CN105512178B - A kind of entity recommended method and device - Google Patents

A kind of entity recommended method and device Download PDF

Info

Publication number
CN105512178B
CN105512178B CN201510828555.XA CN201510828555A CN105512178B CN 105512178 B CN105512178 B CN 105512178B CN 201510828555 A CN201510828555 A CN 201510828555A CN 105512178 B CN105512178 B CN 105512178B
Authority
CN
China
Prior art keywords
entity
recommended
sets
according
description information
Prior art date
Application number
CN201510828555.XA
Other languages
Chinese (zh)
Other versions
CN105512178A (en
Inventor
黄际洲
周里成
Original Assignee
百度在线网络技术(北京)有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 百度在线网络技术(北京)有限公司 filed Critical 百度在线网络技术(北京)有限公司
Priority to CN201510828555.XA priority Critical patent/CN105512178B/en
Publication of CN105512178A publication Critical patent/CN105512178A/en
Application granted granted Critical
Publication of CN105512178B publication Critical patent/CN105512178B/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor

Abstract

The embodiment of the invention provides a kind of entity recommended method and devices.On the one hand, the embodiment of the present invention obtains at least two entity sets by the query word according to acquisition;To generate corresponding description information for each entity sets according to the common trait of each entity sets and the query word;In turn, the corresponding description information of each entity sets is shown in the search results pages to match with the query word, and, the description information clicked according to user shows each recommended entity that the clicked corresponding entity sets of description information are included.Therefore, technical solution provided in an embodiment of the present invention can provide the description information of multiple entity sets in search results pages, and each recommended entity that entity sets are included can be further displayed, the diversity of recommended entity in search results pages is improved, and improves the efficiency for obtaining recommended entity from search results pages.

Description

A kind of entity recommended method and device

[technical field]

The present invention relates to the Internet search technology field more particularly to a kind of entity recommended method and devices.

[background technique]

Currently, user is after input inquiry word (Query), search engine is scanned for according to the query word, and to user Export several search results pages to match with the query word.In the prior art, it can be inserted in the search results pages of display One recommendation card, for recommending several recommended entities related to query word and single classification, user can recommend card One entity of middle click, the search results pages to match with the entity of the click are jumped to from current search result page.

However, the recommended entity exported at present is often the single related entities of classification, and recommended entity and query word Between correlativity it is indefinite, it is therefore desirable to user carried out in recommended entity artificial screening and search required for entity, Cause the classification of recommended entity in the prior art single, and the efficiency of acquisition recommended entity is relatively low from search results pages.

[summary of the invention]

In view of this, being improved in search results pages the embodiment of the invention provides a kind of entity recommended method and device The diversity of recommended entity, and improve the efficiency that recommended entity is obtained from search results pages.

The one side of the embodiment of the present invention provides a kind of entity recommended method, comprising:

According to the query word of acquisition, at least two entity sets are obtained;

According to the common trait of each entity sets and the query word, corresponding description information is generated for each entity sets;

The corresponding description information of each entity sets is shown in the search results pages to match with the query word, and, The description information clicked according to user shows that each recommendation that the clicked corresponding entity sets of description information are included is real Body.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the basis obtain The query word obtained obtains at least two entity sets, comprising:

According to the searching entities for including in the query word, temperature is obtained most in entity relevant to described search entity At least two high entities;And at least two entity is clustered, to obtain at least two entity sets; Alternatively,

According to the searching entities and search need information for including in the query word, obtain and described search entity and described At least two candidate collections that search need information matches are related to described search demand information according to each candidate collection Degree, obtains at least two entity sets from least two candidate collections.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation is looked into described It askes in the search results pages that word matches and shows the corresponding description information of each entity sets, comprising:

Obtain the Target Photo to match with each entity sets;

In the search results pages to match with the query word, carry out the corresponding description information of each entity sets with it is described The corresponding display of Target Photo.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation obtains and each reality The Target Photo that body set matches, comprising:

For each entity sets, the Target Photo to match with the entity sets is obtained;Alternatively,

For each entity sets, at least two candidate pictures to match with the entity sets are obtained, to institute It states at least two candidate pictures and carries out picture mosaic, to generate the Target Photo to match with the entity sets.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, according to user institute The description information of click shows each recommended entity that the clicked corresponding entity sets of description information are included, comprising:

Each recommended entity that the corresponding entity sets of the description information clicked using user are included carries out page rendering, To show the new page comprising each recommended entity.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the new page It is middle to show each recommended entity using the layout of a page without columns;The method also includes:

The recommended entity clicked in each recommended entity shown in the layout of a page without columns according to user, in the new page The search result that display matches with the recommended entity clicked below the layout of a page without columns.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the method is also Include:

The navigation information between clicked recommended entity and the query word is being shown in the layout of a page without columns;

Wherein, the navigation information includes the corresponding description letter of entity sets where the query word, the recommended entity Breath and the recommended entity clicked.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, according to user institute The description information of click shows each recommended entity that the clicked corresponding entity sets of description information are included, comprising:

The description information clicked according to user describes letter described in the search results pages to match with the query word Breath lower section shows combobox, and, the corresponding entity sets institute of description information that user is clicked is shown using the combobox Each recommended entity for including.

