201910472052A CN110232141A CN 110232141 A CN110232141 A CN 110232141A CN 201910472052 A CN201910472052 A CN 201910472052A CN 110232141 A CN110232141 A CN 110232141A
- Authority
- CN
- China
- Prior art keywords
- resource
- feature data
- text
- image
- vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3347—Query execution using vector based model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/56—Information retrieval; Database structures therefor; File system structures therefor of still image data having vectorial format
-
- 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
- G06F16/901—Indexing; Data structures therefor; Storage structures
-
- 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
- G06F16/903—Querying
- G06F16/90335—Query processing
-
- 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
- G06F16/951—Indexing; Web crawling techniques
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the present application provides a kind of resource acquiring method, resource acquisition device, storage medium and electronic equipment, and resource acquiring method includes: when detecting the operational order for image, and identification and described image be associated, predeterminated position text information;The corresponding text feature data of the text information are obtained, and obtain the corresponding image feature data of described image;Associated resource is obtained according to the text feature data and described image characteristic.Using aforementioned operation, retrieval element of the picture material, content of text messages recomposition obtained without user according to understanding for retrieval, therefore user can be reduced to the selection operation of specific content, the switching of retrieval window and the input equiprobable operation of text, improve the use feeling of user.
Description
Technical field
This application involves data retrieval technology fields, and in particular to a kind of information push resource acquiring method, resource acquisition
Device, storage medium and electronic equipment.
Background technique
During using contents such as instant message applications chat, reading mail or notices, user may have needle
The demand of coordinate indexing is carried out to related content.
For example, such as two friend's chats, it is desirable to go to Mount Taishan to climb the mountain, and A wants to buy an alpin-stock, at this time A also needs to exit
With the interface of B chat, then the APP interface queries alpin-stock that goes shopping, operation is especially cumbersome, when A and B thinks that the about especially close time goes
When climbing the mountain, bought in shopping website, it, may be before setting out due to the demand of express delivery, alpin-stock does not dispense simultaneously, and at this moment user has
The possible shop that website screening express delivery can be dispensed into before setting out that must go shopping must either go for the quotient near Mount Taishan
?.
Summary of the invention
The embodiment of the present application provides a kind of resource acquiring method, resource acquisition device, to avoid user according to background technique
The step of referring to is operated, and the efficiency that the retrieval of user's chosen content obtains resource is improved.
On the one hand, the embodiment of the present application provides a kind of resource acquiring method, comprising:
When detecting the operational order for image, identification and described image be associated, predeterminated position text information;
The corresponding text feature data of the text information are obtained, and obtain the corresponding characteristics of image number of described image
According to;
Associated resource is obtained according to the text feature data and described image characteristic.
On the other hand, the embodiment of the present application provides a kind of resource acquiring method, comprising:
When detecting the operational order for the first object, the corresponding fisrt feature data of first object are obtained, with
And obtain the corresponding second feature data of the second object;
Based on the fisrt feature data and the associated resource of second feature data query;
Wherein: first object is associated with second object context.
In another aspect, the embodiment of the present application provides a kind of resource acquisition device, comprising:
Text identification unit, for identifying associated, pre- with described image when detecting the operational order for image
If the text information of position;
Characteristic acquiring unit, for obtaining the corresponding text feature data of the text information, and described in obtaining
The corresponding image feature data of image;
Resource acquisition unit, for obtaining associated money according to the text feature data and described image characteristic
Source.
In another aspect, the embodiment of the present application provides a kind of resource acquisition device, comprising:
Characteristic acquiring unit, for obtaining described first when detecting operational order of the user to the first object
The corresponding fisrt feature data of object, and obtain the corresponding second feature data of the second object;
Resource acquisition unit, for being based on the fisrt feature data and the associated money of second feature data query
Source;
Wherein: first object is associated with second object context.
The embodiment of the present application also provides a kind of storage medium, and the storage medium is stored with a plurality of instruction, and described instruction is suitable
For being loaded by processor and executing foregoing resource acquiring method.
The embodiment of the present application also provides a kind of electronic equipment, including memory and processor;
The memory is stored with a plurality of instruction;Described instruction is suitable for being loaded by the processor and being executed as previously described
Resource acquiring method.
Using resource acquiring method provided by the embodiments of the present application, in a certain image of user's click, attempt to trigger accordingly
It is automatic together to obtain and image is associated, text information of predetermined position when search operaqtion, then generated using text information
Text feature data;After extracting the image feature data of image, text feature data and characteristics of image number are automatically utilized
Corresponding resource is retrieved according to the retrieval element formed for retrieval, and using retrieval element.Using aforementioned operation, it is not necessarily to user
The picture material that is obtained according to understanding, content of text messages recomposition are used to retrieve the retrieval elements of specific resources, therefore can be with
User is reduced to the selection operation of specific content, the switching of retrieval window and the input equiprobable operation of text, improves use
The use feeling at family.
Detailed description of the invention
Fig. 1 is the flow chart of resource acquiring method provided by one embodiment of the present invention;
Fig. 2 is the flow chart for the resource acquiring method that another embodiment of the present invention provides;
Fig. 3 is the flow chart for the resource acquiring method that further embodiment of the present invention provides;
Fig. 4 is a kind of structural schematic diagram of resource acquisition device provided in an embodiment of the present invention;
Fig. 5 be another embodiment of the present invention provides resource acquisition device structural schematic diagram;
Wherein: 11- text identification unit, 12- characteristic acquiring unit, 13- resource acquisition unit, 21- characteristic
Acquiring unit, 22- resource acquisition unit.
Specific embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, part relevant to related invention is illustrated only in attached drawing.
Content hereinafter in order to facilitate understanding, to resource acquiring method provided by the embodiments of the present application and resource acquisition device
Before being specifically described, the embodiment of the present application is likely to occur first a scene and currently grasped in the lower user of this scene
It introduces.
In one scenario, immediate communication tool chatting of the user by installation in the electronic device, interlocutor is discontinuously
One or more of message are had sent to user, aforementioned messages may include for picture, text information, comprising the figure of content of text
Piece, voice messaging or video clip, and it is specific to user's reception and registration one to attempt to use one or more of message above-mentioned
Information.
