CN104142999B - Search result methods of exhibiting and device - Google Patents
Search result methods of exhibiting and device Download PDFInfo
- Publication number
- CN104142999B CN104142999B CN201410377405.7A CN201410377405A CN104142999B CN 104142999 B CN104142999 B CN 104142999B CN 201410377405 A CN201410377405 A CN 201410377405A CN 104142999 B CN104142999 B CN 104142999B
- Authority
- CN
- China
- Prior art keywords
- user
- search
- search term
- search result
- classification
- 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.)
- Active
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
- G06F16/951—Indexing; Web crawling techniques
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 present invention provides a kind of search result methods of exhibiting and device.Search result methods of exhibiting comprises determining that target user's classification belonging to the first user that currently used first search term scans for;It is directed to the historical search data of the first search term according to user each under target user's classification, determines the click weight of each object in the first search result for currently corresponding to the first search term;According to the click weight of object each in the first search result, each object in the first search result is ranked up;Each object in the first search result after sequence is presented to the first user.Technical solution of the present invention can be improved user and obtain the efficiency for meeting the object of demand from search result, improve search efficiency.
Description
[technical field]
The present invention relates to Internet technical field more particularly to a kind of search result methods of exhibiting and device.
[background technique]
With the development of internet technology, user can carry out information inquiry by search engine.Specifically, user inputs
Search term (query), search engine scans for obtaining search result based on the search term that user inputs, then by search result
It is supplied to user.
Existing search result usually shows according to the form of paging or waterfall stream, and one page or a screen show search result
In partial objects, user needs page turning backward or turns over screen and browse, until meeting the object of user demand until obtaining.This
The displaying of kind search result is arbitrary, and user can not often obtain the object for meeting demand from search result in time, search
Efficiency is lower.
[summary of the invention]
Many aspects of the invention provide a kind of search result methods of exhibiting and device, to improve user from search result
It is middle to obtain the efficiency for meeting the object of demand, improve search efficiency.
An aspect of of the present present invention provides a kind of search result methods of exhibiting, comprising:
Determine target user's classification belonging to the first user that currently used first search term scans for;
It is directed to the historical search data of first search term according to each user under target user's classification, determines current
Click weight corresponding to each object in the first search result of first search term;
According to the click weight of each object in first search result, object each in first search result is carried out
Sequence;
Each object in first search result after sequence is presented to first user.
Another aspect of the present invention provides a kind of search result displaying device, comprising:
Class of subscriber determining module, for determining mesh belonging to the first user that currently used first search term scans for
Mark class of subscriber;
Weight determination module is clicked, for being directed to first search term according to each user under target user's classification
Historical search data determines the click weight of each object in the first search result for currently corresponding to first search term;
Sorting module ties first search for the click weight according to each object in first search result
Each object is ranked up in fruit;
Display module, for each object in first search result after sequence to be presented to first user.
It can be seen from the above, the present invention is by classifying to the user for using search term to scan for, to currently used
The first user that one search term scans for determines target user's classification belonging to the first user, uses target user's classification
Under each user be directed to the historical search data of the first search term, determine currently correspond to it is each right in the search result of the first search term
The click weight of elephant is ranked up each object according to the click weight of each object, and each object after sequence is presented to first
User.The present invention classifies user, is ranked up for every class user to object each in search result, so that each in search result
Object can click tendency according to such user and successively show, and such user can obtain from search result in time to meet the requirements
Object, and then be conducive to improve search efficiency.
[Detailed description of the invention]
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is some realities of the invention
Example is applied, it for those of ordinary skill in the art, without any creative labor, can also be attached according to these
Figure obtains other attached drawings.
Fig. 1 is the flow diagram for the search result methods of exhibiting that one embodiment of the invention provides;
Fig. 2 is a kind of flow diagram of the embodiment for the step 101 that one embodiment of the invention provides;
Fig. 3 is a kind of flow diagram of the embodiment for the step 102 that one embodiment of the invention provides;
Fig. 4 be another embodiment of the present invention provides step 102 a kind of embodiment flow diagram;
Fig. 5 be another embodiment of the present invention provides search result methods of exhibiting flow diagram;
Fig. 6 is the structural schematic diagram that the search result that one embodiment of the invention provides shows device;
Fig. 7 be another embodiment of the present invention provides search result show device structural schematic diagram.
[specific embodiment]
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
Fig. 1 is the flow diagram for the search result methods of exhibiting that one embodiment of the invention provides.As shown in Figure 1, the party
Method includes:
101, target user's classification belonging to the first user that currently used first search term scans for is determined.