The one side of the embodiment of the present invention provides a kind of entity recommendation apparatus, comprising:

Entity obtains module, for the query word according to acquisition, obtains at least two entity sets;

Data obtaining module is each entity set symphysis for the common trait according to each entity sets and the query word At corresponding description information;

Recommend display module, for showing that each entity sets are corresponding in the search results pages to match with the query word Description information, and, the description information clicked according to user shows the clicked corresponding entity sets institute of description information Each recommended entity for including.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the entity obtain Modulus block, is specifically used for:

According to the searching entities for including in the query word, temperature is obtained most in entity relevant to described search entity At least two high entities;And at least two entity is clustered, to obtain at least two entity sets; Alternatively,

According to the searching entities and search need information for including in the query word, obtain and described search entity and described At least two candidate collections that search need information matches are related to described search demand information according to each candidate collection Degree, obtains at least two entity sets from least two candidate collections.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, described device is also It include: image generating module, for obtaining the Target Photo to match with each entity sets;

The recommendation display module, is specifically used for: in the search results pages to match with the query word, carrying out each reality The display corresponding with the Target Photo of the corresponding description information of body set.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the picture are raw At module, it is specifically used for:

For each entity sets, the Target Photo to match with the entity sets is obtained;Alternatively,

For each entity sets, at least two candidate pictures to match with the entity sets are obtained, to institute It states at least two candidate pictures and carries out picture mosaic, to generate the Target Photo to match with the entity sets.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, it is described to recommend to show Show module, be specifically used for:

Each recommended entity that the corresponding entity sets of the description information clicked using user are included carries out page rendering, To show the new page comprising each recommended entity.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the new page It is middle to show each recommended entity using the layout of a page without columns;The recommendation display module, is also used to:

The recommended entity clicked in each recommended entity shown in the layout of a page without columns according to user, in the new page The search result that display matches with the recommended entity clicked below the layout of a page without columns.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, it is described to recommend to show Show module, be also used to: showing the navigation information between clicked recommended entity and the query word in the layout of a page without columns;

Wherein, the navigation information includes the corresponding description letter of entity sets where the query word, the recommended entity Breath and the recommended entity clicked.

The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, it is described to recommend to show Show module, be specifically used for:

The description information clicked according to user describes letter described in the search results pages to match with the query word Breath lower section shows combobox, and, the corresponding entity sets institute of description information that user is clicked is shown using the combobox Each recommended entity for including.

As can be seen from the above technical solutions, the embodiment of the present invention has the advantages that

Technical solution provided in an embodiment of the present invention can provide the description letter of multiple entity sets in search results pages Breath, and each recommended entity that entity sets are included can be further displayed, it is single to provide classification in the prior art The technical solution of recommended entity compare, technical solution provided by the embodiment of the present invention can export more than multi-class, quantity Recommended entity, to improve the diversity of recommended entity in search results pages.Moreover, with recommended entity in the prior art with look into Ask recommends the indefinite technical solution of relationship to compare between word, passes through the corresponding description of output entity sets in the embodiment of the present invention Information can export the recommendation relationship between query word and recommended entity, and user is helped to be quickly obtained required entity, improve The efficiency of recommended entity is obtained from search results pages.

[Detailed description of the invention]

In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached Figure is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this field For those of ordinary skill, without any creative labor, it can also be obtained according to these attached drawings other attached Figure.

Fig. 1 is the flow diagram of entity recommended method provided by the embodiment of the present invention;

When Fig. 2 (a)~Fig. 2 (h) is that at least two candidate pictures provided by the embodiment of the present invention generate Target Photo Combination exemplary diagram;

Fig. 3 is the first exemplary diagram of the recommended entity in search results pages in the prior art;

Fig. 4 (a)~Fig. 4 (c) is the first example of the recommended entity in search results pages provided by the embodiment of the present invention Figure;

Fig. 5 (a)~Fig. 5 (b) is the second example of the recommended entity in search results pages provided by the embodiment of the present invention Figure;

Fig. 6 is the second exemplary diagram of the recommended entity in search results pages in the prior art;

Fig. 7 (a)~Fig. 7 (c) is the third example of the recommended entity in search results pages provided by the embodiment of the present invention Figure;

Fig. 8 (a)~Fig. 8 (b) is the 4th example of the recommended entity in search results pages provided by the embodiment of the present invention Figure;

Fig. 9 is the functional block diagram of entity recommendation apparatus provided by the embodiment of the present invention.

[specific embodiment]

For a better understanding of the technical solution of the present invention, being retouched in detail to the embodiment of the present invention with reference to the accompanying drawing It states.

It will be appreciated that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Base Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts it is all its Its embodiment, shall fall within the protection scope of the present invention.

The term used in embodiments of the present invention is only to be not intended to be limiting merely for for the purpose of describing particular embodiments The present invention.In the embodiment of the present invention and the "an" of singular used in the attached claims, " described " and "the" It is also intended to including most forms, unless the context clearly indicates other meaning.

It should be appreciated that term "and/or" used herein is only a kind of incidence relation for describing affiliated partner, indicate There may be three kinds of relationships, for example, A and/or B, can indicate: individualism A, exist simultaneously A and B, individualism B these three Situation.In addition, character "/" herein, typicallys represent the relationship that forward-backward correlation object is a kind of "or".

Depending on context, word as used in this " if " can be construed to " ... when " or " when ... When " or " in response to determination " or " in response to detection ".Similarly, depend on context, phrase " if it is determined that " or " if detection (condition or event of statement) " can be construed to " when determining " or " in response to determination " or " when the detection (condition of statement Or event) when " or " in response to detection (condition or event of statement) ".

Embodiment one

The embodiment of the present invention provides a kind of entity recommended method, referring to FIG. 1, it is real provided by the embodiment of the present invention The flow diagram of body recommended method, as shown, method includes the following steps:

S101 obtains at least two entity sets according to the query word of acquisition.

Specifically, can receive the query word of user's input in the embodiment of the present invention, then being searched according to the query word Rope shows several search to user finally by search results pages and ties to obtain several search results to match with query word Fruit.In the search results pages of generation, need to provide a user various dimensions, multi-class recommended entity simultaneously.In order to realize to User provides various dimensions, multi-class recommended entity, in the embodiment of the present invention, needs the query word first according to acquisition, obtains extremely Few two entity sets.