After reading the one or more of message that interlocutor sends, user can be inferred that the letter of characteristic therein
Breath;After obtaining aforementioned information, the possible information of user, which is further obtained, may think root with the resource of information association, such as user
It is believed that breath searches the path for reaching a certain address, type in restaurant etc. in some region;At this point, aforementioned one or more of message
Because not being directly linked together, user can not rapidly be obtained and information association using click message, triggering retrieval
Resource, and intelligence triggers retrieval by being entered information into retrieval window manually again, obtains resource;For example, passing through in information
In the case that picture and text cooperation indicate, user can only be manually entered the information of text and picture reflection in retrieval window,
Obtain corresponding correlated resources.
It is contemplated that because user needs the input in retrieval window that can determine resource, for the information of retrieval,
It is relatively complicated to need to switch operation, the operating process such as software interface, copy stickup or typewriting input.
Several embodiments of the application provide the method for simplified user's operation, when user triggers a certain message content,
I.e. automatically extension needs the content retrieved, and generates the retrieval data for indicating that user may be intended to, and utilize aforementioned retrieval data
Realize the retrieval of resource.
It should be noted that the electronic equipment in aforementioned applications can be smart phone etc. using the equipment of touch screen, it can also be with
It is the equipment that traditional desktop computer etc. needs mouse or keyboard to input, without limitation to this embodiment of the present application.
Embodiment one
Fig. 1 is the flow chart of resource acquiring method provided by one embodiment of the present invention.As shown in Figure 1, the present invention is implemented
The resource acquiring method that example provides includes step S101-S103.
S101: when detecting the operational order for image, identification and image be associated, text information of predeterminated position.
By the electronic equipment that user uses for smart phone.The part that touch screen shows image is touched in user, is made
After the touch screen of electronic equipment generates trigger signal, electronic equipment determines the position of user's point touching screen according to trigger signal,
And according to the determination of the type of action of position and user's point touching screen be the operational order to image, subsequent electronic equipment starts pair
Image and associated text information are handled.
After detecting to the operational order of image, electronic equipment is searched and the associated text information of image, text information
Positioned at predeterminated position;Predeterminated position above-mentioned can be the predeterminated position within the scope of picture, be also possible to except picture range, with
Image has certain display area of context relation or the position of certain character zone;In addition, text information be also possible to be with
Hyperlink is associated, the text information being stored at certain remote servers or electronic equipment other positions.
Text information can be the information indicated in the form of literal code information, it is also possible to the shape of dot matrix picture region
The information that formula indicates, the embodiment of the present application are not also particularly limited.
In the case where text information is indicated using literal code form, identification is with the associated text information of image i.e. to pre-
If the text of position carries out selected operation;In the case where text information is indicated using dot matrix picture, identify associated with image
Text information is to carry out ORC identification to dot matrix picture region, to obtain the corresponding text of text that dot matrix information table shows in picture
Encoded information, and selected operation is carried out to literal code information.
S102: obtaining the corresponding text feature data of the text information, obtains the corresponding image feature data of image.
When detecting operational order of the user to image, electronic equipment is handled image, or image is sent out
It gives and image is handled with other equipment of electronic equipment communication connection, by other equipment, generating indicates picture material
Image feature data.For example, the image that user clicks is the image that a width has word content, indicates Fragrance Hill somewhere landscape;
After handling image, i.e., the landscape content part in image is identified, mark picture material can be obtained and shown
Address is the image feature data of " Fragrance Hill ", " Haidian District, Beijing City dealing street 40 " this address.
After obtaining text information, electronic equipment or the other equipment communicated to connect with electronic equipment are i.e. to text information
It is handled, generates the text feature data for indicating content of text messages.
For example, there is the text information of " going to climb the mountain " in predetermined position in the aforementioned picture for indicating Fragrance Hill landscape,
After obtaining aforementioned " going to climb the mountain " corresponding text information, it can handle text information, identify text feature data
It is " climbing the mountain ".
S103: associated resource is obtained according to text feature data and image feature data.
After obtaining the image feature data and the corresponding text feature data of text information that indicate picture material respectively
To carry out integrated treatment by text feature data and image data, form the retrieval data for retrieval, and utilize this retrieval
Data retrieve corresponding resource database, obtain and text feature data and the associated resource of image feature data.
For example, connecting the example mentioned in abovementioned steps, the address that image feature data is " Fragrance Hill ", Fragrance Hill is being obtained
It is special to image above-mentioned after the text feature data " climbing the mountain " of " Haidian District, Beijing City dealing street 40 " and expression text information
Sign data and text feature data are combined, it can obtaining content representation is " to remove the perfume (or spice) of Haidian District, Beijing City dealing street 40
Climb the mountain on mountain " retrieval data, retrieved using this retrieval data, then available relevant resource, relevant resource can
It can include removing " Haidian District, Beijing City dealing street 40 " navigation circuit of this address, the ticket information in Fragrance Hill, near Fragrance Hill
The resources such as road condition information and the environmental aspect in nearest Fragrance Hill.
Then, aforementioned resource is shown using electronic equipment and is exported, so that user gets resource above-mentioned.
Using resource acquiring method provided by the embodiments of the present application, in a certain image of user's click, attempt to trigger accordingly
It is automatic together to obtain and image is associated, text information of predetermined position when search operaqtion, then generated using text information
Text feature data;After extracting the image feature data of image, text feature data and characteristics of image number are automatically utilized
Corresponding resource is retrieved according to the retrieval element formed for retrieval, and using retrieval element.Using aforementioned operation, it is not necessarily to user
The picture material that is obtained according to understanding, content of text messages recomposition are used to retrieve the retrieval elements of specific resources, therefore can be with
User is reduced to the selection operation of specific content, the switching of retrieval window and the input equiprobable operation of text, improves use
The use feeling at family.
It should be noted that " obtaining the corresponding image feature data of image " in actual implementation, in abovementioned steps S102
Can directly be executed after detecting the operational order for image, i.e., aforementioned " obtaining the corresponding image feature data of image " with
Identification text feature data corresponding with the text information of the associated predeterminated position of image and acquisition text information can be arranged side by side
It executes.
It shall also be noted that " Fragrance Hill " above-mentioned, " Haidian District, Beijing City dealing street 40 " and " climbing the mountain " is only used for
With the linguistic interpretation text feature data and image feature data of image, in practical application, text feature data and characteristics of image
Data may not be aforementioned character representation, but use other kinds of character representation.