The executing subject of the present embodiment can be search result and show device, such as can be various search engines, or
It is connection but is used to show the display equipment of the search result of search engine output independently of search engine, but not limited to this.
In the present embodiment, it is contemplated that even if different user is scanned for using same search word, click in search result
The tendency of object would also vary from, and therefore, classify to the user for using search term to scan for, and it is identical to will click on tendency
User be divided into one kind.Based on this, search result shows that device can use each user under same class of subscriber and search for same
The historical search data of rope word adjusts some current user for the sequence of each object in the search result of the search term
It is whole, so as to which the object priority that the user is more likely to click is presented to the user, the user is improved from search result
The efficiency for meeting the object of demand is obtained, and then improves search efficiency.
In embodiments of the present invention, the search term pair of the search term used and user's click can be searched for by user
The binary combination for the object in search result answered is inclined to characterize the click of user.Wherein, the object in search result is main
Refer to the corresponding URL of object in search result.Two between object that the search term used in view of user and user click
Tuple credit union has very much, and data volume can be very big, is unfavorable for user's classification.Based on this, search term that the present embodiment uses user
It is carried out with the object that user clicks extensive.The embodiment of the present invention to extensive mode without limitation, the search that a kind of couple of user uses
It is that the search term, the object that a kind of couple of user clicks are replaced using search word class belonging to search term that word, which carries out extensive mode,
Carrying out extensive mode is to replace the object using the corresponding website of the object (host).Wherein, search word class, which has altogether, includes
155 types such as film, novel, animation, news, over time, search word class can also increase.
The case where being scanned for for any one user under any one class of subscriber using any one search term,
The method flow that search result shows that device is shown search result is all the same.For ease of description, the embodiment of the present invention with
It is illustrated for the scene that first user is scanned for using the first search term.
Specifically, search result shows that device determines the first user that currently used first search term scans for, later
Determine class of subscriber belonging to the first user that currently used first search term scans for.For ease of description, by the first user
Affiliated class of subscriber is known as target user's classification, which is actually one of class of subscriber.It illustrates
Bright, the first search term can be " iphone ", " world cup " etc..
In an optional embodiment, it is contemplated that influence user and click the parameter of tendency to might have very much, such as can be with
It is that the intrinsic attribute of user itself or user's use are inclined in the click that user shows on search result
Terminal device information and any combination of above-mentioned various parameters etc..Wherein, user shows on search result
Clicking tendency can be determined by user for the historical search data of predetermined search word classification.
Based on above-mentioned, search result show device can according to the first user for the first search term belonging to search for word class
Historical search data, the first user attribute information and/or the information of terminal device that uses of the first user, determine the first use
Target user's classification belonging to family.It is directed to for example, target user's classification belonging to the first user can be expressed as the first user
What the historical search data of search word class, the attribute information of the first user belonging to the first search term and/or the first user used
One function of the information of terminal device.
It is worth noting that above-mentioned first user is for the historical search data of search word class belonging to the first search term
Generated search is scanned for using the various search terms under search word class belonging to the first search term before referring to the first user
Data.Here historical search data may include search use search term, searched based on search term content, search day
The information such as the object that phase and user clicked or visited.In addition, the attribute information of above-mentioned first user may include the first use
The information such as gender, age, interest and the state at family.The information for the terminal device that first user uses can be the first user and use
The information such as type, brand, specifications parameter and the performance indicator of terminal device.
In a preferred embodiment, belonging to search result displaying device can be directed to the first search term according to the first user
The historical search data for searching for word class, determines target user's classification belonging to the first user.
For example, search result show device according to the first user going through for search word class belonging to the first search term
History searches for data, determines a kind of embodiment of target user's classification belonging to the first user as shown in Fig. 2, the embodiment packet
It includes:
1011, the historical search data that word class is searched for belonging to according to the first user for the first search term, determines first
User is directed to the object clicking rate of the first search term.
First user refers to that the first user searches using belonging to the first search term for the object clicking rate of the first search term
In the behavior that each search term under rope word class scans for, the first user clicks the probability of the object in search result.
Specifically, search result shows that device can go through the first user for search word class belonging to the first search term
History search data are counted, and the number that the first user clicks the object in the search result of the historical search data is obtained, with
And in the search result of the historical search data object total number;The search of the historical search data is clicked using the first user
As a result the number of the object in, divided by the total number of object in the search result of the historical search data, using division result as
First user is directed to the object clicking rate of the first search term.It is worth noting that historical search data here includes search knot
Fruit.