For example, according to the query word of acquisition, the method for obtaining at least two entity sets can in the embodiment of the present invention To include but is not limited to following two:

The first: according to the searching entities for including in the query word, obtaining in entity relevant to described search entity Obtain highest at least two entity of temperature;And at least two entity is clustered, it is real to obtain described at least two Body set.

It is understood that illustrate that basis is looked into if only comprising searching entities or query word being searching entities in query word The search need information of user can not be determined by asking word, therefore the search need of user is indefinite, then can from described search reality In the relevant all entities of body, highest at least two entity of temperature is obtained.

Further, it is possible to carry out cluster operation using clustering algorithm at least two entity highest to temperature, by described in extremely Few two entities are divided at least two entity sets, to realize the query word according to acquisition, obtain at least two entity sets.

Second: according to the searching entities and search need information for including in the query word, obtaining real with described search At least two candidate collections that body and described search demand information match, according to each candidate collection and described search demand information The degree of correlation, at least two entity sets are obtained from least two candidate collections.

It is understood that illustrating can be with according to query word if in query word including searching entities and search need information It determines the search need information of user, therefore the search need of user is clear, then can be believed according to searching entities and search need Breath obtains and highest at least two candidate collection of search need information correlation.

, can be according to the degree of correlation of each candidate collection and search need information during a concrete implementation, and press According to the sequence that the degree of correlation is descending, each candidate collection is ranked up, to obtain ranking results, is finally obtained from ranking results Highest at least two candidate collection of sequence is taken, using as at least two entity sets.

For example, the search need of user is indefinite, when the query word of user's input is " Ha Shiqi " in order to excite user's Potential search need can obtain and " Ha Shiqi " degree of correlation highest, the highest several entity sets of temperature, such as " and Ha Shiqi The same IQ allows people to catch anxious dog dog ", " dog of wolf is looked like as Ha Shiqi ", " as the Ha Shiqi nothing of personality docility Attack medium-sized dog ".

In another example the search need of user is very clear, i.e., to Kazakhstan when the query word of user's input is " Ha Shiqi is supported well " The feeding manner of scholar's surprise and other related pets is interested, if can obtain highest with the search need information correlation of user Dry entity sets, such as " the medium-sized dog of more difficult raising as Ha Shiqi ", " and Ha Shiqi is suitble to domestic small-sized pet on the contrary Dog ", " the loyal trained pet of understanding as Ha Shiqi ".

In this method, the different search needs of the query word of user's input are divided, will be matched with search need , user's entity interested be orderly grouped according to classification, different entity sets are formed, finally by these entity sets User is recommended, for meeting the different search need of user, wherein the sequence of each entity sets can according to entity sets with The degree of correlation of search need, the highest entity set of the search need degree of correlation with user are directed between the query word of user's input It closes, comes near preceding position.

S102 generates corresponding description according to the common trait of each entity sets and the query word for each entity sets Information.

Specifically, in the embodiment of the present invention after obtaining at least two entity sets, each entity sets can be extracted and looked into The common trait of word is ask, then according to the common trait of each entity sets and query word, generates corresponding retouch for each entity sets State information.

It should be noted that generate description information when, need to construct and embody in description information with query word or Existing relationship between search intention information.For example, query word is " Beethovan ", it include recommended entity in entity sets, such as " suddenly Gold " and " Vincent van gogh " etc., the common trait of the entity sets and query word are " broken in body but firm in spirit famous person ", then the entity Gathering corresponding description information can be " the broken in body but firm in spirit famous person as Beethovan ".

In addition, the corresponding description information of entity sets can not only embody the common spy between entity sets and search term Sign, also embodies the reason of recommending the entity sets to user, allows users to more fully understand include in entity sets each Recommended entity, and more fully understand relationship or entity set between the entity sets of recommendation and the query word of user's search The relationship between the search intention of user is closed, to allow users to quickly select interested or required reality Body, and then improve the efficiency of recommended entity required for user obtains.

S103 shows the corresponding description information of each entity sets in the search results pages to match with the query word.

Specifically, in the embodiment of the present invention, after generating corresponding description information for each entity sets, can display with The corresponding description information of each entity sets is shown in the search results pages that the query word matches.For example, can with it is described In the search results pages that query word matches, the middle position of any two search results shows that each entity sets are corresponding and retouches State information.

During a concrete implementation, it is also possible to obtain the Target Photo to match with each entity sets;Then, exist In the search results pages to match with the query word, the corresponding description information of each entity sets and the Target Photo are carried out Corresponding display.

For example, level shows the corresponding description information of each entity sets in the search results pages to match with query word, And the lower position of each description information in display, show the target that entity sets corresponding with the description information match Picture.It, can be by each entity sets pair by the display corresponding with the Target Photo of the corresponding description information of each entity sets The description information answered is associated with corresponding Target Photo, and user is allow to directly perceive the entity set of recommendation by Target Photo It closes.

For example, the method for obtaining the Target Photo to match with each entity sets can wrap in the embodiment of the present invention It includes but is not limited to following two:

The first: for each entity sets, obtaining the Target Photo to match with the entity sets.

Second: for each entity sets, obtaining at least two candidate figures to match with the entity sets Piece carries out picture mosaic to described at least two candidate pictures, to generate the Target Photo to match with the entity sets.

During a concrete implementation, it can will polymerize with more figures that entity sets match, such as can be The polymerization of three candidate pictures, four candidate pictures or five candidate pictures, it might even be possible to be more representative lists The polymerization of the picture of entity.It is polymerize using more pictures and obtains Target Photo, more recommended entities in the entity sets can be transmitted Picture, user can also be enhanced, the recommendations of entity sets is perceived.

For example, can choose in the entity sets, when generating the Target Photo of more figure polymerization patterns for entity sets if preceding In dry recommended entity then these pictures are carried out arrangement polymerization by picture corresponding to each recommended entity.Alternatively, in another example, Also it can choose the picture that can most reflect search need representated by the entity sets in entity sets, and in the view of picture Similar picture is polymerize in the features such as feel, style, color, generates Target Photo.

Fig. 2 (a)~Fig. 2 (h) is please referred to, is that at least two candidate pictures provided by the embodiment of the present invention generate target Combination exemplary diagram when picture, as shown in Fig. 2 (a)~Fig. 2 (f), at least two candidate pictures can be multiple candidate figures Piece combination producing Target Photo.As shown in Fig. 2 (g), at least two candidate pictures can also be subjected to shade superposition and generate target Picture.As shown in Fig. 2 (h), at least two candidate pictures can also be subjected to 3D combination producing Target Photo.It needs to illustrate Be, it is above-mentioned by least two candidate pictures generate Target Photos in the way of by way of example only, the embodiment of the present invention is to the life At mode without being particularly limited to.

S104, the description information clicked according to user show that the clicked corresponding entity sets of description information are wrapped Each recommended entity contained.

Specifically, in the embodiment of the present invention, due to being shown in the search results pages that display matches with the query word Each entity sets corresponding description informations, therefore, if the entity sets understood required for having user in each entity sets, User can be slided to the left using gesture in the corresponding description information of each entity sets or slide entity sets to the right, from And it may browse through the description information of each entity sets.When user is interesting to some entity sets, user clicks the entity The description information of set, in this way, the description information that can be clicked according to user, further displays clicked description information pair Each recommended entity that the entity sets answered are included, in this way, each recommended entity for being included by the entity sets is supplied to user.

For example, in the embodiment of the present invention, the description information clicked according to user shows clicked description information The method for each recommended entity that corresponding entity sets are included can include but is not limited to following two:

The first: each recommended entity that the corresponding entity sets of the description information clicked using user are included carries out page Face rendering, to show the new page comprising each recommended entity.

During a concrete implementation, each recommended entity can be shown using the layout of a page without columns in the new page;Then, The recommended entity clicked in each recommended entity that can be shown in the layout of a page without columns according to user, described in the new page The search result that display matches with the recommended entity clicked below the layout of a page without columns.

It is understood that after user selects a description information in the corresponding description information of each entity sets, it can To carry out primary new search according to the corresponding entity sets of the description information clicked, and it is aobvious in the top position of new page Show that a layout of a page without columns, the layout of a page without columns can be fixed on the top of search results pages, in this way, user can in the layout of a page without columns fast browsing institute point All recommended entities for including in the corresponding entity sets of the description information hit.

User can further click on required recommended entity, and the recommended entity clicked according to user can pass through Asynchronous loading technique renders the search result to match with the recommended entity clicked automatically in new page below the layout of a page without columns, Wherein, clicking recommended entity, rendering the search result to match with the recommended entity clicked is all in the new page It carries out, does not need to jump to other pages, there is no need to toggling between the page, so as to easily click institute The recommended entity needed, and it is quickly obtained the search result to match with the recommended entity clicked, it is multiple to substantially increase acquisition The efficiency of recommended entity reduces search time.

Optionally, can also believe showing the navigation between clicked recommended entity and the query word in the layout of a page without columns Breath;Wherein, the navigation information include the query word, the corresponding description information of entity sets where the recommended entity and The recommended entity clicked.

It is understood that the layout of a page without columns can consist of two parts: navigation information and each recommended entity.Navigation information can With each recommended entity top being shown in the layout of a page without columns, quickly understand by query word convenient for user to the recommended entity clicked Recommend logic, clicked recommended entity is quickly positioned convenient for user.

For example, query word described in the navigation information is located at the left side, entity sets where the recommended entity are corresponding Description information is located at centre, and the recommended entity clicked is located at the right.For example, navigation information can be " Beethovan > and Beethovan Equally broken in body but firm in spirit famous person > Vincent van gogh ", wherein " Beethovan " is the query word of user's input, " and Beethovan one The famous person that sample is broken in body but firm in spirit " is the corresponding description information of entity sets that user is clicked in each entity sets, " Vincent Van gogh " is the recommendation information that user further clicks in each recommendation information that the entity sets include.

Information abstraction is carried out alternatively, it is also possible to the description information to entity sets, simplifies information as leading for after refinement Description information in information of navigating, such as navigation information or " the broken in body but firm in spirit famous person of Beethovan > similar > Vincent is of ancient India It is high ", to reduce information redundancy, improve the efficiency for obtaining information.

In this method, when showing each recommended entity in the layout of a page without columns, can be shown with horizontal direction the title of each recommended entity with And the corresponding picture of each recommended entity.Wherein it is possible to the degree of correlation according to each recommended entity and search need information, to each recommendation Entity is ranked up, and the higher recommended entity of the degree of correlation more comes forward position, user can in the layout of a page without columns sliding to the left or Person slides to browse each recommended entity to the right.Alternatively, it is also possible to be returned and inquired by the above-mentioned new page comprising each recommended entity The search results pages that word matches, and further select other entity sets.

Second: the description information clicked according to user, the institute in the search results pages to match with the query word It states and shows combobox below description information, and, the corresponding reality of description information that user is clicked is shown using the combobox Each recommended entity that body set is included.

During a concrete implementation, when user is interested according to selecting in the corresponding description information of each entity sets Entity sets when, can use slip gesture to the left or slip gesture move each entity sets to the right, then browse to Each entity sets can match when the user clicks when the corresponding description information of certain interested entity sets with query word Search results pages in show combobox below the corresponding description information of each entity sets, and then user institute is shown in combobox Each recommended entity included in the corresponding entity sets of the description information of click, for example, may include 12 in the entity sets ~20 recommended entities can be ranked up each recommended entity according to recommended entity and the degree of correlation of search need, the degree of correlation Higher recommended entity more comes forward position.