Embodiment two
Fig. 2 is the flow chart for the resource acquiring method that another embodiment of the present invention provides.As indicated with 2, the application is implemented
The resource acquiring method that example provides includes step S201-S207.
S201: when detecting the operational order for image, identification and image be associated, text envelope of predetermined position
Breath.
For ease of understanding, the embodiment of the present application fabricates a scene: assuming that user is soft using the communication in instant smart phone
Part and interlocutor, which chat, to link up;Chat person sends a text message " this afternoon goes to climb the mountain " to user first, subsequent user
It has replied one " where, do not have climbing boot ", chat person has sent a figure that is typical, indicating Fragrance Hill landscape then to user
Picture.
It is directed to after smart phone detects the operational order for image in the touch screen that user clicks smart phone
Image predetermined position is inquired, and the text information of predetermined position is obtained.In the embodiment of the present application, the text of predetermined position
This information is the text information in context, and predeterminated position is set to 100 alphabetic characters of context, therefore, is obtained
Text information be " this afternoon goes to climb the mountain ", " where, do not have climbing boot ".
S202: natural language processing is carried out to text information, obtains the name entity in text information.
After getting text information, natural language processing is carried out for text information, i.e., obtains text using segmenting method
Name entity in this information.
For example, it is directed to " this afternoon goes to climb the mountain ", " where, do not have climbing boot " carry out word segmentation processing, available life
Name entity includes " today ", " afternoon ", " climbing the mountain ", " climbing boot " etc..
S203: the corresponding intention of text information is obtained based on name entity.
Then, after according to " today " above-mentioned, " afternoon ", " climbing the mountain " and " climbing boot ", according to aforementioned name object query
The default database limited can determine the sportswear for being intended that purchase such as " climbing boot " one kind of user in greater probability
It is standby, it is thus determined that user's is intended that " athletic equipment of the purchase for climbing the mountain before climbing the mountain this afternoon ".
S204: the corresponding Text eigenvector of text information is obtained based on intention.
In the embodiment of the present application, using Text eigenvector as text feature data above-mentioned.Determining text information pair
After that answers is intended that " before climbing the mountain this afternoon purchase athletic equipment " for climbing the mountain, it can according to " being climbed in this afternoon
The athletic equipment for climbing the mountain is bought in front of the mountains " and it is such as raw according to temporal information, the possible address information in aforementioned intention
At the corresponding Text eigenvector of aforementioned texts information.Text eigenvector can be the feature formed using Text Flag herein
Vector is also possible to the feature vector using other identifier composition, and the application is simultaneously not specially limited, as long as it can be embodied
" athletic equipment of the purchase for climbing the mountain before this afternoon ".
S205: the corresponding image feature vector of image is obtained.
In the embodiment of the present application, using image feature vector as image feature data above-mentioned.
The corresponding image feature vector of image is obtained, is that dissection process is carried out to image, extracts the notable feature in image
Information, and image feature vector is formed using notable feature information.
Still by taking the aforementioned image for indicating " Fragrance Hill " as an example, after handling image, image feature vector is produced,
And having in image feature vector indicates " Fragrance Hill ", Fragrance Hill address " Haidian District, Beijing City dealing street 40 " or similar data
Mark
After the execution for completing S204 and S205, S206 is executed.
S206: resource vector is generated according to Text eigenvector and image feature vector.
Resource vector is generated according to Text eigenvector and image feature vector, can be Text eigenvector and image
Feature vector is spliced, and a resource vector is obtained;The aforementioned splicing to Text eigenvector and image feature vector, Ke Yili
Solution carries out the element in Text eigenvector and image feature vector to require according to attribute of an element each in resource vector
Selection forms resource vector.
For example, obtained resource vector can be expression and " meet and go to Fragrance Hill to climb the mountain this afternoon, need according to previous example
The vector of athletic equipment of the purchase for climbing the mountain before climbing the mountain in the afternoon ".
S207: associated resource is obtained according to resource vector.
On the basis of determining resource vector, electronic equipment can retrieve locally or remotely database based on resource vector
In can satisfy the resource of the purpose that aforementioned resource vector indicates.
For example, being to indicate " to meet and go to Fragrance Hill to climb the mountain this afternoon, purchase is used before needing to climb the mountain in the afternoon in resource vector
In the athletic equipment climbed the mountain " in the case where, resource can be the navigation routine in Fragrance Hill, and in the certain model in navigation routine periphery
The address information of sport equipment store in a certain range near the address information or Fragrance Hill of interior sport equipment store is enclosed, or
Person is a certain range of sport equipment store information of user current location, even can satisfy the delivery before this afternoon
To the information in the network movement equipment shop of user current location.
Using resource acquiring method provided by the embodiments of the present application, after user clicks a certain image, electronic equipment is together
The text information of image context is obtained, and extracts the intention of text representation using text information, and text is generated according to intention
Feature vector;In addition, electronic equipment, which carries out processing to image, obtains image feature vector.It is special using image feature vector and text
After sign vector combines and to form resource vector, it can using resource vector as the corresponding resource of retrieval elements acquisition.Because
User click image while, electronic equipment together automatically obtain image context text information, generate text feature to
The process of amount, image feature vector and final resource vector has electronic equipment or other processing equipments to realize together,
Selection of the user to context text information can be reduced, the operation such as switching retrieval window improves the use feeling of user.
In the embodiment of the present application, generating resource vector process according to Text eigenvector and image feature vector is to text
Eigen vector sum image feature vector carries out splicing, in other embodiments, special according to Text eigenvector and image
The process that sign vector generates resource vector is also possible to: by the element of Text eigenvector and image feature vector corresponding position
It is added or is multiplied and obtain the element of corresponding position, obtaining resource vector according to aforementioned elements;In the previous case, text
Number of elements in feature vector and image feature vector is preferably identical, also, Text eigenvector and image feature vector pair
Answer the element property at position identical.
Embodiment three
Fig. 3 is the flow chart for the resource acquiring method that further embodiment of the present invention provides.As described in Figure 3, the application is real
The resource acquiring method for applying example offer includes step S301-S30N.
S301: when detecting the operational order for the first object, obtaining the corresponding fisrt feature data of the first object, with
And obtain the corresponding second feature data of the second object.
In the embodiment of the present application, the first object can be the associated various types of objects of content that electronic equipment is shown;
It may be text, image, audio fragment or video clip in concrete application.
After detecting the operational order for the first object, electronic equipment can be handled for the first object,
Extract the corresponding fisrt feature data of the first object.
Meanwhile electronic equipment can also search the second object, the second object is and the associated object of the first object context;This
Apply embodiment in, the second object be also possible to it is that electronic equipment show, except the first object display the associated object of content its
Its various object, the second object are also possible to text, image, audio fragment or video clip, and the quantity of the second object can be with
It is one, is also possible to multiple.
For ease of understanding, the embodiment of the present application fabricates a scene: assuming that user is soft using the communication in instant smart phone
Part and interlocutor, which chat, to link up;Chat person first to user send a text message " climb the mountain, Fragrance Hill, how? ", subsequent user
Replied one " when ", interlocutor then replys " how this afternoon first buys athletic equipment? ".
When the user clicks " how this afternoon first buys athletic equipment? " (this text is that the first object above-mentioned is corresponding
Display content) after, electronic equipment detects the operational order to the first object, then extracts the first object, and inquires
The relevant context of first object, find " it climbs the mountain, Fragrance Hill, how? " as the second object.
Then, electronic equipment or with electronic equipment communication connection other equipment to the first object and the second object at
Reason, respectively obtains fisrt feature data and second feature data;Then, S303 is executed.
S302: fisrt feature data and the associated resource of second feature data query are based on.
After obtaining fisrt feature data and second feature data, electronic equipment or its with electronic equipment communication connection
He is based on fisrt feature data and second feature data at equipment, is associated to resource database, determining while and fisrt feature
The resource of data and second feature data correlation.
Continued access previous example, fisrt feature data and second feature data are respectively to indicate that content is " this afternoon, perfume (or spice)
The characteristic of mountain, athletic equipment " and " climbing the mountain, Fragrance Hill ", it is available about can be under today according to fisrt feature data
Noon goes to Fragrance Hill to climb the mountain and buys the relevant resource of athletic equipment, such as a certain range of sports shops position of user current location
Information;And according to the second feature data available navigation routines about current location to Fragrance Hill, estimated Xiangshan District
Meteorological condition, the ticket information in Fragrance Hill, sport equipment store information near Fragrance Hill etc., in summary, available Fragrance Hill is attached
The resource of close sport equipment store.
Using resource acquiring method provided by the embodiments of the present application, user is in selected first object, it is intended that retrieval is corresponding
When resource, the second object of the first Object Query context that can be selected according to user;In practical application, the first object and
Two objects have the probability of strong incidence relation very big, thus using the fisrt feature data of the first object and the second object acquisition and
The association probability that the resource of second feature data combined retrieval and user are intended to the resource obtained is higher, can using method above-mentioned
To improve the hit probability of query resource.
In addition, after selected first object, being directly based upon the first Object Selection may using resource acquiring method above-mentioned
Associated second object so that user is not necessarily to obtain two objects respectively to retrieve resource, or switches to searched page, will
The corresponding information input of two objects is retrieved into frame retrieval, improves the use feeling of user.
In the aforementioned embodiment, the first object and the second object are text object.In one embodiment, first is obtained
The method of the corresponding fisrt feature data of object can be executed according to step S401-S403.
S401: carrying out natural language processing to the first object, obtains the first name entity in the first object.
After getting the first object, natural language processing is carried out for the first object, i.e., obtains the using segmenting method
Name entity in an object.
For example, for " how this afternoon, Fragrance Hill first buy athletic equipment? " carry out word segmentation processing, available life
Name entity includes " today ", " afternoon ", " Fragrance Hill ", " buying ", " athletic equipment ".
S402: the corresponding first intention of the first object is obtained based on the first name entity.
For example, being directed to " today " above-mentioned, " afternoon ", " Fragrance Hill ", " buying ", " athletic equipment " available first intention
Athletic equipment is bought for this afternoon.
S403: the corresponding first eigenvector of the first object is obtained based on first intention.
First eigenvector is fisrt feature data above-mentioned.In the embodiment of the present application, first eigenvector be can be
One include 128 elements feature vector, expression first intention that can be detailed-oriented.
It is similar with the method for the corresponding fisrt feature data of the first object of aforementioned acquisition, obtain the second object corresponding the
The method of two feature vectors can be executed according to step S501-S503.
S501: carrying out natural language processing to the second object, obtains the second name entity in the second object.
For example, for " it climbs the mountain, Fragrance Hill, how? " how climb the mountain? ", available name entity include " climbing the mountain ",
" Fragrance Hill ".
S502: the corresponding second intention of the second object is obtained based on the second name entity.
For example, being directed to " climbing the mountain " above-mentioned and " Fragrance Hill ", available second intention is to climb Fragrance Hill.
S503: the corresponding second feature vector of the second object is obtained based on second intention.
Second feature vector is second feature data above-mentioned.In the embodiment of the present application, second feature vector be can be
One include 128 elements feature vector, expression second intention that can be detailed-oriented.
In the embodiment of the present application, the first object and the second object are text object, according to the first object and the second object
Corresponding characteristic is obtained, is to carry out word segmentation processing to the first object and the second object to obtain name entity and subsequent spy
Levy data.
In other embodiments, the first object and the second object may be audio fragment, and it is right first that processing is carried out to it
Audio fragment carries out identification operation, determines content of text therein, then carry out word segmentation processing to content of text and obtain characteristic.
In other embodiments, the first object and the second object may be image, and processing is carried out to it is extracted in image
Significant characteristics, and using significant characteristics as characteristic.
In other embodiments, the first object and the second object are also possible to be video clip, and it is to extract that processing is carried out to it
Key frame in video clip extracts the significant characteristics in key, and using significant characteristics as characteristic.
It in the embodiment of the present application, is respectively first eigenvector and second in fisrt feature data and second feature data
It include step S601- according to fisrt feature data and the associated resource of second feature data acquisition in the case where feature vector
S602。
S601: resource vector is formed according to first eigenvector and second feature vector.
S602: associated resource is obtained according to resource vector.