1012, by the first user, for the object clicking rate of the first search term, to be expressed as condition related with class of subscriber general
Rate model.
1013, above-mentioned condition probabilistic model is solved, obtains conditional probability of first user under each class of subscriber.
Specifically EM algorithm (Expectation Maximization can be carried out to above-mentioned condition probabilistic model
Algorithm, EM) it solves, obtain conditional probability of first user under each class of subscriber.
1014, the conditional probability according to the first user under each class of subscriber determines target user belonging to the first user
Classification.
For example, search result shows that device can choose the corresponding class of subscriber of maximal condition probability as the first user institute
Target user's classification of category.
In another example being greater than predetermined probabilities in maximal condition probability of first user in the conditional probability under each class of subscriber
When thresholding, determine that the corresponding class of subscriber of the maximal condition probability is target user's classification belonging to the first user.If maximum
Conditional probability is not more than predetermined probabilities thresholding, then can give up first user.Wherein, the historical search data of some users
Relatively fewer, conditional probability of the user under each class of subscriber based on determined by the historical search data of the user compares
It is low, possibly can not class of subscriber belonging to the accurate characterization user, the present embodiment, can be by this part by setting probability threshold
The corresponding lower user of conditional probability can directly show the search knot of this certain customers to filtering out according to existing way
Fruit does not need to be ranked up search result, is conducive to mitigate processing load.
It is worth noting that can be determined using target user's class method for distinguishing belonging to above-mentioned the first user of determination
Belong to the other users of target user's classification.
102, it is directed to the historical search data of the first search term according to each user under above-mentioned target user's classification, determines current
Click weight corresponding to each object in the first search result of the first search term.
Illustrate first, each user refers to target user for the historical search data of the first search term under target user's classification
Each user scans for the search data generated using the first search term before current under classification.Here historical search data
It also may include the search term that search uses, the content, search date and the user's click or visit that are searched based on search term
The information such as the object crossed.
Wherein, the click weight of each object represents the first user to a certain extent and clicks the object in the first search result
Tendency degree.The click weight of usually one object is bigger, and first user's point more tends to click the object.
In an optional embodiment, a kind of optional embodiment of above-mentioned steps 102 is as shown in figure 3, the embodiment
Include:
1021, each user under target user's classification is counted for the historical search data of the first search term, is determined
Each user is for the corresponding meeting of object each in the search result of the historical search data of the first search term under target user's classification
Talk about number, the weight of secondary, total number of clicks and each session is clicked in tail portion;
1022, according to each user under above-mentioned target user's classification for the search knot of the historical search data of the first search term
The corresponding session number of each object, tail portion number of clicks, total number of clicks and the weight of each session, calculate first in fruit
The click weight of corresponding objects in search result.
It is worth noting that session here refers to the complete search process for completing user demand, which can
It can include one or many search behaviors.The corresponding session number of each object refers to that this is related to of the session of the object
Number.The weight of each session can be obtained ahead of time, the present embodiment to specific acquisition pattern without limitation.The corresponding tail of each object
Portion's number of clicks refers to that the object is located at the bottom of search result, and the number clicked by user;The corresponding total point of each object
It hits number and refers to the total degree that the object is clicked.
Further, the another embodiment of step 102 is as shown in figure 4, after 1022 further include:
1023, it is directed to the historical search data of the first search term according to users all within the current preset time, determines
The interim amendment weight of each object in first search result.
1024, using the interim amendment weight of each object in the first search result, respectively to each right in the first search result
The click weight of elephant is modified.
103, according to the click weight of object each in the first search result, each object in the first search result is arranged
Sequence.
104, each object in the first search result after sequence is presented to the first user.
In obtaining the first search result after the click weight of each object, search result shows that device can be according to first
The click weight of each object in search result is ranked up (referred to as point is adjusted) to each object, can be incited somebody to action by the sorting operation
User is inclined to the object clicked and is discharged to front, preferentially to show user.For example, search result shows that device can be according to point
The descending sequence of weight is hit, each object is ranked up.
After sequence, search result shows that each object after sequence can be showed the first user, the first user by device
Information needed for checking each object acquisition can be clicked.Since after sequence, user can be tended to the object priority clicked
Show user, it is possible to improve user and obtain the efficiency for meeting the object of demand from search result, and then be conducive to mention
High search efficiency.