Further, in combobox, user also can use slip gesture to the left or slip gesture moves to the right Each recommended entity makes user can be clear in search results pages to browse to all recommended entities for including in the entity sets It lookes at multiple recommended entities into the entity sets and entity sets of each dimension, improves the efficiency for obtaining recommended entity.

Embodiment two

(the comprehensive expression of person of classicism music style, impoverished, the residual will of body as user input query word " Beethovan " Hard music artist), in the case where main search demand obtains satisfaction, other possible interested search needs of user are also wrapped Include: as Beethovan broken in body but firm in spirit famous person artist impoverished as Beethovan possess with Beethovan as create Make the musician etc. of style.

Referring to FIG. 3, it is the first exemplary diagram of the recommended entity in search results pages in the prior art, as shown, In conventional solution, to recommend " Mozart ", " Chopin ", " Shu Baite ", " Bach " etc. several around Beethovan for recommended entity Entity, these are all musicians well-known in the world, and the dimension of recommended entity is single and " related person " this description information pair Recommended entity it is explanatory weak, for user " the impoverished artist as Beethovan ", " body is residual as Beethovan The two search needs of the famous person of will heavily fortified point " are not able to satisfy.

Fig. 4 (a)~Fig. 4 (c) is please referred to, is the provided recommended entity in search results pages of the embodiment of the present invention First exemplary diagram.As shown in Fig. 4 (a), the entities such as " Hayden ", " Mozart ", " Chopin ", " Bach " are summarized in an entity set In conjunction, extracts common trait and generate description information " musician of creation style as possessing with Beethovan ", with traditional technology Scheme is compared, and increases the entity sets of compositions such as " van gogh, Huo Jin, Shi Tiesheng, Ah bright ", and increase description for the entity sets Information " the broken in body but firm in spirit famous person as Beethovan ", and increase " Andersen, Xu Beihong, Michelangelo, Shu Baite " The entity sets of equal compositions, and increase description information " the impoverished artist as Beethovan " for the entity sets, with And will most there be generation in each entity sets as shown in Fig. 4 (a) for the picture that these three entity sets are mixed respectively after more figure polymerizations The picture of three entities of table is combined, and compared with conventional solution, technical solution provided by the embodiment of the present invention is done To the upgrading for recommending to entity sets recommendation from entity.

As shown in Fig. 4 (b), when the musician of creation style is interested as user possesses other with Beethovan, point The description information " musician of creation style as possessing with Beethovan " for hitting entity sets, then can show new page, and Show the layout of a page without columns in new page, all recommended entities included in the entity sets shown in the layout of a page without columns, as Hayden, Mozart, Chopin, Bach etc..User clicks interested recommended entity by selection, and current page will show the search knot of the recommended entity Fruit.

Former search results pages can be returned when user wants to continue to browse other entity sets, and further click on description letter It ceases " the broken in body but firm in spirit famous person as Beethovan " or " the impoverished artist as Beethovan ", such as Fig. 4 (c) institute Show, can be jumped in the corresponding entity sets of display " the broken in body but firm in spirit famous person as Beethovan " by former search results pages again The new page for the recommended entity for being included.

Fig. 5 (a)~Fig. 5 (b) is please referred to, is the provided recommended entity in search results pages of the embodiment of the present invention Second exemplary diagram.As shown in Fig. 5 (a), if user is interested in " musician of creation style as possessing with Beethovan ", point The description information is hit, combobox is shown below each physical model shown in search results pages relevant to query word, under Draw and show each recommended entity in the corresponding entity sets of the description information in frame, such as " Mozart ", " Chopin ", " Hayden " and Recommended entities such as " Bach ".In addition, as shown in Fig. 5 (a), user can in each recommended entity using slip gesture to the left or Slip gesture browses to all recommended entities to the right.As shown in Fig. 5 (b), if user is to " broken in body but firm in spirit as Beethovan When famous person " is interested, the description information is clicked, below each physical model shown in search results pages relevant to query word It shows combobox, each recommended entity in the corresponding entity sets of the description information is shown in combobox, such as " van gogh ", " suddenly The recommended entities such as gold ", " Shi Tiesheng " and " Ah bright ".

Embodiment three

When user input query word " Ah sweet's main story ", (film of pursuing a goal with determination that Tom's hanks are acted the leading role, nineteen ninety-five obtain Oscar 6 Item Grand Prix), in the case where main search demand obtains satisfaction, other possible interested search needs of user further include: other Obtain other works that the cinematographic work, classical film of pursuing a goal with determination, Tom's hanks of Oscar are acted the leading role.

Referring to FIG. 6, it is the second exemplary diagram of the recommended entity in search results pages in the prior art, as shown, In conventional solution, recommended entity recommends " The Shawshank Redemption ", " Rudy ", " succeeds around Ah sweet's formal dress Several entities such as people ", these all be obtain Oscar famous movies, the dimension of recommended entity it is single and " associated recommendation " this Description information is not able to satisfy other search needs of user the explanatory weak of recommended entity.

Fig. 7 (a)~Fig. 7 (c) is please referred to, is the provided recommended entity in search results pages of the embodiment of the present invention Third exemplary diagram.As shown in Fig. 7 (a), by " bird people ", " leaping madhouse ", " Legends of the Fall ", " the magical drift of teenager group " etc. Entity is summarized in an entity sets, is extracted common trait and is generated description information " as Ah sweet's main story Austria at high box office This card award-winning work ".By the composition such as " The Shawshank Redemption ", " when happiness is knocked at the door ", " never giving up ", " Les Choristes " Entity sets, and increase description information " as Ah sweet's main story classical must see film of pursuing a goal with determination " for the entity sets.In addition, The entity sets of compositions such as " mermaids ", " Philips captain ", " The Philadelphia Story " and " Cast Away " can also be increased, and to be somebody's turn to do Entity sets increase description information " being all the popular film that Tom's hanks are acted the leading role ".

