CN106776860A - One kind search abstraction generating method and device - Google Patents
One kind search abstraction generating method and device Download PDFInfo
- Publication number
- CN106776860A CN106776860A CN201611070196.7A CN201611070196A CN106776860A CN 106776860 A CN106776860 A CN 106776860A CN 201611070196 A CN201611070196 A CN 201611070196A CN 106776860 A CN106776860 A CN 106776860A
- Authority
- CN
- China
- Prior art keywords
- user
- search
- portrait
- query word
- network address
- 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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06F16/34—Browsing; Visualisation therefor
- G06F16/345—Summarisation for human users
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides one kind search abstraction generating method and device, the method includes：Obtain the target network address related to the query word of user input；And user's portrait of user is obtained, the user draws a portrait at least one in the search custom, searching preferences, search rate, consumption habit, the search focus that describe user；Finally, according to above-mentioned query word and the corresponding search summary of the above-mentioned user portrait above-mentioned target network address of generation, and show that the search is made a summary.During user's portrait is incorporated the generation of search summary and shown by the embodiment of the present invention, the foundation that user's portrait is made a summary as generation search, the search summary that generation can be effectively ensured more meets search custom, preference, the focus of user, so that user according to search summary can be fast and accurately identified the link whether the link for needed for user, and then improve user search efficiency and accuracy.
Description
Technical field
The present invention relates to search engine technique field, in particular to one kind search abstraction generating method and device.
Background technology
At present, with the development of the network information technology, based on required information requirement quickly and is accurately found, search is drawn
Hold up and arise at the historic moment.When user is retrieved in search system, the retrieval request that search engine is submitted to according to user provides a large amount of
The URL (Uniform Resource Locator, URL, also referred to as network address) and summary of correspondence website, Yong Hutong
Reading summary is crossed to decide whether to browse the corresponding website of the summary.Therefore, the meaning of summary is to instruct user to judge URL's
Value.
In general, summary refers to an A brief introduction for article, accurately can comprehensively reflect a certain piece article center
The succinct coherent short essay of content, but the summary concept that search engine is used is different, and it is not necessarily brief Jie of article
Continue, it is the segment information extracted in webpage according to certain rule.Summary in search engine is according to the mode for obtaining
Difference can be divided into two kinds of summaries, and one kind is static abstract, and another kind is dynamic abstract, wherein, static abstract is only about page
The general information of face content, the query word with user input is unrelated, centered on the content of document；Dynamic abstract is exactly in response
When user inquires about, according to the position that query word occurs in a document, word related around query word is extracted, will in display
Query word is marked and returns to user, centered on inquiry.When using dynamic abstract, for a document, work as user
When the query word of input is different, the summary for ultimately generating is different.
For existing technical scheme, although dynamic abstract is associated with the query word of user input, by document
Some fragments related to query word are constituted, and the dynamic abstract contains the information related to the query word of user input, but,
Each corresponding dynamic abstract of link of the prior art has that specific aim is poor for different users, causes to use
Family can not be fast and accurately identified according to existing dynamic abstract the link whether the link for needed for user, and then reduce
The search efficiency and accuracy of user.
The content of the invention
In view of this, the purpose of the embodiment of the present invention is to provide a kind of search abstraction generating method and device, to solve
Existing dynamic abstract can not completely, accurately reflect out whether link is user's content interested, easily mislead user, and then
Reduce the search efficiency of user and the problem of accuracy.
In a first aspect, the embodiment of the invention provides a kind of search abstraction generating method, the method includes：
Acquisition according to the query word of user input retrieve the target network address related to the query word for obtaining；
User's portrait of the user is obtained, the search that the user draws a portrait for describing the user is accustomed to, search is inclined
At least one in good, search rate, consumption habit, search focus；
According to the query word and the corresponding search summary of the user portrait generation target network address；
Show the corresponding search summary of each described target network address.
With reference in a first aspect, the embodiment of the invention provides the first possible implementation method of first aspect, wherein, institute
Before stating the user's portrait for obtaining the user, also include：
The historical behavior data of the user are obtained, the historical behavior data include that user basic information, history are logged in
Information, historical search record, historical viewings record, history collection record, history Download History, historical review record and history disappear
Take at least one in record；And/or,
Obtain the contextual information of user's current search behavior, the contextual information includes current time, current
At least one in place, current environment, the user webpage that the word of search, the user browse recently recently；
Historical behavior data and/or the contextual information according to the user set up user's portrait of the user.
With reference in a first aspect, the embodiment of the invention provides second possible implementation method of first aspect, wherein, institute
State according to the query word and the corresponding search summary of the user portrait generation target network address, including：
According to the query word, multiple candidate's summaries are extracted in the document in the corresponding webpage of the target network address
Sentence；
The weighted value of each feature tag in user's portrait is set；
According to candidate's summary sentence and user portrait degree of association order from high to low according to the power of each feature tag
Weight values are ranked up to multiple candidate's summary sentences；
The candidate summary sentence generation target network address for choosing the forward predetermined number of sequence searches for summary accordingly.
With reference to second possible implementation method of first aspect, the third of first aspect is the embodiment of the invention provides
Possible implementation method, wherein, it is described according to the query word, extracted in the document in the corresponding webpage of the target network address
Go out multiple candidate's summary sentences, including：
Participle is carried out to the document in the corresponding webpage of the target network address；
The query word is matched with the document after participle, the query word is calculated and is gone out within said document
Existing position；And the near synonym of the query word are matched with the document after participle, calculate the query word
The position that occurs within said document of near synonym；
Within said document, the position that the position for occurring with the query word and the near synonym of the query word occur is as base
Standard, the multiple candidate's summary sentences of interception.
With reference to second possible implementation method of first aspect, the 4th kind of first aspect is the embodiment of the invention provides
Possible implementation method, wherein, the weighted value that each feature tag in user's portrait is set, including：
Calculate the operation concentration degree of each feature tag in being drawn a portrait to the user in user's history；
The weighted value of each feature tag is determined according to the operation concentration degree.
With reference to second possible implementation method of first aspect, the 4th kind of first aspect is the embodiment of the invention provides
Possible implementation method, wherein, it is described to draw a portrait degree of association order from high to low according to each with user according to candidate's summary sentence
The weighted value of feature tag is ranked up to multiple candidate's summary sentences, including：
The degree of association is drawn a portrait by height with user according to candidate's summary sentence using Logic Regression Models or gradient lifting tree-model
Multiple candidate's summary sentences are ranked up according to the weighted value of each feature tag to low order.