Fig. 5 be another embodiment of the present invention provides search result methods of exhibiting flow diagram.As shown in figure 5, should
Method includes:
501, target user's classification belonging to the first user that currently used first search term scans for is determined.
502, it is directed to the historical search data of the first search term according to each user under above-mentioned target user's classification, obtains target
Distribution letter of each user for the object being clicked in the search result of the historical search data of the first search term under class of subscriber
Cease entropy, the click accounting of front three object and the click accounting of types of objects.
503, according to each user under above-mentioned target user's classification for the search knot of the historical search data of the first search term
In the click accounting of the distributed intelligence entropy for the object being clicked in fruit, the click accounting of front three object and types of objects extremely
It is one few, determine that the first search term is the search term for needing to be ranked up search result.
504, it is directed to the historical search data of the first search term according to each user under above-mentioned target user's classification, determines current
Click weight corresponding to each object in the first search result of the first search term.
505, according to the click weight of object each in the first search result, each object in the first search result is arranged
Sequence.
506, each object in the first search result after sequence is presented to the first user.
About step 501,504,505 and 506, reference can be made in embodiment illustrated in fig. 1 step 101-104 description, herein
It repeats no more.
In the present embodiment, it is contemplated that some search terms show as requirement report one, i.e., all users are directed to this kind of search
The object that the search result of word is clicked is all same or similar, such as inquires the search term of weather, therefore be not necessarily to this
The search result of class search term does personalized differentiation (not needing to sort).In order to mitigate the processing that search result shows device
It bears, economize on resources, search result shows that device can classify to search term, is divided into and needs to be ranked up its search result
Search term and do not need the search term being ranked up to its search result.
Based on above-mentioned, the present embodiment each object in corresponding first search result of current first search term is ranked up
Before, it is first determined the first search term is the search term for needing to be ranked up its search result.
Specifically, search result shows that device can be to each user under above-mentioned target user's classification for the first search term
Historical search data is counted, and each user is directed to searching for the historical search data of the first search term under acquisition target user's classification
The click accounting of the distributed intelligence entropy for the object being clicked in hitch fruit, the click accounting of front three object and types of objects.
Wherein, historical search data here includes search result.
The click accounting of above-mentioned front three object refers to that each user is for the history of the first search term under target user's classification
It searches in the search result of data and occupies the percentage for clicking total degree in family positioned at the number that the object of front three is clicked by user;
The click accounting of above-mentioned types of objects refers to each user under target user's classification for the historical search data of the first search term
In search result it is different classes of under the number clicked by user of object occupy the percentage that total degree is clicked at family.
Later, search result shows device according to the distributed intelligence entropy of the object of acquisition being clicked, front three object
At least one of accounting and the click accounting of types of objects are clicked, determines that the first search term is to need to carry out search result
The search term of sequence.
Specifically, search result show device can by the distributed intelligence entropy for the object of above-mentioned acquisition being clicked, first three
Click accounting and at least one of the click accounting of types of objects of position object, as the parameter of dichotomy, and by first
Object of the search term as dichotomy, classifies to the first search term, determines that the first search term is to belong to need to search for it
As a result the search term being ranked up still falls within the search term for not needing to be ranked up to its search result.The present embodiment is with
One search term is judged as the search term for needing to be ranked up its search result.
It further, can if the first search term is judged as the search term for not needing to be ranked up its search result
The corresponding search result of the first search term is directly showed user.
In the present embodiment, classify to search term, it only need to be to the search term for needing to be ranked up its search result
It is ranked up processing, is conducive to mitigate the processing load that search result shows device, economize on resources.
It should be noted that for the various method embodiments described above, for simple description, therefore, it is stated as a series of
Combination of actions, but those skilled in the art should understand that, the present invention is not limited by the sequence of acts described because
According to the present invention, some steps may be performed in other sequences or simultaneously.Secondly, those skilled in the art should also know
It knows, the embodiments described in the specification are all preferred embodiments, and related actions and modules is not necessarily of the invention
It is necessary.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, there is no the portion being described in detail in some embodiment
Point, reference can be made to the related descriptions of other embodiments.
Fig. 6 is the structural schematic diagram that the search result that one embodiment of the invention provides shows device.As shown in fig. 6, the dress
Set includes: class of subscriber determining module 61, click weight determination module 62, sorting module 63 and display module 64.
Class of subscriber determining module 61, for determining belonging to the first user that currently used first search term scans for
Target user's classification.