As shown in Fig. 7 (a), the picture after more figure polymerizations is mixed respectively for these three entity sets, it will be in each entity sets Most there is the picture of five entities of representative to be combined, compared with conventional solution, technology provided by the embodiment of the present invention Scheme has accomplished the upgrading that entity sets recommendation is recommended to from entity.

As shown in Fig. 7 (b), when user to other classics pursue a goal with determination film it is interested when, click the description information of entity sets " the classical film of pursuing a goal with determination that must be seen as Ah sweet's main story ", then can show new page, and show the layout of a page without columns in new page, lead to Show all recommended entities included in the entity sets in column, such as " The Shawshank Redemption ", " when happiness is knocked at the door ", " forever Do not abandon ", " Les Choristes " etc..User clicks interested recommended entity by selection, and current page will show that this is pushed away Recommend the search result of entity.

Former search results pages can be returned when user wants to continue to browse other entity sets, and further click on description letter It ceases " the Oscar award-winning work at high box office as Ah sweet's main story ", as shown in Fig. 7 (c), can be jumped again by former search results pages It goes to and recommends in fact included in the corresponding entity sets of display " the Oscar award-winning work at high box office as Ah sweet's main story " The new page of body.

Fig. 8 (a)~Fig. 8 (b) is please referred to, is the provided recommended entity in search results pages of the embodiment of the present invention 4th exemplary diagram.As shown in Fig. 8 (a), if user is interested in " the classical film of pursuing a goal with determination that must be seen as Ah sweet's main story ", point The description information is hit, combobox is shown below each physical model shown in search results pages relevant to query word, under Draw and show each recommended entity in the corresponding entity sets of the description information in frame, as " never compromising ", " Like Mike ", " after After someone " and the recommended entities such as " weakness ".As shown in Fig. 8 (b), if user is to " as the Ah sweet's main story Oscar at high box office When award-winning work " is interested, the description information is clicked, each physical model shown in search results pages relevant to query word Lower section shows combobox, and each recommended entity in the corresponding entity sets of the description information is shown in combobox, such as " bird people ", The recommended entities such as " stealing dream space ", " the magical drift of teenager group ", " imitating game ".

The embodiment of the present invention, which further provides, realizes the Installation practice of each step and method in above method embodiment.

Referring to FIG. 9, its functional block diagram for entity recommendation apparatus provided by the embodiment of the present invention.As shown, The device includes:

Entity obtains module 91, for the query word according to acquisition, obtains at least two entity sets;

Data obtaining module 92 is each entity sets for the common trait according to each entity sets and the query word Generate corresponding description information;

Recommend display module 93, for showing each entity sets pair in the search results pages to match with the query word The description information answered, and, the description information clicked according to user shows the clicked corresponding entity sets of description information Each recommended entity for being included.

During a concrete implementation, the entity obtains module 91, is specifically used for:

According to the searching entities for including in the query word, temperature is obtained most in entity relevant to described search entity At least two high entities;And at least two entity is clustered, to obtain at least two entity sets; Alternatively,

According to the searching entities and search need information for including in the query word, obtain and described search entity and described At least two candidate collections that search need information matches are related to described search demand information according to each candidate collection Degree, obtains at least two entity sets from least two candidate collections.

During a concrete implementation, described device can also include: image generating module 94, for obtaining and respectively The Target Photo that entity sets match;

The recommendation display module 93, is specifically used for: in the search results pages to match with the query word, carrying out each The display corresponding with the Target Photo of the corresponding description information of entity sets.

During a concrete implementation, the image generating module 94 is specifically used for:

For each entity sets, the Target Photo to match with the entity sets is obtained;Alternatively,

For each entity sets, at least two candidate pictures to match with the entity sets are obtained, to institute It states at least two candidate pictures and carries out picture mosaic, to generate the Target Photo to match with the entity sets.

During a concrete implementation, the recommendation display module 93 is specifically used for:

Each recommended entity that the corresponding entity sets of the description information clicked using user are included carries out page rendering, To show the new page comprising each recommended entity.

During a concrete implementation, each recommended entity is shown using the layout of a page without columns in the new page;It is described to recommend to show Show module 93, be also used to:

The recommended entity clicked in each recommended entity shown in the layout of a page without columns according to user, in the new page The search result that display matches with the recommended entity clicked below the layout of a page without columns.

During a concrete implementation, the recommendation display module 93 is also used to: showing institute's point in the layout of a page without columns Navigation information between the recommended entity hit and the query word;

Wherein, the navigation information includes the corresponding description letter of entity sets where the query word, the recommended entity Breath and the recommended entity clicked.

Alternatively, the recommendation display module 93 is specifically used for during a concrete implementation:

The description information clicked according to user describes letter described in the search results pages to match with the query word Breath lower section shows combobox, and, the corresponding entity sets institute of description information that user is clicked is shown using the combobox Each recommended entity for including.

Method shown in FIG. 1 is able to carry out by each unit in this present embodiment, the part that the present embodiment is not described in detail, It can refer to the related description to Fig. 1.

The technical solution of the embodiment of the present invention has the advantages that

In the embodiment of the present invention, by the query word according to acquisition, at least two entity sets are obtained;To according to each The common trait of entity sets and the query word generates corresponding description information for each entity sets;In turn, it is looked into described It askes in the search results pages that word matches and shows the corresponding description information of each entity sets, and, it is retouched according to what user was clicked Information is stated, shows each recommended entity that the clicked corresponding entity sets of description information are included.