In aforementioned S601, include: according to the possibility mode that first eigenvector and second feature vector form resource vector
(1) first eigenvector is added to the member as resource vector corresponding position with the element of second feature vector corresponding position
Element obtains resource vector;(2) using first eigenvector and the element multiplication of second feature vector corresponding position as resource to
The element for measuring corresponding position, obtains resource vector;(3) first eigenvector and second feature vector are spliced, forms resource
Vector;The wherein splicing of (3) to resource vector, it may be possible to the Partial Elements in selected first eigenvector and second feature vector
Spliced, it is also possible to regard all elements of first eigenvector and second feature vector as vector element, form money
Source vector.
In the embodiment of the present application, step S301 obtains the corresponding fisrt feature data of the first object, and obtains second pair
As there are two types of possible modes, respectively step S3011-S3012 and S3013-S3015 for corresponding second feature data.
First way
S3011: while obtaining the first object and the second object.
Herein, the first object and the second object are obtained while described, are for subsequent to the first object and the second object
For processing;In practical application, it may be possible to after user selectes the first object, be based on the first object retrieval context relation
Content obtains the second object.
S3012: the first object of parsing and obtain corresponding fisrt feature data, parse the second object and obtain corresponding the
Two characteristics.
Herein, according to the difference of electronic equipment or associated processing equipment, the process of the first object and the second object is parsed
It may be parallel, it is also possible to successively successively carry out.
The second way
S3013: the corresponding fisrt feature data of the first object are obtained.
S3014: judge whether fisrt feature data include scheduled resource data;If so, executing S2035;If it is not, executing
S2036。
S3015: the associated resource of fisrt feature data acquisition is used.
S3016: searching the second object, obtains the corresponding second feature data of the second object.
In the second way, there is no obtaining directly to search the second object, but it is being directed to the first object corresponding the
One characteristic is analyzed, and judges whether fisrt feature data include the data for being sufficiently used for retrieval resource;If it is not, searching again
Whether second have the second object, and second feature data are obtained in the case where there is the second object.
For example, in one example, if the first object is " we go to Fragrance Hill to climb the mountain two 10 points of this morning ", passing through
Analysis has included that personage, time, place and movement are several for characterize the characteristics of intention, then can determine the
The resource data of one characteristic has been sufficiently used for obtaining specific resource, without searching the second object;And at second
In example, if the first object is " our today, two 10 AMs went to open air ", in addition to personage, time and substantially
It acts outside the characteristic being intended to, the characteristic not indicating place and specifically acting, therefore can not be directly used in and look into
It askes and is intended to corresponding resource, therefore search the second object " climbing the mountain, Fragrance Hill ", determine that the second of specific place and type of action is special
Data are levied, then can use fisrt feature data and the specific resource of second feature data search.
Other than providing resource acquiring method above-mentioned, the embodiment of the present application also provides a kind of resource acquisition device.
Fig. 4 is a kind of structural schematic diagram for resource acquisition device that embodiment provides.As shown in figure 4, at the application one
In embodiment, resource acquisition device includes text identification unit 11, characteristic acquiring unit 12 and resource acquisition unit 13.
Text identification unit 11, for when detecting the operational order for image, identification and described image be associated,
The text information of predeterminated position;
Characteristic acquiring unit 12, for obtaining the corresponding text feature data of the text information, and acquisition institute
State the corresponding image feature data of image;
Resource acquisition unit 13, for obtaining associated money according to the text feature data and described image characteristic
Source.
Further, the corresponding text feature data of the acquisition of characteristic acquiring unit 12 text information include:
Natural language processing is carried out to the text information, obtains the name entity in the text information;
The corresponding intention of the text information is obtained based on the name entity;
The corresponding Text eigenvector of the text information is obtained based on the intention;Using the Text eigenvector as
The text feature data;
Corresponding, characteristic obtains the big warm corresponding image feature data of image that obtains and includes:
Obtain the corresponding image feature vector of described image;Using described image feature vector as described image characteristic
According to;
Associated resource is obtained according to the text feature data and described image characteristic, comprising:
Resource vector is formed according to the Text eigenvector and described image feature vector;
The associated resource is obtained according to the resource vector.
Further, resource acquisition unit 13 forms resource vector packet according to Text eigenvector and image feature vector
It includes:
The value that the Text eigenvector is added or is multiplied with the element of described image feature vector corresponding position
As the element of resource vector corresponding position, the resource vector is obtained;Alternatively,
Splice the Text eigenvector and described image feature vector, obtains the resource vector.
Fig. 5 is the structural schematic diagram for the resource acquisition device that another embodiment of the application provides.As shown in figure 5, in addition
In one embodiment, resource acquisition unit includes characteristic acquiring unit 21 and resource acquisition unit 22.
Wherein, characteristic acquiring unit 21 is used for when detecting operational order of the user to the first object, obtains institute
The corresponding fisrt feature data of the first object are stated, and obtain the corresponding second feature data of the second object;
Resource acquisition unit 22 is used to be based on the fisrt feature data and the associated money of second feature data query
Source;Wherein: first object is associated with second object context.
It is in a preferred embodiment, described to obtain the corresponding fisrt feature data of first object, comprising:
Natural language processing is carried out to first object, obtains the first name entity in first object;
The corresponding first intention of first object is obtained based on the first name entity;
The corresponding first eigenvector of first object is obtained based on the first intention;By the fisrt feature to
Amount is used as the fisrt feature data;
The corresponding second feature data of the second object of the acquisition, comprising:
Obtain the second name entity in second object;
The corresponding second intention of second object is obtained based on the second name entity;
The corresponding second feature vector of second object is obtained based on the second intention;By the second feature vector
As the second feature data.
It is described to obtain the corresponding fisrt feature data of first object in a kind of scheme, comprising:
Natural language processing is carried out to first object, obtains the first name entity in first object;
The corresponding first intention of first object is obtained based on the first name entity;
The corresponding first eigenvector of first object is obtained based on the first intention;By the fisrt feature to
Amount is used as the fisrt feature data;
The corresponding second feature data of the second object of the acquisition, comprising:
Obtain the second name entity in second object;
The corresponding second intention of second object is obtained based on the second name entity;
The corresponding second feature vector of second object is obtained based on the second intention;By the second feature vector
As the second feature data.
Optionally, described to be based on the fisrt feature data and the second feature data acquisition in resource acquisition device
Associated resource, comprising:
Resource vector is formed according to the first eigenvector and the second feature vector;
The associated resource is obtained according to the resource vector.
Optionally, in resource acquisition device, resource is formed according to the first eigenvector and the second feature vector
Vector, comprising:
The value that the first eigenvector is added or is multiplied with the element of second feature vector corresponding position
As the element of resource vector corresponding position, the resource vector is obtained;Alternatively,
Splice the first eigenvector and the second feature vector, forms the resource vector.
Optionally, described to obtain the corresponding fisrt feature data of first object in resource acquisition device, and obtain
The corresponding second feature data of second object, comprising:
First object and second object are obtained simultaneously;
It parses first object and obtains the corresponding fisrt feature data, parse second object and obtain pair
The second feature data answered.
Optionally, described to obtain the corresponding fisrt feature data of first object in resource acquisition device, and obtain
The corresponding second feature data of second object, comprising:
Obtain the corresponding fisrt feature data of the first object;
Judge whether the fisrt feature data include scheduled resource characteristic data;
If it is not, searching second object, the corresponding second feature data of second object are obtained.
The embodiment of the present application provides a kind of medium, and media storage has item instruction, and said instruction can be used for being added by processor
It carries and executes resource acquiring method above-mentioned.
The embodiment of the present application also provides a kind of electronic equipment, and electronic equipment includes storing itself and processor;Memory storage
There is a plurality of instruction;Said instruction is suitable for having processor to load and executing resource acquiring method above-mentioned.
A1. the embodiment of the present application provides a kind of resource acquiring method, and the resource acquiring method includes:
When detecting the operational order for image, identification and described image be associated, predeterminated position text information;
The corresponding text feature data of the text information are obtained, and obtain the corresponding characteristics of image number of described image
According to;
Associated resource is obtained according to the text feature data and described image characteristic.
A2. information-pushing method according to a1,
It is described to obtain the corresponding text feature data of the text information, comprising:
Natural language processing is carried out to the text information, obtains the name entity in the text information;
The corresponding intention of the text information is obtained based on the name entity;
The corresponding Text eigenvector of the text information is obtained based on the intention;Using the Text eigenvector as
The text feature data;
The corresponding image feature data of the acquisition described image, comprising:
Obtain the corresponding image feature vector of described image;Using described image feature vector as described image characteristic
According to;
Associated resource is obtained according to the text feature data and described image characteristic, comprising:
Resource vector is formed according to the Text eigenvector and described image feature vector;
The associated resource is obtained according to the resource vector.
A3. the information-pushing method according to A2, according to the Text eigenvector and described image feature vector shape
At resource vector, comprising:
The value that the Text eigenvector is added or is multiplied with the element of described image feature vector corresponding position
As the element of resource vector corresponding position, the resource vector is obtained;Alternatively,
Splice the Text eigenvector and described image feature vector, obtains the resource vector.
B1. a kind of resource acquiring method, comprising:
When detecting the operational order for the first object, the corresponding fisrt feature data of first object are obtained, with
And obtain the corresponding second feature data of the second object;
Based on the fisrt feature data and the associated resource of second feature data query;
Wherein: first object is associated with second object context.
B2. the resource acquiring method according to B1,
It is described to obtain the corresponding fisrt feature data of first object, comprising:
Natural language processing is carried out to first object, obtains the first name entity in first object;
The corresponding first intention of first object is obtained based on the first name entity;
The corresponding first eigenvector of first object is obtained based on the first intention;By the first eigenvector
As the fisrt feature data;
The corresponding second feature data of the second object of the acquisition, comprising:
Natural language processing is carried out to second object, obtains the second name entity in second object;
The corresponding second intention of second object is obtained based on the second name entity;
The corresponding second feature vector of second object is obtained based on the second intention;By the second feature vector
As the second feature data.
B3 resource acquiring method according to B2, it is described to be based on the fisrt feature data and the second feature data
Obtain associated resource, comprising:
Resource vector is formed according to the first eigenvector and the second feature vector;
The associated resource is obtained according to the resource vector.
B4 resource acquiring method according to B3 is formed according to the first eigenvector and the second feature vector
Resource vector, comprising:
The value that the first eigenvector is added or is multiplied with the element of second feature vector corresponding position
As the element of resource vector corresponding position, the resource vector is obtained;Alternatively,
Splice the first eigenvector and the second feature vector, forms the resource vector.
B5. according to the described in any item resource acquiring methods of claim B1-B4, acquisition first object is corresponding
Fisrt feature data, and obtain the corresponding second feature data of the second object, comprising:
First object and second object are obtained simultaneously;
It parses first object and obtains the corresponding fisrt feature data, parse second object and obtain pair
The second feature data answered.
B6. according to the described in any item resource acquiring methods of claim B1-B4, acquisition first object is corresponding
Fisrt feature data, and obtain the corresponding second feature data of the second object, comprising:
Obtain the corresponding fisrt feature data of the first object;
Judge whether the fisrt feature data include scheduled resource characteristic data;
If it is not, searching second object, the corresponding second feature data of second object are obtained.
C1. the application provides a kind of resource acquisition device, comprising:
Text identification unit, for identifying associated, pre- with described image when detecting the operational order for image
If the text information of position;
Characteristic acquiring unit, for obtaining the corresponding text feature data of the text information, and described in obtaining
The corresponding image feature data of image;
Resource acquisition unit, for obtaining associated money according to the text feature data and described image characteristic
Source.
C2. the resource acquisition device according to C1,
It is described to obtain the corresponding text feature data of the text information, comprising:
Natural language processing is carried out to the text information, obtains the name entity in the text information;
The corresponding intention of the text information is obtained based on the name entity;
The corresponding Text eigenvector of the text information is obtained based on the intention;Using the Text eigenvector as
The text feature data;
The corresponding image feature data of the acquisition described image, comprising:
Obtain the corresponding image feature vector of described image;Using described image feature vector as described image characteristic
According to;
Associated resource is obtained according to the text feature data and described image characteristic, comprising:
Resource vector is formed according to the Text eigenvector and described image feature vector;
The associated resource is obtained according to the resource vector.
C3. the resource acquisition device according to C2,
Resource vector is formed according to the Text eigenvector and described image feature vector, comprising:
The value that the Text eigenvector is added or is multiplied with the element of described image feature vector corresponding position
As the element of resource vector corresponding position, the resource vector is obtained;Alternatively,
Splice the Text eigenvector and described image feature vector, obtains the resource vector.
D1. a kind of resource acquisition device, comprising:
Characteristic acquiring unit, for obtaining described first when detecting operational order of the user to the first object
The corresponding fisrt feature data of object, and obtain the corresponding second feature data of the second object;
Resource acquisition unit, for being based on the fisrt feature data and the associated money of second feature data query
Source;
Wherein: first object is associated with second object context.
D2. the resource acquisition device according to D1, it is described to obtain the corresponding fisrt feature data of first object, packet
It includes:
Natural language processing is carried out to first object, obtains the first name entity in first object;
The corresponding first intention of first object is obtained based on the first name entity;
The corresponding first eigenvector of first object is obtained based on the first intention;By the fisrt feature to
Amount is used as the fisrt feature data;
The corresponding second feature data of the second object of the acquisition, comprising:
Obtain the second name entity in second object;
The corresponding second intention of second object is obtained based on the second name entity;
The corresponding second feature vector of second object is obtained based on the second intention;By the second feature vector
As the second feature data.
D3. the resource acquisition device according to D2, it is described to be based on the fisrt feature data and the second feature number
According to the associated resource of acquisition, comprising:
Resource vector is formed according to the first eigenvector and the second feature vector;
The associated resource is obtained according to the resource vector.
D4. the resource acquisition device according to D3, according to the first eigenvector and the second feature vector shape
At resource vector, comprising:
The value that the first eigenvector is added or is multiplied with the element of second feature vector corresponding position
As the element of resource vector corresponding position, the resource vector is obtained;Alternatively,
Splice the first eigenvector and the second feature vector, forms the resource vector.
D5. according to the described in any item resource acquisition devices of claim D1-D4,
It is described to obtain the corresponding fisrt feature data of first object, and obtain the corresponding second feature of the second object
Data, comprising:
First object and second object are obtained simultaneously;
It parses first object and obtains the corresponding fisrt feature data, parse second object and obtain pair
The second feature data answered.
D6. according to the described in any item resource acquisition devices of claim D1-D4,
It is described to obtain the corresponding fisrt feature data of first object, and obtain the corresponding second feature of the second object
Data, comprising:
Obtain the corresponding fisrt feature data of the first object;
Judge whether the fisrt feature data include scheduled resource characteristic data;
If it is not, searching second object, the corresponding second feature data of second object are obtained.
E1. a kind of storage medium, the storage medium are stored with a plurality of instruction, and described instruction is suitable for being loaded by processor
And execute such as the described in any item resource acquiring methods of A1-A3.
F1. a kind of storage medium, the storage medium are stored with a plurality of instruction, and described instruction is suitable for being loaded by processor
And execute such as the described in any item resource acquiring methods of B1-B6.
G1. a kind of electronic equipment, including memory and processor;The memory is stored with a plurality of instruction;Described instruction
Suitable for being loaded by the processor and being executed such as the described in any item resource acquiring methods of A1-A3.
H1. a kind of electronic equipment, including memory and processor;The memory is stored with a plurality of instruction;Described instruction
Suitable for being loaded by the processor and being executed such as the described in any item resource acquiring methods of B1-B6.
Above description is only the preferred embodiment of the application 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 application, 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 foregoing invention design, 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 and disclosed herein but be not limited to similar functions
Technical characteristic replaced mutually and the technical solution that is formed.
Claims (10)
1. a kind of resource acquiring method characterized by comprising
When detecting the operational order for image, identification and described image be associated, predeterminated position text information;
The corresponding text feature data of the text information are obtained, and obtain the corresponding image feature data of described image;
Associated resource is obtained according to the text feature data and described image characteristic.
2. information-pushing method according to claim 1, which is characterized in that
It is described to obtain the corresponding text feature data of the text information, comprising:
Natural language processing is carried out to the text information, obtains the name entity in the text information;
The corresponding intention of the text information is obtained based on the name entity;
The corresponding Text eigenvector of the text information is obtained based on the intention;Using the Text eigenvector as described in
Text feature data;
The corresponding image feature data of the acquisition described image, comprising:
Obtain the corresponding image feature vector of described image;Using described image feature vector as described image characteristic;
Associated resource is obtained according to the text feature data and described image characteristic, comprising:
Resource vector is formed according to the Text eigenvector and described image feature vector;
The associated resource is obtained according to the resource vector.
3. a kind of resource acquiring method characterized by comprising
When detecting the operational order for the first object, the corresponding fisrt feature data of first object are obtained, and obtain
Take the corresponding second feature data of the second object;
Based on the fisrt feature data and the associated resource of second feature data query;
Wherein: first object is associated with second object context.
4. a kind of resource acquisition device characterized by comprising
Text identification unit, for when detecting the operational order for image, identification and described image to be associated, preset position
The text information set;
Characteristic acquiring unit, for obtaining the corresponding text feature data of the text information, and acquisition described image
Corresponding image feature data;
Resource acquisition unit, for obtaining associated resource according to the text feature data and described image characteristic.
5. resource acquisition device according to claim 4, it is characterised in that:
It is described to obtain the corresponding text feature data of the text information, comprising:
Natural language processing is carried out to the text information, obtains the name entity in the text information;
The corresponding intention of the text information is obtained based on the name entity;
The corresponding Text eigenvector of the text information is obtained based on the intention;Using the Text eigenvector as described in
Text feature data;
The corresponding image feature data of the acquisition described image, comprising:
Obtain the corresponding image feature vector of described image;Using described image feature vector as described image characteristic;
Associated resource is obtained according to the text feature data and described image characteristic, comprising:
Resource vector is formed according to the Text eigenvector and described image feature vector;
The associated resource is obtained according to the resource vector.
6. a kind of resource acquisition device characterized by comprising
Characteristic acquiring unit, for obtaining first object when detecting operational order of the user to the first object
Corresponding fisrt feature data, and obtain the corresponding second feature data of the second object;
Resource acquisition unit, for being based on the fisrt feature data and the associated resource of second feature data query;
Wherein: first object is associated with second object context.
7. a kind of storage medium, which is characterized in that the storage medium is stored with a plurality of instruction, and described instruction is suitable for by handling
Device is loaded and is executed such as the described in any item resource acquiring methods of claim 1-2.
8. a kind of storage medium, which is characterized in that the storage medium is stored with a plurality of instruction, and described instruction is suitable for by handling
Device loads and executes resource acquiring method as claimed in claim 3.