Weight determination module 62 is clicked, is connect with class of subscriber determining module 61, for according to class of subscriber determining module
Each user is directed to the historical search data of the first search term under the 61 target user's classifications determined, determines that currently corresponding to first searches
The click weight of each object in first search result of rope word.
Sorting module 63 is connect with weight determination module 62 is clicked, for what is determined according to click weight determination module 62
The click weight of each object in first search result, is ranked up each object in the first search result.
Display module 64 is connect with sorting module 63, for each in the first search result after sorting module 63 sorts
Object is presented to the first user.
In an optional embodiment, class of subscriber determining module 61, which is particularly used in, is searched according to the first user for first
The terminal dress that the historical search data of search word class, the attribute information of the first user belonging to rope word and/or the first user use
The information set determines target user's classification belonging to the first user.
Further, class of subscriber determining module 61 is particularly used according to the first user for search belonging to the first search term
The historical search data of word class determines target user's classification belonging to the first user.Then as shown in fig. 7, class of subscriber determines
A kind of realization structure of module 61 includes: clicking rate determination unit 611, probability obtaining unit 612 and class of subscriber determination unit
613。
Clicking rate determination unit 611, for according to the first user for the history for searching for word class belonging to the first search term
Data are searched for, determine that the first user is directed to the object clicking rate of the first search term.
Probability obtaining unit 612 is connect with clicking rate determination unit 611, for will click on the determination of rate determination unit 611
First user is expressed as conditional probability model related with class of subscriber for the object clicking rate of the first search term, and to condition
Probabilistic model is solved, and conditional probability of first user under each class of subscriber is obtained.
Class of subscriber determination unit 613 is connect with probability obtaining unit 612, for being obtained according to probability obtaining unit 612
Conditional probability of first user under each class of subscriber, determine target user's classification belonging to the first user.
In an optional embodiment, as shown in fig. 7, a kind of realization structure for clicking weight determination module 62 includes: ginseng
Number determination unit 621 and computing unit 622.
Parameter determination unit 621, for being searched to each user under target user's classification for the history of first search term
Rope data are counted, and determine that each user is for the search result of the historical search data of the first search term under target user's classification
In the corresponding session number of each object, tail portion number of clicks, total number of clicks and the weight of each session.
Computing unit 622 is connect with parameter determination unit 621, and the target for being determined according to parameter determination unit 621 is used
Under the classification of family each user for the corresponding session number of each object in the search result of the historical search data of the first search term,
The weight of tail portion number of clicks, total number of clicks and each session calculates the click power of corresponding objects in the first search result
Weight.Computing unit 622 is also connect with sorting module 63.
In an optional embodiment, as shown in fig. 7, the device further include: amendment weight determining module 65 and amendment mould
Block 66.
Weight determining module 65 is corrected, for being directed to the first search term according to users all within the current preset time
Historical search data determines the interim amendment weight of each object in the first search result.
Correction module 66 is connect with amendment weight determining module 65 and computing unit 622, for being determined using amendment weight
The interim amendment weight of each object in the first search result that module 65 determines, respectively to computing unit 622 calculated first
The click weight of each object is modified in search result.Correction module 66 is also connect with sorting module 63, is used for sequence mould
Block 63 provides each object revised click weight.
In an optional embodiment, as shown in fig. 7, the device further include: gain of parameter module 67 and search term determine
Module 68.
Gain of parameter module 67, for being directed to the historical search number of the first search term according to user each under target user's classification
According to each user is for pair being clicked in the search result of the historical search data of the first search term under acquisition target user's classification
The click accounting of the distributed intelligence entropy of elephant, the click accounting of front three object and types of objects.
Search term determining module 68 is connect with gain of parameter module 67, the mesh for being obtained according to gain of parameter module 67
Distribution of each user for the object being clicked in the search result of the historical search data of the first search term under mark class of subscriber
At least one of comentropy, the click accounting of front three object and click accounting of types of objects, determine the first search term
It is the search term for needing to be ranked up search result.Search term determining module 68 is also connect with click weight determination module 62,
For providing what whether identified first search term needed to be ranked up search result to click weight determination module 62
The result of search term.
Search result provided in this embodiment shows device, by dividing the user for using search term to scan for
Class determines target user's classification belonging to the first user to the first user that currently used first search term scans for, and uses
Each user is directed to the historical search data of the first search term under target user's classification, and determination currently corresponds to the first search term
The click weight of each object in search result, is ranked up each object according to the click weight of each object, will be each after sequence
Object is presented to the first user.Search result provided in this embodiment shows that device classifies user, for every class user to searching
Each object is ranked up in hitch fruit, and each object in search result is successively showed according to such user click tendency,
User can obtain satisfactory object from search result in time in this way, and then be conducive to improve search efficiency.