Technical solution provided in an embodiment of the present invention can provide the description letter of multiple entity sets in search results pages Breath, and each recommended entity that entity sets are included can be further displayed, it is single to provide classification in the prior art The technical solution of recommended entity compare, technical solution provided by the embodiment of the present invention can export more than multi-class, quantity Recommended entity, to improve the diversity of recommended entity in search results pages.Moreover, with recommended entity in the prior art with look into Ask recommends the indefinite technical solution of relationship to compare between word, passes through the corresponding description of output entity sets in the embodiment of the present invention Information can export the recommendation relationship between query word and recommended entity, and user is helped to be quickly obtained required entity, improve The efficiency of recommended entity is obtained from search results pages.

It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description, The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.

In several embodiments provided by the present invention, it should be understood that disclosed system, device and method can be with It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit It divides, only a kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or group Part can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown Or the mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, device or unit it is indirect Coupling or communication connection can be electrical property, mechanical or other forms.

The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme 's.

It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list Member both can take the form of hardware realization, can also realize in the form of hardware adds SFU software functional unit.

The above-mentioned integrated unit being realized in the form of SFU software functional unit can store and computer-readable deposit at one In storage media.Above-mentioned SFU software functional unit is stored in a storage medium, including some instructions are used so that a computer It is each that device (can be personal computer, server or network equipment etc.) or processor (Processor) execute the present invention The part steps of embodiment the method.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (Read- Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic or disk etc. it is various It can store the medium of program code.

The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention Within mind and principle, any modification, equivalent substitution, improvement and etc. done be should be included within the scope of the present invention.

Claims (14)

1. a kind of entity recommended method, which is characterized in that the described method includes:
According to the query word of acquisition, at least two entity sets are obtained;
According to the common trait of each entity sets and the query word, corresponding description information is generated for each entity sets;
The corresponding description information of each entity sets is shown in the search results pages to match with the query word, and, according to The description information that user is clicked shows each recommended entity that the clicked corresponding entity sets of description information are included;
Wherein, the query word according to acquisition obtains at least two entity sets, comprising:
According to the searching entities for including in the query word, it is highest that temperature is obtained in entity relevant to described search entity At least two entities;And at least two entity is clustered, to obtain at least two entity sets;Alternatively,
According to the searching entities and search need information for including in the query word, obtain and described search entity and described search At least two candidate collections that demand information matches, according to the degree of correlation of each candidate collection and described search demand information, from At least two entity sets are obtained at least two candidate collections.
2. the method according to claim 1, wherein being shown in the search results pages to match with the query word Show the corresponding description information of each entity sets, comprising:
Obtain the Target Photo to match with each entity sets;
In the search results pages to match with the query word, the corresponding description information of each entity sets and the target are carried out The corresponding display of picture.
3. according to the method described in claim 2, it is characterized in that, the Target Photo that acquisition matches with each entity sets, packet It includes:
For each entity sets, the Target Photo to match with the entity sets is obtained;Alternatively,
For each entity sets, obtain at least two candidate pictures to match with the entity sets, to it is described extremely Few two candidate pictures carry out picture mosaic, to generate the Target Photo to match with the entity sets.
4. the method according to claim 1, wherein the description information clicked according to user, display is clicked The corresponding entity sets of description information each recommended entity for being included, comprising:
Each recommended entity that the corresponding entity sets of the description information clicked using user are included carries out page rendering, with exhibition It now include the new page of each recommended entity.
5. according to the method described in claim 4, it is characterized in that, showing each recommended entity using the layout of a page without columns in the new page; The method also includes:
The recommended entity clicked in each recommended entity shown in the layout of a page without columns according to user, described in the new page The search result that display matches with the recommended entity clicked below the layout of a page without columns.
6. according to the method described in claim 5, it is characterized in that, the method also includes:
The navigation information between clicked recommended entity and the query word is being shown in the layout of a page without columns;
Wherein, the navigation information include the query word, the corresponding description information of entity sets where the recommended entity with And the recommended entity clicked.
7. the method according to claim 1, wherein the description information clicked according to user, display is clicked The corresponding entity sets of description information each recommended entity for being included, comprising:
The description information clicked according to user, under the description information described in the search results pages to match with the query word Side's display combobox, and, included using the corresponding entity sets of description information that the combobox shows that user is clicked Each recommended entity.
8. a kind of entity recommendation apparatus, which is characterized in that described device includes:
Entity obtains module, for the query word according to acquisition, obtains at least two entity sets;
Data obtaining module, for the common trait according to each entity sets and the query word, for the generation pair of each entity sets The description information answered;
Recommend display module, is retouched for showing that each entity sets are corresponding in the search results pages to match with the query word Information is stated, and, the description information clicked according to user shows that the clicked corresponding entity sets of description information are included Each recommended entity;
Wherein, the entity obtains module, is specifically used for:
According to the searching entities for including in the query word, it is highest that temperature is obtained in entity relevant to described search entity At least two entities;And at least two entity is clustered, to obtain at least two entity sets;Alternatively,
According to the searching entities and search need information for including in the query word, obtain and described search entity and described search At least two candidate collections that demand information matches, according to the degree of correlation of each candidate collection and described search demand information, from At least two entity sets are obtained at least two candidate collections.
9. device according to claim 8, which is characterized in that
Described device further include: image generating module, for obtaining the Target Photo to match with each entity sets;
The recommendation display module, is specifically used for: in the search results pages to match with the query word, carrying out each entity set Close the display corresponding with the Target Photo of corresponding description information.
10. device according to claim 9, which is characterized in that the image generating module is specifically used for:
For each entity sets, the Target Photo to match with the entity sets is obtained;Alternatively,
For each entity sets, obtain at least two candidate pictures to match with the entity sets, to it is described extremely Few two candidate pictures carry out picture mosaic, to generate the Target Photo to match with the entity sets.
11. device according to claim 8, which is characterized in that the recommendation display module is specifically used for:
Each recommended entity that the corresponding entity sets of the description information clicked using user are included carries out page rendering, with exhibition It now include the new page of each recommended entity.
12. device according to claim 11, which is characterized in that show that each recommendation is real using the layout of a page without columns in the new page Body;The recommendation display module, is also used to:
The recommended entity clicked in each recommended entity shown in the layout of a page without columns according to user, described in the new page The search result that display matches with the recommended entity clicked below the layout of a page without columns.
13. device according to claim 12, which is characterized in that the recommendation display module is also used to: in the layout of a page without columns The middle navigation information shown between clicked recommended entity and the query word;
Wherein, the navigation information include the query word, the corresponding description information of entity sets where the recommended entity with And the recommended entity clicked.
14. device according to claim 8, which is characterized in that the recommendation display module is specifically used for:
The description information clicked according to user, under the description information described in the search results pages to match with the query word Side's display combobox, and, included using the corresponding entity sets of description information that the combobox shows that user is clicked Each recommended entity.
CN201510828555.XA 2015-11-25 2015-11-25 A kind of entity recommended method and device CN105512178B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510828555.XA CN105512178B (en) 2015-11-25 2015-11-25 A kind of entity recommended method and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510828555.XA CN105512178B (en) 2015-11-25 2015-11-25 A kind of entity recommended method and device
PCT/CN2016/089649 WO2017088497A1 (en) 2015-11-25 2016-07-11 Entity recommendation method, device, apparatus, and computer storage medium