9. a kind of electronic equipment, it is characterised in that: including memory and processor;
The memory is stored with a plurality of instruction;Described instruction is suitable for being loaded by the processor and being executed such as claim 1-
2 described in any item resource acquiring methods.
10. a kind of electronic equipment, it is characterised in that: including memory and processor;
The memory is stored with a plurality of instruction;Described instruction is suitable for being loaded by the processor and executing such as claim 3
The resource acquiring method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910472052.1A CN110232141A (en) | 2019-05-31 | 2019-05-31 | Resource acquiring method, resource acquisition device, storage medium and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910472052.1A CN110232141A (en) | 2019-05-31 | 2019-05-31 | Resource acquiring method, resource acquisition device, storage medium and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110232141A true CN110232141A (en) | 2019-09-13 |
Family
ID=67858903
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910472052.1A Pending CN110232141A (en) | 2019-05-31 | 2019-05-31 | Resource acquiring method, resource acquisition device, storage medium and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110232141A (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102436477A (en) * | 2011-10-11 | 2012-05-02 | 鸿富锦精密工业(深圳)有限公司 | Device with related content search function and method |
US20120131520A1 (en) * | 2009-05-14 | 2012-05-24 | Tang ding-yuan | Gesture-based Text Identification and Selection in Images |
CN105975554A (en) * | 2016-04-29 | 2016-09-28 | 广东小天才科技有限公司 | Mobile terminal-based big data search method and apparatus |
CN106933809A (en) * | 2017-03-27 | 2017-07-07 | 三角兽(北京)科技有限公司 | Information processor and information processing method |
CN106997372A (en) * | 2016-12-07 | 2017-08-01 | 阿里巴巴集团控股有限公司 | The method and apparatus that business operation is realized based on picture |
CN107967110A (en) * | 2017-11-30 | 2018-04-27 | 广东小天才科技有限公司 | Playback method, playing device, electronic equipment and computer-readable recording medium |
CN108108771A (en) * | 2018-01-03 | 2018-06-01 | 华南理工大学 | Image answering method based on multiple dimensioned deep learning |
CN109032491A (en) * | 2018-07-23 | 2018-12-18 | Oppo广东移动通信有限公司 | Data processing method, device and mobile terminal |
-
2019
- 2019-05-31 CN CN201910472052.1A patent/CN110232141A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120131520A1 (en) * | 2009-05-14 | 2012-05-24 | Tang ding-yuan | Gesture-based Text Identification and Selection in Images |
CN102436477A (en) * | 2011-10-11 | 2012-05-02 | 鸿富锦精密工业(深圳)有限公司 | Device with related content search function and method |
CN105975554A (en) * | 2016-04-29 | 2016-09-28 | 广东小天才科技有限公司 | Mobile terminal-based big data search method and apparatus |
CN106997372A (en) * | 2016-12-07 | 2017-08-01 | 阿里巴巴集团控股有限公司 | The method and apparatus that business operation is realized based on picture |
CN106933809A (en) * | 2017-03-27 | 2017-07-07 | 三角兽(北京)科技有限公司 | Information processor and information processing method |
CN107765970A (en) * | 2017-03-27 | 2018-03-06 | 三角兽(北京)科技有限公司 | Information processor and information processing method |
CN107967110A (en) * | 2017-11-30 | 2018-04-27 | 广东小天才科技有限公司 | Playback method, playing device, electronic equipment and computer-readable recording medium |
CN108108771A (en) * | 2018-01-03 | 2018-06-01 | 华南理工大学 | Image answering method based on multiple dimensioned deep learning |
CN109032491A (en) * | 2018-07-23 | 2018-12-18 | Oppo广东移动通信有限公司 | Data processing method, device and mobile terminal |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110362370B (en) | Webpage language switching method and device and terminal equipment | |
CN106982256B (en) | Information pushing method, device, equipment and storage medium | |
CN109086394B (en) | Search ranking method and device, computer equipment and storage medium | |
KR100996311B1 (en) | Method and system for detecting spam user created contentucc | |
CN106326420B (en) | Recommendation method and device for mobile terminal | |
US20130066818A1 (en) | Automatic Crowd Sourcing for Machine Learning in Information Extraction | |
CN110263248A (en) | A kind of information-pushing method, device, storage medium and server | |
KR20060049261A (en) | Method, system, and apparatus for processing information based on the discovery of semantically labeled strings | |
CN112035638B (en) | Information processing method, device, storage medium and equipment | |
CN110019703B (en) | Data marking method and device and intelligent question-answering method and system | |
CN108287862A (en) | Search engine | |
WO2022063123A1 (en) | Interface displaying method, device, and electronic device | |
CN112199526B (en) | Method and device for issuing multimedia content, electronic equipment and storage medium | |
CN109343926A (en) | Application program image target display methods, device, terminal and storage medium | |
CN110069769A (en) | Using label generating method, device and storage equipment | |
CN110489032B (en) | Dictionary query method for electronic book and electronic equipment | |
CN112631437A (en) | Information recommendation method and device and electronic equipment | |
CN112416142A (en) | Method and device for inputting characters and electronic equipment | |
CN112416212B (en) | Program access method, apparatus, electronic device and readable storage medium | |
CN111813236B (en) | Input method, input device, electronic equipment and readable storage medium | |
US8589492B2 (en) | Electronic message object drop feature | |
CN110232141A (en) | Resource acquiring method, resource acquisition device, storage medium and electronic equipment | |
CN109597498A (en) | A kind of input method based on scene, device, equipment | |
CN113918796A (en) | Information searching method, device, server and storage medium | |
CN114237405A (en) | Message correction method and device and electronic equipment |
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: 20200724 Address after: 518000 Nanshan District science and technology zone, Guangdong, Zhejiang Province, science and technology in the Tencent Building on the 1st floor of the 35 layer Applicant after: TENCENT TECHNOLOGY (SHENZHEN) Co.,Ltd. Address before: 100029, Beijing, Chaoyang District new East Street, building No. 2, -3 to 25, 101, 8, 804 rooms Applicant before: Tricorn (Beijing) Technology Co.,Ltd. |
|
TA01 | Transfer of patent application right | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190913 |
|
RJ01 | Rejection of invention patent application after publication |