Second aspect, the embodiment of the present invention additionally provides a kind of search summarization generation device, and the device includes：
Target website acquisition module, the query word for obtaining according to user input retrieve obtaining with the inquiry
The related target network address of word；
User's portrait acquisition module, the user for obtaining the user draws a portrait, and the user draws a portrait described for describing
At least one in search custom, searching preferences, search rate, consumption habit, the search focus of user；
Search summarization generation module, for corresponding with the user portrait generation target network address according to the query word
Search summary.
With reference to second aspect, the first possible implementation method of second aspect is the embodiment of the invention provides, wherein, institute
Stating device also includes：
Historical behavior data acquisition module, the historical behavior data for obtaining the user, the historical behavior data
Downloaded including user basic information, history logon information, historical search record, historical viewings record, history collection record, history
At least one in record, historical review record and history consumer record；And/or,
Contextual information acquisition module, the contextual information for obtaining user's current search behavior is described upper and lower
Literary information includes that the word of search, the user browse recently recently for current time, current location, current environment, the user
Webpage at least one；
User's portrait sets up module, is built for the historical behavior data and/or the contextual information according to the user
Found user's portrait of the user.
With reference to second aspect, second possible implementation method of second aspect is the embodiment of the invention provides, wherein, institute
Stating search summarization generation module includes：
Candidate's summary sentence extracting sub-module, for according to the query word, in the corresponding webpage of the target network address
Multiple candidate's summary sentences are extracted in document；
Feature tag weighted value determination sub-module, the weighted value for setting each feature tag in user's portrait；
Candidate's summary sentence sorting sub-module, for the order according to candidate's summary sentence with user's portrait degree of association from high to low
The weighted value according to each feature tag is ranked up to multiple candidate's summary sentences；
Search summarization generation submodule, the candidate summary sentence generation for choosing the forward predetermined number of sequence is described
Target network address searches for summary accordingly.
With reference to second possible implementation method of second aspect, the third of second aspect is the embodiment of the invention provides
Possible implementation method, wherein, candidate's summary sentence extracting sub-module, specifically for the corresponding webpage of the target network address
In document carry out participle；The query word is matched with the document after participle, the query word is calculated in institute
State the position occurred in document；And the near synonym of the query word are matched with the document after participle, calculate
The position that the near synonym of the query word occur within said document；Within said document, the position for occurring with the query word
On the basis of the position occurred with the near synonym of the query word, the multiple candidate's summary sentences of interception.
With reference to second possible implementation method of second aspect, the 4th kind of second aspect is the embodiment of the invention provides
Possible implementation method, wherein, the feature tag weighted value determination sub-module, specifically for calculating in user's history to described
The operation concentration degree of each feature tag in user's portrait；Each feature tag is determined according to the operation concentration degree
Weighted value.
With reference to second possible implementation method of second aspect, the 4th kind of second aspect is the embodiment of the invention provides
Possible implementation method, wherein, candidate's summary sentence sorting sub-module, specifically for using Logic Regression Models or gradient
Lifting tree-model is according to candidate's summary sentence and user portrait degree of association order from high to low according to each feature tag
Weighted value is ranked up to multiple candidate's summary sentences.
With reference to second aspect, the 5th kind of possible implementation method of second aspect is the embodiment of the invention provides, wherein, institute
Stating device also includes：
Summary display module is searched for, summary is searched for for showing that each described target network address is corresponding, this is searched element and make a summary and is
Generated according to the query word and user's portrait.
In search abstraction generating method provided in an embodiment of the present invention and device, the method includes：Obtain defeated with user
The target network address of the query word correlation for entering；And user's portrait of user is obtained, the user draws a portrait for the search for describing user
At least one in custom, searching preferences, search rate, consumption habit, search focus；Finally, according to above-mentioned query word and
Above-mentioned user's portrait generates the corresponding search summary of the target network address, and shows that the search is made a summary.Compared to prior art, this
Invention by user portrait incorporate search summary generation with displaying during, using user portrait as generation search make a summary according to
According to the search summary that generation can be effectively ensured more meets search custom, preference, the focus of user such that it is able to so that with
Family can be fast and accurately identified according to search summary the link whether link for needed for user, and then improve user
Search efficiency and accuracy.
To enable the above objects, features and advantages of the present invention to become apparent, preferred embodiment cited below particularly, and coordinate
Appended accompanying drawing, is described in detail below.
Brief description of the drawings
Technical scheme in order to illustrate more clearly the embodiments of the present invention, below will be attached to what is used needed for embodiment
Figure is briefly described, it will be appreciated that the following drawings illustrate only certain embodiments of the present invention, thus be not construed as it is right
The restriction of scope, for those of ordinary skill in the art, on the premise of not paying creative work, can also be according to this
A little accompanying drawings obtain other related accompanying drawings.
Fig. 1 shows a kind of schematic flow sheet of search abstraction generating method that the embodiment of the present invention is provided；
Fig. 2 searches for summarization generation process in showing a kind of search abstraction generating method that the embodiment of the present invention is provided
Schematic flow sheet；
Fig. 3 shows a kind of structural representation of search summarization generation device that the embodiment of the present invention is provided；
Fig. 4 searches for summarization generation module in showing a kind of search summarization generation device that the embodiment of the present invention is provided
Structural representation；
Fig. 5 shows the structural representation of another search summarization generation device that the embodiment of the present invention is provided；
Fig. 6 shows a kind of bandwagon effect of the corresponding search summary of each target network address that the embodiment of the present invention is provided
Schematic diagram.
Specific embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
Middle accompanying drawing, is clearly and completely described to the technical scheme in the embodiment of the present invention, it is clear that described embodiment is only
It is a part of embodiment of the invention, rather than whole embodiments.The present invention generally described and illustrated in accompanying drawing herein is real
The component for applying example can be arranged and designed with a variety of configurations.Therefore, it is of the invention to what is provided in the accompanying drawings below
The detailed description of embodiment is not intended to limit the scope of claimed invention, but is merely representative of selected reality of the invention
Apply example.Based on embodiments of the invention, the institute that those skilled in the art are obtained on the premise of creative work is not made
There is other embodiment, belong to the scope of protection of the invention.
In view of for existing technical scheme, although dynamic abstract is associated with the query word of user input, by text
Fragment some related to query word in shelves is constituted, and the dynamic abstract contains the letter related to the query word of user input
Breath, but, there is asking for specific aim difference for different users in each corresponding dynamic abstract of link of the prior art
Topic, cause user can not be fast and accurately identified according to existing dynamic abstract the link whether the link for needed for user,
And then reduce the search efficiency and accuracy of user.Based on this, a kind of search abstraction generating method is the embodiment of the invention provides
And device, it is described below by embodiment.