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, such as multiple units or components
It 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, the indirect coupling of device or unit
It closes or communicates to connect, 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.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features;
And these are modified or replaceed, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (12)
1. a kind of search result methods of exhibiting characterized by comprising
Determine target user's classification belonging to the first user that currently used first search term scans for;
It is directed to the historical search data of first search term according to each user under target user's classification, determines current corresponding
The click weight of each object in the first search result of first search term;
According to the click weight of each object in first search result, each object in first search result is arranged
Sequence;
Each object in first search result after sequence is presented to first user;
Target user's classification belonging to the first user that currently used first search term of determination scans for, comprising:
It is used according to first user for the historical search data of search word class, described first belonging to first search term
The information for the terminal installation that the attribute information at family and/or first user use, determines target belonging to first user
Class of subscriber.
2. the method according to claim 1, wherein the historical search data according to first user,
Determine target user's classification belonging to first user, comprising:
According to first user for the historical search data for searching for word class belonging to first search term, described the is determined
One user is directed to the object clicking rate of first search term;
It is general that the object clicking rate that first user is directed to first search term is expressed as condition related with class of subscriber
Rate model;
The conditional probability model is solved, conditional probability of first user under each class of subscriber is obtained;
According to conditional probability of first user under each class of subscriber, target user's class belonging to first user is determined
Not.
3. according to the method described in claim 2, it is characterized in that, it is described according to first user under each class of subscriber
Conditional probability determines target user's classification belonging to first user, comprising:
When maximal condition probability of first user in the conditional probability under each class of subscriber is greater than predetermined probabilities thresholding,
Determine that the corresponding class of subscriber of the maximal condition probability is target user's classification belonging to first user.
4. method according to claim 1-3, which is characterized in that described according to each under target user's classification
User is directed to the historical search data of first search term, determines the first search knot for currently corresponding to first search term
The click weight of each object in fruit, comprising:
Each user under target user's classification is counted for the historical search data of first search term, determines institute
State each object pair in the search result for the historical search data that each user is directed to first search term under target user's classification
The weight of session number, tail portion number of clicks, total number of clicks and each session answered;
According to each user under target user's classification in the search result of the historical search data of first search term
The corresponding session number of each object, tail portion number of clicks, total number of clicks and the weight of each session, calculate described first
The click weight of corresponding objects in search result.
5. according to the method described in claim 4, it is characterized by further comprising:
It is directed to the historical search data of first search term according to users all within the current preset time, determines described the
The interim amendment weight of each object in one search result;
Using the interim amendment weight of each object in first search result, respectively to each object in first search result
Click weight be modified.
6. method according to claim 1-3, which is characterized in that described according to each under target user's classification
User is directed to the historical search data of first search term, determines the first search knot for currently corresponding to first search term
In fruit before the click weight of each object, comprising:
It is directed to the historical search data of first search term according to each user under target user's classification, obtains the target
Point of each user for the object being clicked in the search result of the historical search data of first search term under class of subscriber
The click accounting of cloth comentropy, the click accounting of front three object and types of objects;
According to each user under target user's classification in the search result of the historical search data of first search term
At least one in the click accounting of the distributed intelligence entropy for the object being clicked, the click accounting of front three object and types of objects
It is a, determine that first search term is the search term for needing to be ranked up search result.
7. a kind of search result shows device characterized by comprising
Class of subscriber determining module, for determining that target belonging to the first user that currently used first search term scans for is used
Family classification;
Weight determination module is clicked, for being directed to the history of first search term according to each user under target user's classification
Data are searched for, determine the click weight of each object in the first search result for currently corresponding to first search term;
Sorting module, for the click weight according to each object in first search result, in first search result
Each object is ranked up;
Display module, for each object in first search result after sequence to be presented to first user;The use
Family category determination module is specifically used for according to first user for the history for searching for word class belonging to first search term
The information of terminal installation that search data, the attribute information of first user and/or first user use, determine described in
Target user's classification belonging to first user.
8. device according to claim 7, which is characterized in that the class of subscriber determining module includes:
Clicking rate determination unit, for according to first user for the history for searching for word class belonging to first search term
Data are searched for, determine that first user is directed to the object clicking rate of first search term;
Probability obtaining unit, the object clicking rate for first user to be directed to first search term is expressed as and user
The related conditional probability model of classification, and the conditional probability model is solved, first user is obtained in each user
Conditional probability under classification;
Class of subscriber determination unit determines described for the conditional probability according to first user under each class of subscriber
Target user's classification belonging to one user.