Publications (2)

Publication Number Publication Date
CN105512178A CN105512178A (en) 2016-04-20
CN105512178B true CN105512178B (en) 2019-02-05

Family

ID=55720160

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510828555.XA CN105512178B (en) 2015-11-25 2015-11-25 A kind of entity recommended method and device

Country Status (2)

Country Link
CN (1) CN105512178B (en)
WO (1) WO2017088497A1 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105512178B (en) * 2015-11-25 2019-02-05 百度在线网络技术(北京)有限公司 A kind of entity recommended method and device

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102279869A (en) * 2010-06-09 2011-12-14 微软公司 The relationship between the entities navigation
CN104281699A (en) * 2014-10-15 2015-01-14 百度在线网络技术(北京)有限公司 Search recommending method and device
CN104778255A (en) * 2015-04-20 2015-07-15 百度在线网络技术(北京)有限公司 Method and device for recommending searching results
CN104969212A (en) * 2013-01-30 2015-10-07 奎克西公司 Performing application search based on entities

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6834308B1 (en) * 2000-02-17 2004-12-21 Audible Magic Corporation Method and apparatus for identifying media content presented on a media playing device
CN101364239B (en) * 2008-10-13 2011-06-29 中国科学院计算技术研究所 Method for auto constructing classified catalogue and relevant system
CN102609433B (en) * 2011-12-16 2013-11-20 北京大学 Method and system for recommending query based on user log
US20150234813A1 (en) * 2013-11-04 2015-08-20 Michael R. Knapp Systems and Methods for Categorizing and Accessing Information Databases and for Displaying Query Results
CN105512178B (en) * 2015-11-25 2019-02-05 百度在线网络技术(北京)有限公司 A kind of entity recommended method and device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102279869A (en) * 2010-06-09 2011-12-14 微软公司 The relationship between the entities navigation
CN104969212A (en) * 2013-01-30 2015-10-07 奎克西公司 Performing application search based on entities
CN104281699A (en) * 2014-10-15 2015-01-14 百度在线网络技术(北京)有限公司 Search recommending method and device
CN104778255A (en) * 2015-04-20 2015-07-15 百度在线网络技术(北京)有限公司 Method and device for recommending searching results

Also Published As

Publication number Publication date
WO2017088497A1 (en) 2017-06-01
CN105512178A (en) 2016-04-20

Similar Documents

Publication Publication Date Title
US9223873B2 (en) Method and system for incrementally selecting and providing relevant search engines in response to a user query
US7739221B2 (en) Visual and multi-dimensional search
US7917514B2 (en) Visual and multi-dimensional search
US7707220B2 (en) Methods and apparatus for interactive searching techniques
US9495461B2 (en) Search assistant system and method
KR101061529B1 (en) Display of the reduced and extended data items
JP5493267B2 (en) Product search device and product search method
JP5632571B2 (en) User interface for providing the automatic generation and ergonomic representation of keywords search criteria
JP5827121B2 (en) The method for searching a plurality of stored digital images, computer programs and apparatus
JP6242967B2 (en) Modification of the structure of search queries on the online social network
WO2012176317A1 (en) Image recognition system-equipped interest graph collection system using relationship search
JP5118283B2 (en) Search user interface having the features of the visual and improvement metaphor to based access ease and ease of use
US20080059897A1 (en) Method and system of social networking through a cloud
US20100146012A1 (en) Previewing search results for suggested refinement terms and vertical searches
US20040215608A1 (en) Search engine supplemented with URL's that provide access to the search results from predefined search queries
CN102541590B (en) A software recommendation method and system recommendation
US20120124071A1 (en) Extensible search term suggestion engine
US20100153881A1 (en) Process and apparatus for selecting an item from a database
US20150339391A1 (en) Method for searching and device thereof
JP6023354B2 (en) Natural language rendering of a structured search query
WO2013019882A1 (en) Multipoint pinch gesture control of search results
KR20130118897A (en) Smartphone-based methods and systems
JP2007034663A (en) Information processor, information processing method, and program
WO2013173626A2 (en) Using content
CN103946838B (en) Interactive multi-mode image search

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
GR01 Patent grant