As shown in figure 1, the embodiment of the invention provides a kind of search abstraction generating method, the method comprising the steps of S102-
S106, it is specific as follows：
Step S102：Acquisition according to the query word of user input retrieve the target network related to the query word for obtaining
Location；
Step S104：Obtain above-mentioned user user portrait, the user draw a portrait for describe above-mentioned user search be accustomed to,
At least one in searching preferences, search rate, consumption habit, search focus；
Step S106：According to above-mentioned query word and the corresponding search summary of the above-mentioned user portrait above-mentioned target network address of generation；
Step S108：Show the corresponding above-mentioned search summary of each target network address.
Specifically, by user terminal receive user input query word, user can by search interface input inquiry word,
Can also by way of voice input inquiry word, then, user terminal sends to server the query word, server according to
The query word is retrieved, and retrieval obtains the target network address related to the query word；
Wherein, the process of generation search summary can be performed by user terminal, it is also possible to be performed by server, specifically
For：
When generation search summary process by user terminal to perform when, server will retrieve related to query word
Target network address webpage corresponding with each target network address is sent to user terminal；User terminal is according to the query word and use for getting
The corresponding search summary of the family above-mentioned target network address of portrait generation, finally, shown successively by user terminal each target network address and
The corresponding search summary of each target network address；
When generation search summary process by server to perform when, server is examined according to the query word for receiving
Rope, obtains the target network address related to the query word；Server is according to the query word for getting and user's portrait above-mentioned mesh of generation
The corresponding search summary of mark network address, and each target network address search summary corresponding with each target network address was sent to user's end
End, finally, each target network address and the corresponding search summary of each target network address is shown by user terminal successively.
Specifically, the corresponding above-mentioned search summary of each target network address of above-mentioned displaying, is specifically included in title displaying
Area shows each target network address corresponding title related to the query word of user input and in search summary exhibition one by one
Show that area shows the corresponding search summary of each target network address one by one, wherein, this is according to above-mentioned query word and user's portrait generation
, target network address network address mark is carried on title；
During due to user's portrait to be incorporated the generation of search summary and is shown, plucked user's portrait as generation search
The foundation wanted, the finally search in the show area displaying of search summary is made a summary and more meets search custom, preference, the focus of user,
User can according to search summary can be fast and accurately identified the link whether the link for needed for user, specifically, work as
When user terminal receives user's click commands, the corresponding web page contents of target network address are opened and show, now user can be with
Browse whole web page contents；
Further, before the corresponding web page contents of display target network address, each target network address of above-mentioned displaying is corresponding
Above-mentioned search summary, also including being marked to content related to search summary in web page contents, the mark includes：Overstriking mark
One or more in note, color mark, lower camber line mark；Then, then show mark after the corresponding webpage of target network address
Content, so enables to user quickly to view the related content near search summary；
In the embodiment of present invention offer, during user's portrait is incorporated the generation of search summary and shown, will use
Family portrait as generation search summary foundation, can be effectively ensured generation search summary more meet user search be accustomed to,
Preference, focus such that it is able to so that user can be fast and accurately identified whether the link is user according to search summary
Required link, and then improve the search efficiency and accuracy of user.
In addition, it is necessary to illustrate, the priority to step S102 and step S104 in the embodiment that the present invention is provided is held
Row order do not limit, further, in order to will obtain the multiple target network address related to query word display order and use
User's portrait at family is associated, the corresponding search summary of each target network address of above-mentioned displaying, including：According to target network address pair
Order of the web page contents answered with user's portrait degree of association from high to low shows each target network address and the target successively
The corresponding search summary of network address, in such manner, it is possible to so that user priority is checked and user's maximally related target network address of portrait.
Further to be drawn a portrait, it is necessary to set up user respectively for each user in advance, user's portrait accurately describes to use
There is certain difference in the important informations such as family behavioural habits, consumption habit, user's portrait of each user, can be to follow-up raw
Search digest procedure into each target network address provides effective theoretical foundation, based on this, the user of the above-mentioned above-mentioned user of acquisition
Before portrait, also include：
The historical behavior data of above-mentioned user are obtained, the historical behavior data include that user basic information, history log in letter
Breath, historical search record, historical viewings record, history collection record, history Download History, historical review record and history consumption
At least one in record；And/or,
The contextual information of above-mentioned user's current search behavior is obtained, the contextual information includes current time, current position
At least one in point, current environment, above-mentioned the user webpage that the word of search, above-mentioned user browse recently recently；
Historical behavior data and/or above-mentioned contextual information according to above-mentioned user set up user's portrait of above-mentioned user.
Specifically, above-mentioned user basic information can include：Age of user, user's educational background, user position information, use
Family sex, user's occupation, in addition, in the embodiment that the present invention is provided, the corresponding user's portrait of each user can be basis
User's is historical behavior data genaration, or according to the generation of the contextual information of user's current search behavior, may be used also
To be to be generated according to the historical behavior data of user and the contextual information of user's current search behavior simultaneously, wherein, it is considered to
What is mainly reflected to user's history behavioral data is to big datas such as user's early stage search customs, to the quick of the current intention of user
Perception is slightly weak, therefore, the contextual information of current search behavior is increased during the user for setting up user draws a portrait, when such as inquiring about
Between, place, environment, nearest term, browse webpage etc. recently, can the more accurately current intention of identifying user so that
Further increase the precision of the search summary of final presentation.
For example, contextual information generation user's portrait of the historical behavior data and user's current search behavior according to user
When, subsequently according to the user portrait generate each target network address it is corresponding search digest procedure in, be current with contextual information
As a example by time, for sports news class search behavior, it is divided into current time for racing season and non-season, when current time is the racing season
When, the search summary of generation should include the live relevant information of real-time race, and the information is probably that active user most pays close attention to
Information.
Further, it is contemplated that the historical behavior data of the user for getting may the data comprising some unformatteds or
The data of person's redundancy, the data not only need huge storage overhead, and are not easy to big data analysis, based on this, are obtaining
After the historical behavior data of above-mentioned user, also include：The historical behavior data of the user to getting are carried out at cleaning duplicate removal
Reason and formatting treatment.
In the embodiment that the present invention is provided, by being drawn a portrait for each user sets up corresponding user in advance, subsequently in life
During search summary corresponding into target network address, the corresponding user's portrait of current search user is taken into account, be directed to same mesh
Mark network address, even if the query word of different user input is identical, but final target network address for showing searches for summary accordingly
May be different so that finally be presented to the search summary of user more targetedly, and with more guiding value.
Wherein, as shown in Fig. 2 above-mentioned steps S106：According to above-mentioned query word and the above-mentioned user portrait above-mentioned target of generation
The corresponding search summary of network address, specifically includes：
Step S1062：According to above-mentioned query word, multiple is extracted in the document in the corresponding webpage of above-mentioned target network address
Candidate's summary sentence；
Step S1064：The weighted value of each feature tag in above-mentioned user's portrait is set；
Step S1066：Degree of association order from high to low is drawn a portrait according to each feature mark with user according to candidate's summary sentence
The above-mentioned weighted value signed is ranked up to multiple above-mentioned candidate's summary sentences；
Step S1068：The above-mentioned candidate summary sentence above-mentioned target network address of generation for choosing the forward predetermined number of sequence is corresponding
Search summary.