9. device according to claim 8, which is characterized in that the class of subscriber determination unit is specifically used for described the
When maximal condition probability of one user in the conditional probability under each class of subscriber is greater than predetermined probabilities thresholding, the maximum is determined
The corresponding class of subscriber of conditional probability is target user's classification belonging to first user.
10. according to the described in any item devices of claim 7-9, which is characterized in that the click weight determination module includes:
Parameter determination unit, for being directed to the historical search number of first search term to each user under target user's classification
According to being counted, determine that each user is for the search of the historical search data of first search term under target user's classification
As a result the corresponding session number of each object, tail portion number of clicks, total number of clicks and the weight of each session in;
Computing unit, for being directed to the historical search data of first search term according to each user under target user's classification
Search result in the corresponding session number of each object, tail portion number of clicks, total number of clicks and the weight of each session,
Calculate the click weight of corresponding objects in first search result.
11. device according to claim 10, which is characterized in that further include:
Weight determining module is corrected, for the going through for first search term according to users all within the current preset time
History searches for data, determines the interim amendment weight of each object in first search result;
Correction module is searched to described first respectively for the interim amendment weight using each object in first search result
The click weight of each object is modified in hitch fruit.
12. according to the described in any item devices of claim 7-9, which is characterized in that further include:
Gain of parameter module, for being directed to the historical search of first search term according to each user under target user's classification
Data obtain each user under target user's classification in the search result of the historical search data of first search term
The click accounting of the distributed intelligence entropy for the object being clicked, the click accounting of front three object and types of objects;
Search term determining module, for being searched according to each user under target user's classification for the history of first search term
The distributed intelligence entropy for the object being clicked in the search result of rope data, the click accounting of front three object and types of objects
At least one of accounting is clicked, determines that first search term is the search term for needing to be ranked up search result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410377405.7A CN104142999B (en) | 2014-08-01 | 2014-08-01 | Search result methods of exhibiting and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410377405.7A CN104142999B (en) | 2014-08-01 | 2014-08-01 | Search result methods of exhibiting and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104142999A CN104142999A (en) | 2014-11-12 |
CN104142999B true CN104142999B (en) | 2019-03-29 |
Family
ID=51852173
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410377405.7A Active CN104142999B (en) | 2014-08-01 | 2014-08-01 | Search result methods of exhibiting and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104142999B (en) |
Families Citing this family (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104778233A (en) * | 2015-03-31 | 2015-07-15 | 北京奇虎科技有限公司 | Searching method and device based on click rate |
CN106156246B (en) * | 2015-04-28 | 2019-10-25 | 腾讯科技(深圳)有限公司 | A kind of querying method of public platform, device and system |
CN105117383A (en) * | 2015-08-14 | 2015-12-02 | 百度在线网络技术(北京)有限公司 | Search result providing method and apparatus |
CN107423298B (en) * | 2016-05-24 | 2021-02-19 | 北京百度网讯科技有限公司 | Searching method and device |
CN106021602B (en) * | 2016-06-15 | 2018-07-06 | 腾讯科技(深圳)有限公司 | A kind of method and device of search results ranking |
CN106126698B (en) * | 2016-06-29 | 2020-01-03 | 武汉斗鱼网络科技有限公司 | Retrieval pushing method and system based on Lucence |
CN107169131A (en) * | 2017-06-08 | 2017-09-15 | 广州优视网络科技有限公司 | A kind of video searching method, device and server |
CN109213942A (en) * | 2017-06-30 | 2019-01-15 | 北京搜狗科技发展有限公司 | A kind of search result methods of exhibiting and device |
CN107562873B (en) * | 2017-08-31 | 2021-02-02 | 北京小度互娱科技有限公司 | Method and device for pushing waterfall flow data |
CN107608792B (en) * | 2017-09-12 | 2020-09-01 | 中国联合网络通信集团有限公司 | Resource scheduling method and device |
CN107977405B (en) * | 2017-11-16 | 2021-01-22 | 北京三快在线科技有限公司 | Data sorting method, data sorting device, electronic equipment and readable storage medium |