Specifically, above-mentioned steps S1062：According to above-mentioned query word, the document in the corresponding webpage of above-mentioned target network address
In extract multiple candidates summary sentences, specifically include：
Participle is carried out to the document in the corresponding webpage of above-mentioned target network address；
Above-mentioned query word is matched with the above-mentioned document after participle, above-mentioned query word is calculated and is gone out in the document mentioned above
Existing position；And the near synonym of above-mentioned query word are matched with the above-mentioned document after participle, calculate above-mentioned query word
The position that occurs in the document mentioned above of near synonym；
In the document mentioned above, the position that the position for occurring with above-mentioned query word and the near synonym of above-mentioned query word occur is as base
Standard, the multiple candidate's summary sentences of interception.
Specifically, each feature tag in being drawn a portrait in view of user to the influence degree of user's synthesis preference not
Together, by way of different weighted values are set for each feature tag, the search for ultimately generating can further be lifted
Summary specific aim and accuracy, based on this, above-mentioned steps S1064：The weight of each feature tag in above-mentioned user's portrait is set
Value, specifically includes：
Calculate the operation concentration degree of each features described above label in being drawn a portrait to above-mentioned user in user's history；
The weighted value of each features described above label is determined according to aforesaid operations concentration degree.
For example, be analyzed by the historical behavior data to a certain user learning, the level of consumption of the user mainly collects
In in the relatively low commodity of purchasing price (belonging to low consumption colony), and the Brang Preference of the user is concentrated mainly on purchase famous brand
Commodity (belonging to famous brand fan), daily product is on the high side in the type of merchandize that the user is bought, but the user
The type of merchandize bought is more various, on the corresponding feature tag of the level of consumption is low consumption in the user of user portrait
Colony, is famous brand fan on the corresponding feature tag of Brang Preference, is day on the corresponding feature tag of type of merchandize
Normal daily class, however, because the operation concentration degree of the level of consumption and Brang Preference in user's portrait of the user is higher than business
The operation concentration degree of kind class, at this point it is possible to be set to the power of the weighted value more than type of merchandize of the level of consumption and Brang Preference
Weight values；
It is corresponding, finally for the search summary of each goods links generation can more be focused on illustrating the introduction on brand,
And brand discounting, the relevant information such as promote, purchase on a barter basis, now, user can quickly and accurately recognize from search summary
Go out the link whether the link for needed for user, contribute to user to open effective link, filter out invalid link.
Specifically, above-mentioned steps S1066：According to candidate's summary sentence and user portrait degree of association order from high to low according to
The above-mentioned weighted value of each feature tag is ranked up to multiple above-mentioned candidate's summary sentences, specifically includes：
The degree of association is drawn a portrait by height with user according to candidate's summary sentence using Logic Regression Models or gradient lifting tree-model
Multiple above-mentioned candidate's summary sentences are ranked up according to the above-mentioned weighted value of each feature tag to low order.
Wherein it is possible to be arranged the candidate's summary sentence for selecting using Logic Regression Models or gradient lifting tree-model
Sequence, the weighted value of each feature tag in candidate is made a summary sentence and user's portrait is input into model as |input paramete, i.e.,
The ranking results of exportable each candidate summary sentence and user's portrait degree of association.
In the embodiment that the present invention is provided, different weights are set by for each feature tag in user's portrait
Value, the degree of association that each candidate makes a summary between sentence and user's portrait is determined with reference to the weighted value of each feature tag, and according to pass
Connection degree order from high to low is ranked up to each candidate summary sentence, finally, chooses the candidate of the forward predetermined number of sequence
Summary sentence is made a summary as the search of final display, in such manner, it is possible to further lift the search summary specific aim for ultimately generating
And accuracy.
In search abstraction generating method provided in an embodiment of the present invention, first, the query word phase with user input is obtained
The target network address of pass；And user's portrait of user is obtained, the search that the user draws a portrait for describing user is accustomed to, search is inclined
At least one in good, search rate, consumption habit, search focus；Finally, drawn a portrait according to above-mentioned query word and above-mentioned user
The corresponding search summary of the target network address is generated, and shows that the search is made a summary.User's portrait is incorporated search by the embodiment of the present invention
During the generation and displaying of summary, the foundation that user's portrait is made a summary as generation search can be effectively ensured searching for generation
Rope summary more meets search custom, preference, the focus of user such that it is able to so that quick by user makes a summary according to search,
Identify exactly the link whether the link for needed for user, and then improve user search efficiency and accuracy；Further
, it is contemplated that what user's history behavioral data mainly reflected is to big datas such as user's early stage search customs, to the current of user
The sensitiveness of intention is slightly weak, therefore, the contextual information of current search behavior is increased during the user for setting up user draws a portrait,
Can the more accurately current intention of identifying user, so as to further increase the precision of the search summary of final presentation；
Further, different weighted values are set by for each feature tag in user's portrait, with reference to each feature tag
Weighted value determines the degree of association between each candidate summary sentence and user's portrait, and according to degree of association order from high to low to each
Individual candidate's summary sentence is ranked up, and finally, chooses candidate's summary sentence searching as final display of the forward predetermined number of sequence
Rope is made a summary, in such manner, it is possible to further lift search summary specific aim and the accuracy for ultimately generating.
The embodiment of the present invention also provides a kind of search summarization generation device, as shown in figure 3, the device includes：
Target website acquisition module 302, the query word for obtaining according to user input retrieve obtain with it is described
The related target network address of query word；
User's portrait acquisition module 304, the user for obtaining above-mentioned user draws a portrait, and the user draws a portrait above-mentioned for describing
At least one in search custom, searching preferences, search rate, consumption habit, the search focus of user；
Search summarization generation module 306, for according to above-mentioned query word and the above-mentioned user portrait above-mentioned target network address of generation
Corresponding search summary.
Specifically, above-mentioned search summarization generation device can be arranged at subscriber terminal side, it is also possible to be arranged at server side,
Specially：
(1) when above-mentioned search summarization generation device is arranged at subscriber terminal side, server is by retrieve and query word
Related target network address webpage corresponding with each target network address is sent to user terminal；User terminal is according to the inquiry for getting
Word and the corresponding search summary of user's portrait above-mentioned target network address of generation, mesh is extracted in each corresponding webpage of target network address
Mark network address searches for summary accordingly, finally, shows each target network address and each target network address correspondence successively by user terminal
Search summary；
(2) when above-mentioned search summarization generation device is arranged at server side, server enters according to the query word for receiving
Row retrieval, obtains the target network address related to the query word；Server is according on the query word for getting and user's portrait generation
The corresponding search summary of target network address is stated, target network address is extracted in each corresponding webpage of target network address, and search is plucked accordingly
Will, and each target network address search summary corresponding with each target network address is sent to user terminal, finally, by user terminal
Show each target network address and the corresponding search summary of each target network address successively.
In the embodiment of present invention offer, during user's portrait is incorporated the generation of search summary and shown, will use
Family portrait as generation search summary foundation, can be effectively ensured generation search summary more meet user search be accustomed to,
Preference, focus such that it is able to so that user can be fast and accurately identified whether the link is user according to search summary
Required link, and then improve the search efficiency and accuracy of user.
Further to be drawn a portrait, it is necessary to set up user respectively for each user in advance, user's portrait accurately describes to use
There is certain difference in the important informations such as family behavioural habits, consumption habit, user's portrait of each user, can be to follow-up raw
Search digest procedure into each target network address provides effective theoretical foundation, and based on this, said apparatus also include：
Historical behavior data acquisition module, the historical behavior data for obtaining above-mentioned user, the historical behavior packet
Include and record under user basic information, history logon information, historical search record, historical viewings record, history collection record, history
At least one in record, historical review record and history consumer record；And/or,
Contextual information acquisition module, the contextual information for obtaining above-mentioned user's current search behavior, the context
Information includes that the word of search, above-mentioned user browse recently recently for current time, current location, current environment, above-mentioned user
At least one in webpage；
User's portrait sets up module, is built for the historical behavior data and/or above-mentioned contextual information according to above-mentioned user
Found user's portrait of above-mentioned user.
In the embodiment that the present invention is provided, by being drawn a portrait for each user sets up corresponding user in advance, subsequently in life
During search summary corresponding into target network address, the corresponding user's portrait of current search user is taken into account, be directed to same mesh
Mark network address, even if the query word of different user input is identical, but final target network address for showing searches for summary accordingly
May be different so that finally be presented to the search summary of user more targetedly, and with more guiding value.
Further, as shown in figure 4, above-mentioned search summarization generation module 306 includes：
Candidate's summary sentence extracting sub-module 3062, for according to above-mentioned query word, in the corresponding webpage of above-mentioned target network address
In document in extract multiple candidates summary sentences；
Feature tag weighted value determination sub-module 3064, the weight for setting each feature tag in above-mentioned user's portrait
Value；
Candidate's summary sentence sorting sub-module 3066, for according to candidate's summary sentence and user's portrait degree of association from high to low
Order is ranked up according to the above-mentioned weighted value of each feature tag to multiple above-mentioned candidate's summary sentences；
Search summarization generation submodule 3068, the above-mentioned candidate summary sentence generation for choosing the forward predetermined number of sequence
Above-mentioned target network address searches for summary accordingly.
Further, above-mentioned candidate's summary sentence extracting sub-module 3062, specifically for the corresponding net of above-mentioned target network address
Document in page carries out participle；Above-mentioned query word is matched with the above-mentioned document after participle, above-mentioned query word is calculated and is existed
The position occurred in above-mentioned document；And the near synonym of above-mentioned query word are matched with the above-mentioned document after participle, calculate
Go out the position that the near synonym of above-mentioned query word occur in the document mentioned above；In the document mentioned above, the position for occurring with above-mentioned query word
Put on the basis of the position occurred with the near synonym of above-mentioned query word, intercept multiple candidates' summary sentences.
Further, it is contemplated that each feature tag in user's portrait to the influence degree of user's synthesis preference not
Together, by way of different weighted values are set for each feature tag, the search for ultimately generating can further be lifted
Summary specific aim and accuracy, based on this, features described above label weighted value determination sub-module 3064 is gone through specifically for calculating user
The operation concentration degree of each features described above label in being drawn a portrait to above-mentioned user in history；Determined on each according to aforesaid operations concentration degree
State the weighted value of feature tag.
Further, above-mentioned candidate's summary sentence sorting sub-module 3066, specifically for using Logic Regression Models or ladder
Degree lifting tree-model draws a portrait degree of association order from high to low according to the upper of each feature tag according to candidate's summary sentence with user
Weighted value is stated to be ranked up multiple above-mentioned candidate's summary sentences.
In the embodiment that the present invention is provided, different weights are set by for each feature tag in user's portrait
Value, the degree of association that each candidate makes a summary between sentence and user's portrait is determined with reference to the weighted value of each feature tag, and according to pass
Connection degree order from high to low is ranked up to each candidate summary sentence, finally, chooses the candidate of the forward predetermined number of sequence
Summary sentence is made a summary as the search of final display, in such manner, it is possible to further lift the search summary specific aim for ultimately generating
And accuracy.
Further, as shown in figure 5, said apparatus also include：
Search summary display module 308, for showing the corresponding search summary of each target network address, it is root that this searches element summary
Generated according to the query word and user's portrait.
Specifically, because above-mentioned search summarization generation device can also be arranged at subscriber terminal side, now, user terminal bag
Above-mentioned generating means are included, is specifically included：
Target website acquisition module 302, the query word for obtaining according to user input retrieve being looked into this of obtaining
The related target network address of word is ask, i.e., the target network address related to query word and each target network address that the reception server sends are corresponding
Webpage；
User's portrait acquisition module 304, the user for obtaining above-mentioned user draws a portrait, and the user draws a portrait above-mentioned for describing
At least one in search custom, searching preferences, search rate, consumption habit, the search focus of user；
Search summarization generation module 306, for according to above-mentioned query word and the above-mentioned user portrait above-mentioned target network address of generation
Corresponding search summary；
Search summary display module 308, for showing the corresponding above-mentioned search summary of each above-mentioned target network address.
Specifically, as shown in fig. 6, the bandwagon effect for giving a kind of corresponding search summary of each target network address is illustrated
Figure, wherein, the query word of user input is received by user terminal, user can be by the query word input area 11 on search interface
Input inquiry word, it is also possible to the input inquiry word by way of phonetic entry area 22 is using voice；
Above-mentioned search summary display module 308, specifically for showing one by one and user input in title show area 331
Related each the corresponding title of target network address of query word and show each one by one in search summary show area 332
The corresponding search summary of target network address, wherein, this is, according to above-mentioned query word and user's portrait generation, to be carried on title
There is target network address network address to identify；
During due to user's portrait to be incorporated the generation of search summary and is shown, plucked user's portrait as generation search
The foundation wanted, the finally search in the displaying of show area 332 of search summary is made a summary and more meets search custom, preference, the concern of user
Point, user can according to search summary can be fast and accurately identified the link whether the link for needed for user, specifically
, when user terminal receives user's click commands, the corresponding web page contents of target network address are opened and show, now user
May browse through whole web page contents；
Further, before the corresponding web page contents of display target network address, above-mentioned search summary display module 308, also
For being marked to content related to search summary in web page contents, the mark includes：Overstriking mark, color mark, under
One or more in camber line mark；Then, then show mark after the corresponding web page contents of target network address, can so make
Obtain user and quickly view the related content searched near summary；
In addition, the display order of the multiple target network address related to query word and the user of user in order to obtain are drawn a portrait
Associated, above-mentioned search summary display module 308 is closed specifically for being drawn a portrait according to the corresponding web page contents of target network address and user
Connection degree order from high to low shows the corresponding above-mentioned search summary of each target network address successively, in such manner, it is possible to so that user is excellent
First check and user's maximally related target network address of portrait.
In search summarization generation device provided in an embodiment of the present invention, first, the query word phase with user input is obtained
The target network address of pass；And user's portrait of user is obtained, the search that the user draws a portrait for describing user is accustomed to, search is inclined
At least one in good, search rate, consumption habit, search focus；Finally, drawn a portrait according to above-mentioned query word and above-mentioned user
The corresponding search summary of the target network address is generated, and shows that the search is made a summary.User's portrait is incorporated search by the embodiment of the present invention
During the generation and displaying of summary, the foundation that user's portrait is made a summary as generation search can be effectively ensured searching for generation
Rope summary more meets search custom, preference, the focus of user such that it is able to so that quick by user makes a summary according to search,
Identify exactly the link whether the link for needed for user, and then improve user search efficiency and accuracy；Further
, it is contemplated that what user's history behavioral data mainly reflected is to big datas such as user's early stage search customs, to the current of user
The sensitiveness of intention is slightly weak, therefore, the contextual information of current search behavior is increased during the user for setting up user draws a portrait,
Can the more accurately current intention of identifying user, so as to further increase the precision of the search summary of final presentation；
Further, different weighted values are set by for each feature tag in user's portrait, with reference to each feature tag
Weighted value determines the degree of association between each candidate summary sentence and user's portrait, and according to degree of association order from high to low to each
Individual candidate's summary sentence is ranked up, and finally, chooses candidate's summary sentence searching as final display of the forward predetermined number of sequence
Rope is made a summary, in such manner, it is possible to further lift search summary specific aim and the accuracy for ultimately generating.
The search summarization generation device that the embodiment of the present invention is provided can be equipment on specific hardware or be installed on
Software or firmware in equipment etc..The technique effect of the device that the embodiment of the present invention is provided, its realization principle and generation and preceding
State embodiment of the method identical, to briefly describe, device embodiment part does not refer to part, refers to phase in preceding method embodiment
Answer content.It is apparent to those skilled in the art that, for convenience and simplicity of description, system described above,
The specific work process of device and unit, may be referred to the corresponding process in above method embodiment, will not be repeated here.
In embodiment provided by the present invention, it should be understood that disclosed apparatus and method, can be by other sides
Formula is realized.Device embodiment described above is only schematical, for example, the division of the unit, only one kind are patrolled
Collect function to divide, there can be other dividing mode when actually realizing, but for example, multiple units or component can combine or can
To be integrated into another system, or some features can be ignored, or not perform.It is another, it is shown or discussed each other
Coupling or direct-coupling or communication connection can be the INDIRECT COUPLING or communication link of device or unit by some communication interfaces
Connect, can be electrical, mechanical or other forms.
The unit that is illustrated as separating component can be or may not be it is physically separate, it is aobvious as unit
The part for showing can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple
On NE.Some or all of unit therein can be according to the actual needs selected to realize the mesh of this embodiment scheme
's.
In addition, during each functional unit in the embodiment that the present invention is provided can be integrated in a processing unit, also may be used
Being that unit is individually physically present, it is also possible to which two or more units are integrated in a unit.
If the function is to realize in the form of SFU software functional unit and as independent production marketing or when using, can be with
Storage is in a computer read/write memory medium.Based on such understanding, technical scheme is substantially in other words
The part contributed to prior art or the part of the technical scheme can be embodied in the form of software product, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used to so that a computer equipment (can be individual
People's computer, server, or network equipment etc.) perform all or part of step of each embodiment methods described of the invention.
And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (Read-Only Memory, ROM), arbitrary access are deposited
Reservoir (Random Access Memory, RAM), magnetic disc or CD etc. are various can be with the medium of store program codes.
It should be noted that：Similar label and letter represents similar terms in following accompanying drawing, therefore, once a certain Xiang Yi
It is defined in individual accompanying drawing, then it need not be further defined and explained in subsequent accompanying drawing, additionally, term " the
One ", " second ", " the 3rd " etc. are only used for distinguishing description, and it is not intended that indicating or implying relative importance.
Finally it should be noted that：Embodiment described above, specific embodiment only of the invention, is used to illustrate the present invention
Technical scheme, rather than its limitations, protection scope of the present invention is not limited thereto, although with reference to the foregoing embodiments to this hair
It is bright to be described in detail, it will be understood by those within the art that：Any one skilled in the art
The invention discloses technical scope in, it can still modify to the technical scheme described in previous embodiment or can be light
Change is readily conceivable that, or equivalent is carried out to which part technical characteristic；And these modifications, change or replacement, do not make
The essence of appropriate technical solution departs from the spirit and scope of embodiment of the present invention technical scheme.Should all cover in protection of the invention
Within the scope of.Therefore, protection scope of the present invention described should be defined by scope of the claims.
Claims (13)
1. it is a kind of to search for abstraction generating method, it is characterised in that methods described includes：
Acquisition according to the query word of user input retrieve the target network address related to the query word for obtaining；
User's portrait of the user is obtained, the user draws a portrait for describing the search custom of the user, searching preferences, searching
At least one in rope frequency, consumption habit, search focus；
According to the query word and the corresponding search summary of the user portrait generation target network address；
Show the corresponding search summary of each described target network address.
2. method according to claim 1, it is characterised in that before user's portrait of the acquisition user, also wrap
Include：
Obtain the historical behavior data of the user, the historical behavior data include user basic information, history logon information,
Historical search record, historical viewings record, history collection record, history Download History, historical review record and history consumption note
At least one in record；And/or,
Obtain the contextual information of user's current search behavior, the contextual information include current time, current location,
At least one in current environment, the user webpage that the word of search, the user browse recently recently；
Historical behavior data and/or the contextual information according to the user set up user's portrait of the user.
3. method according to claim 1, it is characterised in that described according to the query word and user portrait generation
The corresponding search of the target network address is made a summary, including：
According to the query word, multiple candidate's summary sentences are extracted in the document in the corresponding webpage of the target network address；
The weighted value of each feature tag in user's portrait is set；
According to candidate's summary sentence and user portrait degree of association order from high to low according to the weighted value of each feature tag
Multiple candidate's summary sentences are ranked up；
The candidate summary sentence generation target network address for choosing the forward predetermined number of sequence searches for summary accordingly.
4. method according to claim 3, it is characterised in that described according to the query word, in the target network address pair
Multiple candidate's summary sentences are extracted in document in the webpage answered, including：
Participle is carried out to the document in the corresponding webpage of the target network address；
The query word is matched with the document after participle, is calculated what the query word occurred within said document
Position；And the near synonym of the query word are matched with the document after participle, calculate the near of the query word
The position that adopted word occurs within said document；
Within said document, on the basis of the position that the position for occurring by the query word and the near synonym of the query word occur,
The multiple candidate's summary sentences of interception.
5. method according to claim 3, it is characterised in that each feature tag in setting user's portrait
Weighted value, including：
Calculate the operation concentration degree of each feature tag in being drawn a portrait to the user in user's history；
The weighted value of each feature tag is determined according to the operation concentration degree.
6. method according to claim 3, it is characterised in that it is described according to candidate's summary sentence and user draw a portrait the degree of association by
High to Low order is ranked up according to the weighted value of each feature tag to multiple candidate's summary sentences, including：
The degree of association is drawn a portrait from high to low with user according to candidate's summary sentence using Logic Regression Models or gradient lifting tree-model
Order multiple candidates' summary sentences are ranked up according to the weighted value of each feature tag.
7. it is a kind of to search for summarization generation device, it is characterised in that described device includes：
Target website acquisition module, the query word for obtaining according to user input retrieve obtaining with the query word phase
The target network address of pass；
User's portrait acquisition module, the user for obtaining the user draws a portrait, and the user draws a portrait for describing the user
Search custom, searching preferences, search rate, consumption habit, search for focus at least one；
Search summarization generation module, for being generated according to the query word and user portrait, the target network address is corresponding to search
Rope is made a summary.
8. device according to claim 7, it is characterised in that described device also includes：
Historical behavior data acquisition module, the historical behavior data for obtaining the user, the historical behavior data include
Recorded under user basic information, history logon information, historical search record, historical viewings record, history collection record, history
At least one in record, historical review record and history consumer record；And/or,
Contextual information acquisition module, the contextual information for obtaining user's current search behavior, the context letter
Breath includes current time, current location, current environment, the user net that the word of search, the user browse recently recently
At least one in page；
User's portrait sets up module, and institute is set up for the historical behavior data and/or the contextual information according to the user
State user's portrait of user.
9. device according to claim 7, it is characterised in that the search summarization generation module includes：
Candidate's summary sentence extracting sub-module, for according to the query word, the document in the corresponding webpage of the target network address
In extract multiple candidates summary sentences；
Feature tag weighted value determination sub-module, the weighted value for setting each feature tag in user's portrait；
Candidate makes a summary sentence sorting sub-module, for according to candidate's summary sentence and user portrait degree of association order from high to low according to
The weighted value of each feature tag is ranked up to multiple candidate's summary sentences；
Search summarization generation submodule, the candidate summary sentence for choosing the forward predetermined number of sequence generates the target
Network address searches for summary accordingly.
10. device according to claim 9, it is characterised in that the candidate makes a summary sentence extracting sub-module, specifically for right
Document in the corresponding webpage of the target network address carries out participle；The document after the query word and participle is carried out
Match somebody with somebody, calculate the position that the query word occurs within said document；And by after the near synonym of the query word and participle
The document is matched, and calculates the position that the near synonym of the query word occur within said document；Within said document,
On the basis of the position that the position occurred by the query word and the near synonym of the query word occur, the multiple candidate's summaries of interception
Sentence.
11. devices according to claim 9, it is characterised in that the feature tag weighted value determination sub-module, it is specific to use
The operation concentration degree of each feature tag in being drawn a portrait to the user in calculating user's history；According in the operation set
Degree determines the weighted value of each feature tag.
12. devices according to claim 9, it is characterised in that candidate's summary sentence sorting sub-module, specifically for adopting
Order of the tree-model according to candidate's summary sentence with user's portrait degree of association from high to low is lifted with Logic Regression Models or gradient
The weighted value according to each feature tag is ranked up to multiple candidate's summary sentences.
13. devices according to claim 7, it is characterised in that described device also includes：
Search summary display module, for showing the corresponding search summary of each described target network address, the search summary
It is according to the query word and user's portrait generation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611070196.7A CN106776860A (en) | 2016-11-28 | 2016-11-28 | One kind search abstraction generating method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611070196.7A CN106776860A (en) | 2016-11-28 | 2016-11-28 | One kind search abstraction generating method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106776860A true CN106776860A (en) | 2017-05-31 |
Family
ID=58904161
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611070196.7A Pending CN106776860A (en) | 2016-11-28 | 2016-11-28 | One kind search abstraction generating method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106776860A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107277630A (en) * | 2017-07-20 | 2017-10-20 | 海信集团有限公司 | The display methods and device of information of voice prompt |
| CN107832414A (en) * | 2017-11-07 | 2018-03-23 | 百度在线网络技术（北京）有限公司 | Method and apparatus for pushed information |
| CN108415970A (en) * | 2018-02-08 | 2018-08-17 | 北京三快在线科技有限公司 | Retrieval result sort method, device, electronic equipment and storage medium |
| CN108920717A (en) * | 2018-07-27 | 2018-11-30 | 百度在线网络技术（北京）有限公司 | For showing the method and device of information |
| CN109325180A (en) * | 2018-09-21 | 2019-02-12 | 北京字节跳动网络技术有限公司 | Article abstract method for pushing, device, terminal device, server and storage medium |
| WO2019128364A1 (en) * | 2017-12-29 | 2019-07-04 | 北京京东尚科信息技术有限公司 | Object searching method and system, computer system and computer readable storage medium |
| CN110147505A (en) * | 2017-08-24 | 2019-08-20 | 北京国双科技有限公司 | A kind of page display method, server and storage medium |
| CN110245289A (en) * | 2019-05-20 | 2019-09-17 | 中国平安财产保险股份有限公司 | A kind of information search method and relevant device |
| CN110929021A (en) * | 2018-08-31 | 2020-03-27 | 阿里巴巴集团控股有限公司 | Text information generating method and text information generating device |
| CN111080339A (en) * | 2019-11-18 | 2020-04-28 | 口口相传(北京)网络技术有限公司 | Method and device for generating category preference data based on scene |
| CN112182451A (en) * | 2020-09-18 | 2021-01-05 | 武汉绿色网络信息服务有限责任公司 | Webpage content abstract generation method, equipment, storage medium and device |
| CN112825076A (en) * | 2019-11-20 | 2021-05-21 | 北京搜狗科技发展有限公司 | Information recommendation method and device and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102163229A (en) * | 2011-04-13 | 2011-08-24 | 北京百度网讯科技有限公司 | Method and equipment for generating abstracts of searching results |
| CN102184221A (en) * | 2011-05-06 | 2011-09-14 | 北京航空航天大学 | Real-time video abstract generation method based on user preferences |
| CN104090886A (en) * | 2013-12-09 | 2014-10-08 | 深圳市腾讯计算机系统有限公司 | Method and device for constructing real-time portrayal of user |
| CN104268142A (en) * | 2014-08-05 | 2015-01-07 | 河海大学 | Meta search result ranking algorithm based on rejection strategy |
| CN105447191A (en) * | 2015-12-21 | 2016-03-30 | 北京奇虎科技有限公司 | Intelligent abstracting method for providing graphic guidance steps and corresponding device |
-
2016
- 2016-11-28 CN CN201611070196.7A patent/CN106776860A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102163229A (en) * | 2011-04-13 | 2011-08-24 | 北京百度网讯科技有限公司 | Method and equipment for generating abstracts of searching results |
| CN102184221A (en) * | 2011-05-06 | 2011-09-14 | 北京航空航天大学 | Real-time video abstract generation method based on user preferences |
| CN104090886A (en) * | 2013-12-09 | 2014-10-08 | 深圳市腾讯计算机系统有限公司 | Method and device for constructing real-time portrayal of user |
| CN104268142A (en) * | 2014-08-05 | 2015-01-07 | 河海大学 | Meta search result ranking algorithm based on rejection strategy |
| CN105447191A (en) * | 2015-12-21 | 2016-03-30 | 北京奇虎科技有限公司 | Intelligent abstracting method for providing graphic guidance steps and corresponding device |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107277630A (en) * | 2017-07-20 | 2017-10-20 | 海信集团有限公司 | The display methods and device of information of voice prompt |
| CN110147505A (en) * | 2017-08-24 | 2019-08-20 | 北京国双科技有限公司 | A kind of page display method, server and storage medium |
| CN107832414B (en) * | 2017-11-07 | 2021-10-22 | 百度在线网络技术（北京）有限公司 | Method and device for pushing information |
| CN107832414A (en) * | 2017-11-07 | 2018-03-23 | 百度在线网络技术（北京）有限公司 | Method and apparatus for pushed information |
| WO2019128364A1 (en) * | 2017-12-29 | 2019-07-04 | 北京京东尚科信息技术有限公司 | Object searching method and system, computer system and computer readable storage medium |
| CN108415970A (en) * | 2018-02-08 | 2018-08-17 | 北京三快在线科技有限公司 | Retrieval result sort method, device, electronic equipment and storage medium |
| CN108920717A (en) * | 2018-07-27 | 2018-11-30 | 百度在线网络技术（北京）有限公司 | For showing the method and device of information |
| CN110929021B (en) * | 2018-08-31 | 2024-05-31 | 阿里巴巴集团控股有限公司 | Text information generation method and text information generation device |
| CN110929021A (en) * | 2018-08-31 | 2020-03-27 | 阿里巴巴集团控股有限公司 | Text information generating method and text information generating device |
| CN109325180A (en) * | 2018-09-21 | 2019-02-12 | 北京字节跳动网络技术有限公司 | Article abstract method for pushing, device, terminal device, server and storage medium |
| CN109325180B (en) * | 2018-09-21 | 2021-01-05 | 北京字节跳动网络技术有限公司 | Article abstract pushing method and device, terminal equipment, server and storage medium |
| CN110245289A (en) * | 2019-05-20 | 2019-09-17 | 中国平安财产保险股份有限公司 | A kind of information search method and relevant device |
| CN111080339A (en) * | 2019-11-18 | 2020-04-28 | 口口相传(北京)网络技术有限公司 | Method and device for generating category preference data based on scene |
| CN111080339B (en) * | 2019-11-18 | 2024-01-30 | 口口相传(北京)网络技术有限公司 | Scene-based category preference data generation method and device |
| CN112825076A (en) * | 2019-11-20 | 2021-05-21 | 北京搜狗科技发展有限公司 | Information recommendation method and device and electronic equipment |
| CN112825076B (en) * | 2019-11-20 | 2024-03-01 | 北京搜狗科技发展有限公司 | Information recommendation method and device and electronic equipment |
| CN112182451A (en) * | 2020-09-18 | 2021-01-05 | 武汉绿色网络信息服务有限责任公司 | Webpage content abstract generation method, equipment, storage medium and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106776860A (en) | One kind search abstraction generating method and device | |
| US10902077B2 (en) | Search result aggregation method and apparatus based on artificial intelligence and search engine | |
| US9348935B2 (en) | Systems and methods for augmenting a keyword of a web page with video content | |
| KR100650404B1 (en) | On-line Advertising System And Method | |
| CN107862553A (en) | Advertisement real-time recommendation method, device, terminal device and storage medium | |
| CN102693271B (en) | A kind of network information recommending method and system | |
| US8209616B2 (en) | System and method for interfacing a web browser widget with social indexing | |
| US9798820B1 (en) | Classification of keywords | |
| US20140172415A1 (en) | Apparatus, system, and method of providing sentiment analysis result based on text | |
| US20130054672A1 (en) | Systems and methods for contextualizing a toolbar | |
| US20130054356A1 (en) | Systems and methods for contextualizing services for images | |
| CA3070612A1 (en) | Click rate estimation | |
| CN103339623A (en) | Internet search related methods and apparatus | |
| CN103092923A (en) | Menu-based advertisement of search engine | |
| US11681765B2 (en) | System and method for integrating content into webpages | |
| CN103365839A (en) | Recommendation search method and device for search engines | |
| US11113722B2 (en) | Providing content related to sentiment of product feature | |
| US20120246561A1 (en) | Systems and methods for extended content harvesting for contextualizing | |
| WO2012051470A1 (en) | Systems and methods for using a behavior history of a user to augment content of a webpage | |
| US20100318427A1 (en) | Enhancing database management by search, personal search, advertising, and databases analysis efficiently using core-set implementations | |
| US11698904B2 (en) | Query rewrite for low performing queries based on customer behavior | |
| CN107688616A (en) | Show unique fact of entity | |
| CN103718178A (en) | Utilization of features extracted from structured documents to improve search relevance | |
| CN103412958A (en) | Display method and device for searching result | |
| KR20050050016A (en) | On-line advertising system and method |
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 |