CN110020206B (en) * | 2019-04-12 | 2021-10-15 | 北京搜狗科技发展有限公司 | Search result ordering method and device |
CN110020209B (en) * | 2019-04-18 | 2022-03-22 | 北京奇艺世纪科技有限公司 | Method and system for determining correlation between content and search word and method and system for displaying correlation |
CN110209927B (en) * | 2019-04-25 | 2020-12-04 | 北京三快在线科技有限公司 | Personalized recommendation method and device, electronic equipment and readable storage medium |
CN110929088B (en) * | 2019-10-25 | 2023-08-25 | 哈尔滨师范大学 | Music search system |
CN111581452B (en) * | 2020-03-26 | 2023-10-17 | 浙江口碑网络技术有限公司 | Recommendation object data obtaining method and device and electronic equipment |
CN111949788B (en) * | 2020-08-14 | 2024-03-19 | 中国工商银行股份有限公司 | Search method, search device, search system and medium |
CN112434183B (en) * | 2020-11-30 | 2024-05-17 | 北京达佳互联信息技术有限公司 | Method, device, equipment and storage medium for ordering search results |
CN114428902B (en) * | 2021-12-31 | 2023-11-14 | 北京百度网讯科技有限公司 | Information searching method, device, electronic equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101334796A (en) * | 2008-02-29 | 2008-12-31 | 浙江师范大学 | Personalized and synergistic integration network multimedia search and enquiry method |
CN101477554A (en) * | 2009-01-16 | 2009-07-08 | 西安电子科技大学 | User interest based personalized meta search engine and search result processing method |
CN101556603A (en) * | 2009-05-06 | 2009-10-14 | 北京航空航天大学 | Coordinate search method used for reordering search results |
CN103593353A (en) * | 2012-08-15 | 2014-02-19 | 阿里巴巴集团控股有限公司 | Information search method and display information sorting weight value determination method and device |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103886090B (en) * | 2014-03-31 | 2018-01-02 | 北京搜狗科技发展有限公司 | Content recommendation method and device based on user preferences |
-
2014
- 2014-08-01 CN CN201410377405.7A patent/CN104142999B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101334796A (en) * | 2008-02-29 | 2008-12-31 | 浙江师范大学 | Personalized and synergistic integration network multimedia search and enquiry method |
CN101477554A (en) * | 2009-01-16 | 2009-07-08 | 西安电子科技大学 | User interest based personalized meta search engine and search result processing method |
CN101556603A (en) * | 2009-05-06 | 2009-10-14 | 北京航空航天大学 | Coordinate search method used for reordering search results |
CN103593353A (en) * | 2012-08-15 | 2014-02-19 | 阿里巴巴集团控股有限公司 | Information search method and display information sorting weight value determination method and device |
Also Published As
Publication number | Publication date |
---|---|
CN104142999A (en) | 2014-11-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104142999B (en) | Search result methods of exhibiting and device | |
CN105808685B (en) | Promotion information pushing method and device | |
CN105447730B (en) | Target user orientation method and device | |
WO2018041168A1 (en) | Information pushing method, storage medium and server | |
CN105701216B (en) | A kind of information-pushing method and device | |
CN102043833B (en) | Search method and device based on query word | |
US20110145234A1 (en) | Search method and system | |
CN109299344A (en) | The generation method of order models, the sort method of search result, device and equipment | |
US11514063B2 (en) | Method and apparatus of recommending information based on fused relationship network, and device and medium | |
CN107305637B (en) | Data clustering method and device based on K-Means algorithm | |
CN111324789B (en) | Method for calculating network information data heat | |
CN110968767B (en) | Ranking engine training method and device, and business card ranking method and device | |
CN104462293A (en) | Search processing method and method and device for generating search result ranking model | |
CN109242002A (en) | High dimensional data classification method, device and terminal device | |
CN111444395A (en) | Method, system and equipment for obtaining relation expression between entities and advertisement recalling system | |
CN109471978B (en) | Electronic resource recommendation method and device | |
CN107995285B (en) | Information display method, device, server and readable storage medium | |
WO2016155493A1 (en) | Data processing method and apparatus | |
CN112052387B (en) | Content recommendation method, device and computer readable storage medium | |
CN108664515B (en) | A kind of searching method and device, electronic equipment | |
CN102693229A (en) | Analysis method, recommendation method, analysis apparatus and recommendation apparatus for software | |
CN110209928A (en) | A kind of information recommendation method, device and storage medium | |
CN102364467A (en) | Network search method and system | |
CN106021423B (en) | META Search Engine personalization results recommended method based on group division | |
CN104462347A (en) | Keyword classifying method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |