CN102810117B - A kind of for providing the method and apparatus of Search Results - Google Patents
A kind of for providing the method and apparatus of Search Results Download PDFInfo
- Publication number
- CN102810117B CN102810117B CN201210226803.XA CN201210226803A CN102810117B CN 102810117 B CN102810117 B CN 102810117B CN 201210226803 A CN201210226803 A CN 201210226803A CN 102810117 B CN102810117 B CN 102810117B
- Authority
- CN
- China
- Prior art keywords
- search result
- result
- order models
- search
- amount threshold
- 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
Abstract
The object of this invention is to provide a kind of for providing the method and apparatus of Search Results.Computer equipment obtains the initial search result corresponding with the search sequence that user inputs; Utilize the first order models, in described initial search result, filter out optimal search result; Utilize the second order models, in described optimal search result, filter out optimum search result; Described optimum search result is supplied to described user.Compared with prior art, the present invention, by multiple order models, carries out sizing screening to Search Results, achieves and optimum search result is supplied to user, thus take into account precision and the efficiency of Search Results, reach the optimization of search efficiency and effect.Further, utilize the method determination order models of machine learning, and utilize submodel to generate upper strata model by machine learning, thus optimize the setting of order models, ensure that the real-time of system order models, intelligibility and controllability.
Description
Technical field
The present invention relates to computer realm, particularly relating to a kind of for providing the technology of Search Results.
Background technology
Current, for the mode providing employing one minor sort mostly of Search Results, namely according to the inquiry request of user, by the inquiry to background data base, utilize preset order models, the Search Results of respective user inquiry request is supplied to user.This mode also exists certain problem, and namely a minor sort is not easy to reach the optimization in efficiency and effect.From efficiency comes first angle, the inquiry mode that precision is lower can be utilized to carry out, but the pin-point accuracy of Query Result cannot be ensured; From the preferential angle of effect, the inquiry mode that precision is higher can be utilized to carry out, but cannot search efficiency be ensured simultaneously.
Summary of the invention
The object of this invention is to provide a kind of for providing the method and apparatus of Search Results.
According to an aspect of the present invention, provide a kind of by computer implemented for providing the method for Search Results, the method comprises the following steps:
A obtains the initial search result corresponding with the search sequence that user inputs;
B utilizes the first order models, in described initial search result, filter out optimal search result;
C utilizes the second order models, filters out optimum search result in described optimal search result;
Described optimum search result is supplied to described user by d.
According to a further aspect in the invention, additionally provide a kind of for providing the result of Search Results to provide equipment, this equipment comprises:
Result acquisition device, for the initial search result that the search sequence obtained with user inputs is corresponding;
First screening plant, for utilizing the first order models, filters out optimal search result in described initial search result;
Second screening plant, for utilizing the second order models, filters out optimum search result in described optimal search result;
Result generator, for being supplied to described user by described optimum search result.
In accordance with a further aspect of the present invention, additionally providing a kind of search engine, comprising described above for providing the result of Search Results to provide equipment.
In accordance with a further aspect of the present invention, additionally providing a kind of search engine plug-in unit, comprising described above for providing the result of Search Results to provide equipment.
In accordance with a further aspect of the present invention, additionally providing a kind of browser, comprising described above for providing the result of Search Results to provide equipment.
In accordance with a further aspect of the present invention, additionally providing a kind of browser plug-in, comprising described above for providing the result of Search Results to provide equipment.
Compared with prior art, the present invention, by multiple order models, carries out sizing screening to Search Results, achieves and optimum search result is supplied to user, thus take into account precision and the efficiency of Search Results, reach the optimization of search efficiency and effect.Further, utilize the method determination order models of machine learning, and utilize submodel to generate upper strata model by machine learning, thus optimize the setting of order models, ensure that the real-time of system order models, intelligibility and controllability.In addition, the current feature selecting for order models, be put together by all low-level image features of various angle mostly, the problem brought like this is to weaken real-time, intelligibility and controllability, be unfavorable for the location of problem, before being also unfavorable for, algorithm accumulation is multiplexing; Given this, the present invention also uses different features for different order models, and ensureing search efficiency and effect optimized while, the feature selecting of Optimal scheduling model, further improves the optimization of search efficiency and effect.
Accompanying drawing explanation
By reading the detailed description done non-limiting example done with reference to the following drawings, other features, objects and advantages of the present invention will become more obvious:
It is a kind of for providing the result of Search Results to provide equipment schematic diagram that Fig. 1 illustrates according to one aspect of the invention;
It is a kind of for providing the result of Search Results to provide equipment schematic diagram that Fig. 2 illustrates in accordance with a preferred embodiment of the present invention;
Fig. 3 illustrates a kind of method flow diagram for providing Search Results providing equipment to realize by result according to a further aspect of the present invention;
Fig. 4 illustrates a kind of method flow diagram for providing Search Results providing equipment to realize by result in accordance with a preferred embodiment of the present invention.
In accompanying drawing, same or analogous Reference numeral represents same or analogous parts.
Embodiment
Below in conjunction with accompanying drawing, the present invention is described in further detail.
It is a kind of for providing the result of Search Results to provide equipment schematic diagram that Fig. 1 illustrates according to one aspect of the invention; Wherein, this result provides equipment to comprise result acquisition device 11, first screening plant 12, second screening plant 13, result generator 14.Result acquisition device 11 obtains the initial search result corresponding with the search sequence that user inputs; First screening plant 12 utilizes the first order models, in described initial search result, filter out optimal search result; Second screening plant 13 utilizes the second order models, filters out optimum search result in described optimal search result; Described optimum search result is supplied to described user by result generator 14.Wherein, result provides equipment not only can work alone, and can also be integrated in the network equipment, subscriber equipment or the network equipment with subscriber equipment by the mutually integrated equipment formed of network.Wherein, the described network equipment its include but not limited to computing machine, network host, single network server, cloud that multiple webserver collection or multiple server are formed; At this, cloud is formed by based on a large amount of computing machine of cloud computing (CloudComputing) or the webserver, and wherein, cloud computing is the one of Distributed Calculation, the virtual supercomputer be made up of the loosely-coupled computing machine collection of a group.Described subscriber equipment its include but not limited to that any one can to carry out the electronic product of man-machine interaction, such as computing machine, smart mobile phone, PDA, game machine or IPTV etc. with user by keyboard, telepilot, touch pad or voice-operated device.Described network includes but not limited to internet, wide area network, Metropolitan Area Network (MAN), LAN (Local Area Network), VPN, wireless self-organization network (AdHoc network) etc.Those skilled in the art will be understood that other result provides equipment to be equally applicable to the present invention, within also should being included in scope, and are contained in this at this with way of reference.
Wherein, result acquisition device 11 obtains the initial search result corresponding with the search sequence that user inputs.Particularly, result acquisition device 11 such as passes through page technology, as JSP, ASP, the page technology such as PHP, or, the application programming interfaces (API) that the equipment of described search sequence can be provided to provide by invoke user equipment or other, or http, the communication mode of other agreements such as https, carry out alternately with user, obtain the search sequence of user's input, and by such as carrying out participle to the search sequence of user's input, and in Query Database, the mode of searching for is carried out for the search sequence after participle, obtain the initial search result corresponding with the search sequence that user inputs, wherein, user is by such as keyboard, touch-screen, speech input device and result acquisition device 11 carry out alternately, input the search sequence that it wishes inquiry, thus initiate search, or, result acquisition device 11 passes through based on various communication protocol (CommunicationsProtocol), the transportation protocol of compunication is referred in this " communication protocol ", as: TCP/IP, UDP, FTP, ICMP, NetBEUI etc., also comprise other form communications be present in computing machine, such as: the communication inside object based programming between object simultaneously, message transmission protocol in operating system between distinct program or computing machine disparate modules, can provide the equipment of described initial search result with other, as search engine, carries out alternately to obtain the initial search result corresponding with the search sequence that user inputs.Preferably, result acquisition device 11 can also intercept the Search Results of some in the obtained Search Results corresponding with the search sequence that user inputs, using as described initial search result.Such as, user proposes the search sequence of " the niciest Sichuan cuisine " to result acquisition device 11 by page technology, result acquisition device 11 carries out participle to " the niciest Sichuan cuisine ", retrieves respectively in a database to " the niciest " and " Sichuan cuisine ", obtains 1000 initial search results.It will be understood by those skilled in the art that mode and several communication transport protocols of above-mentioned acquisition initial search result are only citing; the mode of other acquisition initial search results that are existing or that may occur from now on or communication transport protocols are as being applicable to the present invention; also within scope should being included in, and this is contained at this with way of reference.
First screening plant 12 utilizes the first order models, in described initial search result, filter out optimal search result.Particularly, the described initial search result that the first screening plant 12 provides for result acquisition device 11, utilizes described first order models, calculates priority or the sequencing information of each initial search result; Again according to these priority or sequencing information, described initial search result is screened, to obtain described optimal search result, as priority or sequencing information are met the initial search result of certain threshold requirement as optimal search result, or by these initial search results by its priority or sequencing information descending sort, and top n initial search result will be come as optimal search result.Wherein, sort algorithm, sequencing feature vector etc. is included but not limited in described first order models.
At this, the first order models calculating priority of initial search result or the mode of sequencing information is utilized to include but not limited to: to utilize the first order models, such as comprise the proper vector of one or more characteristic component and weight thereof, determine the assignment of each characteristic component corresponding to initial search result, thus the proper vector obtained corresponding to this initial search result, namely this proper vector comprises assignment and the weight thereof of these characteristic components, using as the priority of this initial search result or sequencing information; Preferably, the assignment of each characteristic component included by this proper vector and weight thereof can also carry out the assignment that this proper vector is determined in weighting, using as the priority of this initial search result or sequencing information.
At this, the mode that initial search result is undertaken arranging by its priority or sequencing information is included but not limited to: be not general, can suppose that priority or the sequencing information of initial search result comprise the proper vector corresponding with this initial search result, this proper vector comprises assignment and the weight thereof of one or more characteristic component, according to the size of the assignment of each proper vector (as determined by the assignment weighting of its characteristic component), the sequence of corresponding Search Results can be determined; Or, according to weight and the assignment size (as dictionary sequence) thereof of each characteristic component of each proper vector, determine the sequence of corresponding Search Results, such as first sort by the assignment of the highest characteristic component of weight, then for the initial search result that the assignment of the highest characteristic component of its weight is identical, can sort, until complete the sequence of all initial search results by the assignment of the secondary high characteristic component of weight.Such as, user proposes the search sequence of " the niciest Sichuan cuisine " to result acquisition device 11 by page technology, result acquisition device 11 provides 1000 initial search results, first order models is defined as two components carrying out based on weight each 50% to the initial search result of user's search sequence, authoritative analysis and semantic analysis, sort, first screening plant 12 obtains 1000 described initial search results that result acquisition device 11 provides, and utilize method that is authoritative and Semantic Ranking to screen described 1000, obtain front 100 results of semantic and authoritative sequence, as described optimal search result.
Second screening plant 13 utilizes the second order models, filters out optimum search result in described optimal search result.Particularly, in the described optimal search result that the second screening plant 13 provides at the first screening plant 12, described second order models is utilized to carry out further Screening Treatment to the search sequence of described user, to obtain described optimum search result.Those skilled in the art will be understood that, except the difference of the second order models and the first order models, the implementation of the second screening plant 13 or basic simlarity identical with the first screening plant 12, therefore for simplicity, repeat no more, be only contained in this by reference.Such as, user proposes the search sequence of " the niciest Sichuan cuisine " to result acquisition device 11 by page technology, result acquisition device 11 provides 1000 initial search results, first screening plant 12 provides 100 optimal search result, second order models be defined as to the preferred result of user's search sequence carry out based on weight be respectively 25% four components: user requirements analysis, user behavior analysis of statistical results, authoritative analysis and semantic analysis are sorted, second screening plant 13 obtains the first screening plant 12 and provides 100 optimal search result, and utilize the second order models, the integrated ordered of 4 components is carried out to described 100 optimal search result, obtain further ranking results, and using the result of first 10 of sequence as described optimum search result.
Described optimum search result is supplied to described user by result generator 14.Particularly, result generator 14 obtains the optimum search result that the second screening plant 13 filters out, and utilization and user carry out alternately, or the call format of the communication mode of the application programming interfaces provided according to subscriber equipment (API) or other agreements such as http, https, is supplied to described user by described optimum search result.Such as, 10 the described optimum search results obtained by second screening plant 13 present to user as the homepage of user search result, or 10 the described optimum search results obtained by the second screening plant 13 present to user as the homepage of user search result, and residue optimal search result is presented to user successively as required after homepage.At this, those skilled in the art will be understood that the display for initial search result, optimal search result or optimum search result, can carry out successively according to optimum search result, optimal search result, initial search result; Also can show from optimum search result, optimal search result, the optional one of initial search result; Also can be by both or triplicity in optimum search result, optimal search result, initial search result, carry out showing according to user's request.
At this, it will be understood by those skilled in the art that result provides equipment can also comprise three screening device and even more multilevel screening device, thus multi-stage sequencing is carried out to initial search result, multi-stage sequencing model is such as adopted to sort step by step to initial search result, to be supplied to the optimum search result of user to obtain.Preferably, the present invention can also determine the rank of required order models according to different application demands, and according to the multi-stage sequencing model of appropriate level, as secondary order models, three or more level order models, initial search result is sorted step by step, to be supplied to the optimum search result of user to obtain.
Preferably, the first screening plant 12 can also utilize described first order models, determines the priority of described initial search result; According to the first predetermined amount threshold, based on the priority of described initial search result, from described initial search result, filter out described optimal search result, wherein, the quantity of described optimal search result meets described first amount threshold.Particularly, first screening plant 12 utilizes described first order models, determine the priority of described initial search result, proper vector such as corresponding to this initial search result or its assignment, again according to the first predetermined amount threshold, based on the priority of these initial search results, described optimal search result is filtered out from described initial search result, wherein, the quantity of described optimal search result meets described first amount threshold, from these initial search results, such as directly filter out the higher initial search result of the priority of some as described optimal search result, or first by its priority, descending sort is carried out to these initial search results, then using the initial search result of some of standing out as described optimal search result.Such as, suppose that the first amount threshold is 100, user proposes the search sequence of " the niciest Sichuan cuisine " to result acquisition device 11 by page technology, result acquisition device 11 provides 1000 initial search results, first order models is defined as two components carrying out based on weight each 50% to the initial search result of user's search sequence, authoritative analysis and semantic analysis, sort, first screening plant 12 obtains 1000 described initial search results that result acquisition device 11 provides, and utilize method that is authoritative and Semantic Ranking to determine the priority of these 1000 initial search results, and according to the first amount threshold 100, screen in these 1000 initial search results are by the sequence of its priority descending sort, obtain front 100 results of semantic and authoritative sequence, as described optimal search result.
More preferably, result provides equipment also to comprise first threshold determining device (not shown), and wherein, first threshold determining device determines rule according to the first predetermined quantity, determines described first amount threshold; Wherein, described first quantity determines that rule comprises following at least any one: based on the quantity of described initial search result, determines described first amount threshold; Based on predetermined for determining that the amount threshold of described optimum search result is established rules then really, determine described first amount threshold.Particularly, first threshold determining device, according to the quantity of described initial search result or predetermined for determining that the amount threshold of described optimum search result is established rules then really, dynamically determines described first amount threshold.Such as, the first quantity determines that arranging described first amount threshold in rule is directly proportional to the quantity of described initial search result, then described initial search result quantity is more, and described first amount threshold is larger; Or, the amount threshold that first quantity determines to arrange in rule described first amount threshold and described optimum search result is established rules then positive correlation really, as the restriction according to interface display, it is fixing that the amount threshold of optimum search result limits, then the first amount threshold is directly proportional to the amount threshold of described fixing optimum search result, such as, represent quantity according to every page and determine that the first amount threshold is every page of integral multiple representing quantity; Or the determination the form of the rules positive feedback relation of the determination of described first amount threshold and the quantity of described initial search result and the predetermined amount threshold for determining described optimum search result, finally reaches balance, thus determines described first amount threshold.
Preferably, the second screening plant 13 can also utilize described second order models, determines the priority of described optimal search result; According to the second predetermined amount threshold, based on the priority of described optimal search result, filter out described optimum search result from described optimal search result, wherein, the quantity of described optimum search result meets described second amount threshold.Particularly, second screening plant 13 utilizes described second order models, determine the priority of described optimal search result, proper vector such as corresponding to this optimal search result or its assignment, again according to the second predetermined amount threshold, based on the priority of these optimal search result, described optimum search result is filtered out from described optimal search result, wherein, the quantity of described optimum search result meets described second amount threshold, from these optimal search result, such as directly filter out the higher optimal search result of the priority of some as described optimum search result, or first by its priority, descending sort is carried out to these optimal search result, then using the optimal search result of some of standing out as described optimum search result.
More preferably, result provides equipment also to comprise Second Threshold determining device (not shown), and wherein, Second Threshold determining device determines rule according to the second predetermined quantity, determines described second amount threshold; Wherein, described second quantity determines that rule comprises following at least any one: based on the terminal attribute of the subscriber equipment of described user, determines described second amount threshold; Based on the type information of described search sequence, determine described second amount threshold; Based on the quantity of described optimal search result, determine described second amount threshold.Particularly, Second Threshold determining device determines rule according to described the second predetermined quantity, based on the terminal attribute of the subscriber equipment of described user, or based on the type information of described search sequence, or based on the quantity of described optimal search result, determine described second amount threshold.Such as, different according to the terminal attribute of subscriber equipment, the second amount threshold is corresponding difference also, as PC holds display screen relatively large, every page can present 10 results, then the second amount threshold is 10, the display screen of mobile device is relatively little, then the second amount threshold is 6; Or it is different according to the type information of described search sequence, second amount threshold is corresponding difference also, type as search sequence is Bus Schedules, then present the most a small amount of result, second amount threshold is relatively little, type as search sequence is catering information, then present the demand that relatively many results just may meet user; Or be directly proportional to the quantity of described optimal search result, determine described second amount threshold.
It is a kind of for providing the result of Search Results to provide equipment schematic diagram that Fig. 2 illustrates in accordance with a preferred embodiment of the present invention; Wherein, this result provides equipment to comprise result acquisition device 11 ', first screening plant 12 ', the second screening plant 13 ', result generator 14 ', model determining device 15 '.Particularly, model determining device 15 ' according to through mark sequence the first Search Results training data, by the mode of machine learning, determine order models, wherein, described order models comprises following at least any one: described first order models, described second order models; Result acquisition device 11 ' obtains the initial search result corresponding with the search sequence that user inputs; First screening plant 12 ' utilizes the first order models, in described initial search result, filter out optimal search result; Second screening plant 13 ' utilizes the second order models, filters out optimum search result in described optimal search result; Described optimum search result is supplied to described user by result generator 14 '.Wherein, result provides result acquisition device 11 ', first screening plant 12 ' in equipment, the second screening plant 13 ', result generator 14 ' identical with corresponding intrument shown in Fig. 1 or substantially identical respectively, so place repeats no more, and be contained in this by way of reference.
Constant work between above-mentioned each device, at this, it will be understood by those skilled in the art that " continuing " refer to above-mentioned each device respectively according to setting or the mode of operation of in real time adjustment require to carry out the determination of order models, the acquisition of initial search result, the screening of optimal search result, the screening of optimum search result and providing of optimum search result, until result provides equipment to stop obtaining the initial search result corresponding with the search sequence that user inputs.
Model determining device 15 ' is according to the first Search Results training data through mark sequence, and by the mode of machine learning, determine order models, wherein, described order models comprises following at least any one: described first order models, described second order models.Particularly, model determining device 15 ' is according to marking the first Search Results training data sorted, from initial order models or an optional characteristic component as initial order models, and according to the parameter in demand utilization linear model, nonlinear model or its combination constantly adjustment order models, by the mode of machine learning, determine order models, as described in the first order models or as described in the second order models.Wherein, described first Search Results training data includes but not limited to the Search Results (url) of query string (query) and correspondence, every bar query string is indicated and the correlation level numeral of Search Results, or the height relation etc. of described multiple Search Results is indicated to the multiple Search Results wherein having correlativity height to distinguish.Wherein, if described order models is linear model, then include but not limited in described order models each characteristic component and with the weights corresponding to described characteristic component; If described order models is nonlinear model, then can comprise decision-making value as corresponding in the some threshold point with characteristic component in described order models, whole order models is made up of some decision-making values, such as form a decision tree by the decision-making value of multiple characteristic component, then form order models by many decision trees, comprehensively give a mark for Search Results.Such as the first Search Results training data is constantly updated to current order models, as the middle order models that initial order models or study obtain, calculate the sequencing information of this training data, the such as weighted sum of multiple characteristic component with weight information, or to be given a mark the score value obtained by be made up of multiple characteristic component with decision-making value one or many decision trees, and the difference of the sequencing information marked according to this sequencing information and its, front order models is deserved in adjustment, deserve the characteristic component of front order models as increase and decrease or adjust weight information or the decision-making value of its characteristic component, such as adjust or adjust simultaneously weight information or the decision-making value of multiple characteristic component in turn.Those skilled in the art will be understood that at this, by machine learning mode determination order models, not only make the error of order models on the first Search Results training data little as far as possible, also have certain extensive Generalization Ability.
Preferably, in described first Search Results training data, include but not limited to search sequence, Search Results, and the mapping relations between search sequence and Search Results, as the priority of this Search Results under corresponding search sequence, sequence maybe must grade; Preferably, the mapping relations between search sequence with Search Results also comprise weight or the decision-making value of the characteristic component of this Search Results under corresponding search sequence.Such as, for given model prototype, as one comprises the proper vector of multiple characteristic component, but not yet demarcate the parameter of each characteristic component, as weights or the decision-making value of this characteristic component, then utilize the training set comprising the mapping relations marked between its search sequence and Search Results, by machine learning algorithms such as genetic algorithm, neural network, decision tree, support vector machine, determine the model parameter comprising parameter corresponding to each characteristic component, namely obtain order models, as described in the first order models or as described in the second order models.
Preferably, this result provides equipment also to comprise submodel determining device 16 ', wherein, submodel determining device 16 ' is according to the second Search Results training data through mark sequence, by the mode of machine learning, determine one or more sequence submodel for determining characteristic component in described order models.Those skilled in the art will be understood that, except the difference of sequence submodel and order models, the implementation of submodel determining device 16 ' or basic simlarity identical with model determining device 15 ', therefore for simplicity, repeat no more, be only contained in this by reference.Wherein, in described second Search Results training data, include but not limited to search sequence, Search Results, and the mapping relations between search sequence and Search Results, as the priority of this Search Results under corresponding search sequence, sequence maybe must grade; Preferably, the mapping relations between search sequence with Search Results also comprise weight or the decision-making value of the characteristic component of this Search Results under corresponding search sequence.Such as, for given submodel prototype, as one comprises the proper vector of multiple characteristic component, but not yet demarcate the parameter of each characteristic component, as weights or the decision-making value of this characteristic component, then utilize the training set comprising the mapping relations marked between its search sequence and Search Results, by machine learning algorithms such as genetic algorithm, neural network, decision tree, support vector machine, determine the model parameter comprising parameter corresponding to each characteristic component, namely obtain sub-order models.
Preferably, model determining device 15 ' can also according to the first Search Results training data through mark sequence, and by the mode of machine learning, determine described second order models, wherein, described second order models comprises user behavior characteristic component.Particularly, user behavior characteristic information includes but not limited to that user carries out the temporal information of searching for, according to the address information that IP address confirms, user is by clicking, touch, swipe, the operation information etc. about Search Results that the page residence time etc. generates, model determining device 15 ' can utilize the first Search Results training data through mark sequence, utilize genetic algorithm, the machine learning algorithms such as neural network, machine learning is carried out to described user behavior characteristic information, determine weight or the decision-making value of each characteristic component under different user behavior characteristic information, namely the second order models is obtained.At this, user behavior characteristic information can be contained in described first Search Results training data, also can be stored in the search engine providing equipment to be connected via network and result or search in the third party devices such as log database, and from these third party devices, obtaining described user behavior characteristic information by the application programming interfaces (API) that these third party devices provide.
In another preferred embodiment, can by above-mentioned for providing the result of Search Results to provide equipment, combine with existing search engine, form a kind of new search engine, existing search engine can be the Google search engine of such as Google company, the baidu search engine etc. of company of Baidu.
In another preferred embodiment, can by above-mentioned for providing the result of Search Results to provide equipment, combine with existing search engine plug-in unit, form a kind of new search engine plug-in unit, the MSNToolBar etc. that existing search engine plug-in unit can be the GoogleToolBar of such as Google company, despot, Microsoft are searched by the Baidu of company of Baidu.
In another preferred embodiment, equipment can be provided by the above-mentioned result of Search Results that provides, combine with existing browser, form a kind of new browser, existing browser can be IE browser, the netscape browser of Netscape company, Firefox browser, the Chrome browser of Google company, the Maxthon browser of company of roaming, the opera browser of Opera company, 360 browsers of 360 companies, the search dog browser of Sohu.com Inc., the Tengxun TT browser etc. of company of Tengxun of Mozilla company of such as Microsoft company.
In another preferred embodiment, can by above-mentioned for providing the result of Search Results to provide equipment, combine with existing browser plug-in, form a kind of new browser plug-in, existing browser plug-in can be such as Flash plug-in unit, RealPlayer plug-in unit, MMS plug-in unit, MIDI staff plug-in unit, ActiveX plug-in unit etc.
Fig. 3 illustrates a kind of method flow diagram for providing Search Results providing equipment to realize by result according to a further aspect of the present invention; Particularly, result provides equipment in step s1, obtains the initial search result corresponding with the search sequence that user inputs; In step s2, utilize the first order models, in described initial search result, filter out optimal search result; In step s3, utilize the second order models, in described optimal search result, filter out optimum search result; In step s4, described optimum search result is supplied to described user.Wherein, result provides equipment not only can work alone, and can also be integrated in the network equipment, subscriber equipment or the network equipment with subscriber equipment by the mutually integrated equipment formed of network.Wherein, the described network equipment its include but not limited to computing machine, network host, single network server, cloud that multiple webserver collection or multiple server are formed; At this, cloud is formed by based on a large amount of computing machine of cloud computing (CloudComputing) or the webserver, and wherein, cloud computing is the one of Distributed Calculation, the virtual supercomputer be made up of the loosely-coupled computing machine collection of a group.Described subscriber equipment its include but not limited to that any one can to carry out the electronic product of man-machine interaction, such as computing machine, smart mobile phone, PDA, game machine or IPTV etc. with user by keyboard, telepilot, touch pad or voice-operated device.Described network includes but not limited to internet, wide area network, Metropolitan Area Network (MAN), LAN (Local Area Network), VPN, wireless self-organization network (AdHoc network) etc.Those skilled in the art will be understood that other result provides equipment to be equally applicable to the present invention, within also should being included in scope, and are contained in this at this with way of reference.
Wherein, in step s1, result provides equipment to obtain the initial search result corresponding with the search sequence that user inputs.Particularly, in step s1, result provides equipment such as to pass through page technology, as JSP, ASP, the page technology such as PHP, or, the application programming interfaces (API) that the equipment of described search sequence can be provided to provide by invoke user equipment or other, or http, the communication mode of other agreements such as https, carry out alternately with user, obtain the search sequence of user's input, and by such as carrying out participle to the search sequence of user's input, and in Query Database, the mode of searching for is carried out for the search sequence after participle, obtain the initial search result corresponding with the search sequence that user inputs, wherein, user is by such as keyboard, touch-screen, speech input device and result provide equipment to carry out alternately, input the search sequence that it wishes inquiry, thus initiate search, or, result provides equipment to pass through based on various communication protocol (CommunicationsProtocol), the transportation protocol of compunication is referred in this " communication protocol ", as: TCP/IP, UDP, FTP, ICMP, NetBEUI etc., also comprise other form communications be present in computing machine, such as: the communication inside object based programming between object simultaneously, message transmission protocol in operating system between distinct program or computing machine disparate modules, can provide the equipment of described initial search result with other, as search engine, carries out alternately to obtain the initial search result corresponding with the search sequence that user inputs.Preferably, in step s1, result provides equipment can also intercept the Search Results of some in the obtained Search Results corresponding with the search sequence that user inputs, using as described initial search result.Such as, user provides equipment to propose the search sequence of " the niciest Sichuan cuisine " by page technology to result, result provides equipment to carry out participle to " the niciest Sichuan cuisine ", retrieves respectively in a database to " the niciest " and " Sichuan cuisine ", obtains 1000 initial search results.It will be understood by those skilled in the art that mode and several communication transport protocols of above-mentioned acquisition initial search result are only citing; the mode of other acquisition initial search results that are existing or that may occur from now on or communication transport protocols are as being applicable to the present invention; also within scope should being included in, and this is contained at this with way of reference.
In step s2, result provides equipment utilization first order models, in described initial search result, filter out optimal search result.Particularly, in step s2, result provides equipment in its described initial search result provided in step s1, utilizes described first order models, calculates priority or the sequencing information of each initial search result; Again according to these priority or sequencing information, described initial search result is screened, to obtain described optimal search result, as priority or sequencing information are met the initial search result of certain threshold requirement as optimal search result, or by these initial search results by its priority or sequencing information descending sort, and top n initial search result will be come as optimal search result.Wherein, sort algorithm, sequencing feature vector etc. is included but not limited in described first order models.
At this, the first order models calculating priority of initial search result or the mode of sequencing information is utilized to include but not limited to: to utilize the first order models, such as comprise the proper vector of one or more characteristic component and weight thereof, determine the assignment of each characteristic component corresponding to initial search result, thus the proper vector obtained corresponding to this initial search result, namely this proper vector comprises assignment and the weight thereof of these characteristic components, using as the priority of this initial search result or sequencing information; Preferably, the assignment of each characteristic component included by this proper vector and weight thereof can also carry out the assignment that this proper vector is determined in weighting, using as the priority of this initial search result or sequencing information.
At this, the mode that initial search result is undertaken arranging by its priority or sequencing information is included but not limited to: be not general, can suppose that priority or the sequencing information of initial search result comprise the proper vector corresponding with this initial search result, this proper vector comprises assignment and the weight thereof of one or more characteristic component, according to the size of the assignment of each proper vector (as determined by the assignment weighting of its characteristic component), the sequence of corresponding Search Results can be determined; Or, according to weight and the assignment size (as dictionary sequence) thereof of each characteristic component of each proper vector, determine the sequence of corresponding Search Results, such as first sort by the assignment of the highest characteristic component of weight, then for the initial search result that the assignment of the highest characteristic component of its weight is identical, can sort, until complete the sequence of all initial search results by the assignment of the secondary high characteristic component of weight.
Such as, user provides equipment to propose the search sequence of " the niciest Sichuan cuisine " by page technology to result, result provides equipment in step s1, obtain 1000 initial search results, first order models is defined as two components carrying out based on weight each 50% to the initial search result of user's search sequence, authoritative analysis and semantic analysis, sort, result provides equipment according to obtain in step s1 1000 described initial search results, and utilize method that is authoritative and Semantic Ranking to screen described 1000, obtain front 100 results of semantic and authoritative sequence, as described optimal search result.
In step s3, result provides equipment utilization second order models, filters out optimum search result in described optimal search result.Particularly, in step s3, in the described optimal search result that result provides equipment to provide in step s2, described second order models is utilized to carry out further Screening Treatment to the search sequence of described user, to obtain described optimum search result.Except those skilled in the art will be understood that the difference except the second order models and the first order models, the implementation of step s3 or basic simlarity identical with step s2, therefore for simplicity, repeat no more, be only contained in this by reference.Such as, user provides equipment to propose the search sequence of " the niciest Sichuan cuisine " by page technology to result, result provides equipment in step s1, provide 1000 initial search results, result provides equipment in step s2, provide 100 optimal search result, second order models be defined as to the preferred result of user's search sequence carry out based on weight be respectively 25% four components: user requirements analysis, user behavior analysis of statistical results, authoritative analysis and semantic analysis are sorted, in step s3, result provides equipment acquisition 100 optimal search result that result provides equipment to provide in step s2, and utilize the second order models, the integrated ordered of 4 components is carried out to described 100 optimal search result, obtain further ranking results, and using the result of first 10 of sequence as described optimum search result.
In step s4, result provides equipment that described optimum search result is supplied to described user.Particularly, in step s4, result provides the equipment acquisition optimum search result that result provides equipment to filter out in step s3, and utilization and user carry out alternately, or the call format of the communication mode of the application programming interfaces provided according to subscriber equipment (API) or other agreements such as http, https, is supplied to described user by described optimum search result.Such as, 10 that provide equipment to obtain in step s3 result described optimum search results present to user as the homepage of user search result, or obtain in step s3 10 described optimum search results are presented to user as the homepage of user search result, and residue optimal search result is presented to user successively as required after homepage.At this, those skilled in the art will be understood that the display for initial search result, optimal search result or optimum search result, can carry out successively according to optimum search result, optimal search result, initial search result; Also can show from optimum search result, optimal search result, the optional one of initial search result; Also can be by both or triplicity in optimum search result, optimal search result, initial search result, carry out showing according to user's request.
At this, it will be understood by those skilled in the art that this exemplary method can also comprise other for the step of screening intermediate search results and even more multistage screening step, thus multi-stage sequencing is carried out to initial search result, multi-stage sequencing model is such as adopted to sort step by step to initial search result, to be supplied to the optimum search result of user to obtain.Preferably, the present invention can also determine the rank of required order models according to different application demands, and according to the multi-stage sequencing model of appropriate level, as secondary order models, three or more level order models, initial search result is sorted step by step, to be supplied to the optimum search result of user to obtain.
Preferably, in step s2, result provides equipment can also utilize described first order models, determines the priority of described initial search result; According to the first predetermined amount threshold, based on the priority of described initial search result, from described initial search result, filter out described optimal search result, wherein, the quantity of described optimal search result meets described first amount threshold.Particularly, in step s2, result provides the first order models described in equipment utilization, determine the priority of described initial search result, proper vector such as corresponding to this initial search result or its assignment, again according to the first predetermined amount threshold, based on the priority of these initial search results, described optimal search result is filtered out from described initial search result, wherein, the quantity of described optimal search result meets described first amount threshold, from these initial search results, such as directly filter out the higher initial search result of the priority of some as described optimal search result, or first by its priority, descending sort is carried out to these initial search results, then using the initial search result of some of standing out as described optimal search result.Such as, suppose that the first amount threshold is 100, user provides equipment to propose the search sequence of " the niciest Sichuan cuisine " by page technology to result, result provides equipment in step s1, provide 1000 initial search results, first order models is defined as two components carrying out based on weight each 50% to the initial search result of user's search sequence, authoritative analysis and semantic analysis, sort, in step s2, 1000 described initial search results that result provides equipment to obtain it to provide in step s1, and utilize method that is authoritative and Semantic Ranking to determine the priority of these 1000 initial search results, and according to the first amount threshold 100, screen in these 1000 initial search results are by the sequence of its priority descending sort, obtain front 100 results of semantic and authoritative sequence, as described optimal search result.
More preferably, this embodiment also comprises step s7 (not shown), and wherein, in step s7, result provides equipment to determine rule according to the first predetermined quantity, determines described first amount threshold; Wherein, described first quantity determines that rule comprises following at least any one: based on the quantity of described initial search result, determines described first amount threshold; Based on predetermined for determining that the amount threshold of described optimum search result is established rules then really, determine described first amount threshold.Particularly, in step s7, result provides the quantity of equipment according to described initial search result or predetermined for determining that the amount threshold of described optimum search result is established rules then really, dynamically determines described first amount threshold.Such as, the first quantity determines that arranging described first amount threshold in rule is directly proportional to the quantity of described initial search result, then described initial search result quantity is more, and described first amount threshold is larger; Or, the amount threshold that first quantity determines to arrange in rule described first amount threshold and described optimum search result is established rules then positive correlation really, as the restriction according to interface display, it is fixing that the amount threshold of optimum search result limits, then the first amount threshold is directly proportional to the amount threshold of described fixing optimum search result, such as, represent quantity according to every page and determine that the first amount threshold is every page of integral multiple representing quantity; Or the determination the form of the rules positive feedback relation of the determination of described first amount threshold and the quantity of described initial search result and the predetermined amount threshold for determining described optimum search result, finally reaches balance, thus determines described first amount threshold.
Preferably, in step s3, result provides equipment can also utilize described second order models, determines the priority of described optimal search result; According to the second predetermined amount threshold, based on the priority of described optimal search result, filter out described optimum search result from described optimal search result, wherein, the quantity of described optimum search result meets described second amount threshold.Particularly, in step s3, result provides the second order models described in equipment utilization, determine the priority of described optimal search result, proper vector such as corresponding to this optimal search result or its assignment, again according to the second predetermined amount threshold, based on the priority of these optimal search result, described optimum search result is filtered out from described optimal search result, wherein, the quantity of described optimum search result meets described second amount threshold, from these optimal search result, such as directly filter out the higher optimal search result of the priority of some as described optimum search result, or first by its priority, descending sort is carried out to these optimal search result, then using the optimal search result of some of standing out as described optimum search result.
More preferably, this embodiment also comprises step s8 (not shown), and wherein, in step s8, result provides equipment to determine rule according to the second predetermined quantity, determines described second amount threshold; Wherein, described second quantity determines that rule comprises following at least any one: based on the terminal attribute of the subscriber equipment of described user, determines described second amount threshold; Based on the type information of described search sequence, determine described second amount threshold; Based on the quantity of described optimal search result, determine described second amount threshold.Particularly, in step s8, result provides equipment to determine rule according to described the second predetermined quantity, based on the terminal attribute of the subscriber equipment of described user, or based on the type information of described search sequence, or based on the quantity of described optimal search result, determine described second amount threshold.Such as, different according to the terminal attribute of subscriber equipment, the second amount threshold is corresponding difference also, as PC holds display screen relatively large, every page can present 10 results, then the second amount threshold is 10, the display screen of mobile device is relatively little, then the second amount threshold is 6; Or it is different according to the type information of described search sequence, second amount threshold is corresponding difference also, type as search sequence is Bus Schedules, then present the most a small amount of result, second amount threshold is relatively little, type as search sequence is catering information, then present the demand that relatively many results just may meet user; Or be directly proportional to the quantity of described optimal search result, determine described second amount threshold.
Fig. 4 illustrates a kind of method flow diagram for providing Search Results providing equipment to realize by result in accordance with a preferred embodiment of the present invention; Particularly, in step s5 ', result provides equipment according to the first Search Results training data through mark sequence, by the mode of machine learning, determine order models, wherein, described order models comprises following at least any one: described first order models, described second order models; In step s1 ', result provides equipment to obtain the initial search result corresponding with the search sequence that user inputs; In step s2 ', result provides equipment utilization first order models, in described initial search result, filter out optimal search result; In step s3 ', result provides equipment utilization second order models, filters out optimum search result in described optimal search result; In step s4 ', result provides equipment that described optimum search result is supplied to described user.Wherein, result provides the step s1 ' in equipment, step s2 ', step s3 ' and step s4 ' identical with step corresponding shown in Fig. 3 or substantially identical respectively, so place repeats no more, and is contained in this by way of reference.
Constant work between above steps, at this, it will be understood by those skilled in the art that " continuing " refer to above steps respectively according to setting or the mode of operation of in real time adjustment require to carry out the determination of order models, the acquisition of initial search result, the screening of optimal search result, the screening of optimum search result and providing of optimum search result, until result provides equipment to stop obtaining the initial search result corresponding with the search sequence that user inputs.
In step s5 ', result provides equipment according to the first Search Results training data through mark sequence, by the mode of machine learning, determine order models, wherein, described order models comprises following at least any one: described first order models, described second order models.Particularly, step s5 ' is according to marking the first Search Results training data sorted, from initial order models or an optional characteristic component as initial order models, and according to the parameter in demand utilization linear model, nonlinear model or its combination constantly adjustment order models, by the mode of machine learning, determine order models, as described in the first order models or as described in the second order models.Wherein, described first Search Results training data includes but not limited to the Search Results (url) of query string (query) and correspondence, every bar query string is indicated and the correlation level numeral of Search Results, or the height relation etc. of described multiple Search Results is indicated to the multiple Search Results wherein having correlativity height to distinguish.Wherein, if described order models is linear model, then include but not limited in described order models each characteristic component and with the weights corresponding to described characteristic component; If described order models is nonlinear model, then can comprise decision-making value as corresponding in the some threshold point with characteristic component in described order models, whole order models is made up of some decision-making values, such as form a decision tree by the decision-making value of multiple characteristic component, then form order models by many decision trees, comprehensively give a mark for Search Results.Such as the first Search Results training data is constantly updated to current order models, as the middle order models that initial order models or study obtain, calculate the sequencing information of this training data, the such as weighted sum of multiple characteristic component with weight information, or to be given a mark the score value obtained by be made up of multiple characteristic component with decision-making value one or many decision trees, and the difference of the sequencing information marked according to this sequencing information and its, front order models is deserved in adjustment, deserve the characteristic component of front order models as increase and decrease or adjust weight information or the decision-making value of its characteristic component, such as adjust or adjust simultaneously weight information or the decision-making value of multiple characteristic component in turn.Those skilled in the art will be understood that at this, by machine learning mode determination order models, not only make the error of order models on the first Search Results training data little as far as possible, also have certain extensive Generalization Ability.
Preferably, in described first Search Results training data, include but not limited to search sequence, Search Results, and the mapping relations between search sequence and Search Results, as the priority of this Search Results under corresponding search sequence, sequence maybe must grade; Preferably, the mapping relations between search sequence with Search Results also comprise weight or the decision-making value of the characteristic component of this Search Results under corresponding search sequence.Such as, for given model prototype, as one comprises the proper vector of multiple characteristic component, but not yet demarcate the parameter of each characteristic component, as weights or the decision-making value of this characteristic component, then utilize the training set comprising the mapping relations marked between its search sequence and Search Results, by machine learning algorithms such as genetic algorithm, neural network, decision tree, support vector machine, determine the model parameter comprising parameter corresponding to each characteristic component, namely obtain order models, as described in the first order models or as described in the second order models.
Preferably, this embodiment also comprises step s6 ', wherein, in step s6 ', result provides equipment according to the second Search Results training data through mark sequence, by the mode of machine learning, determines one or more sequence submodel for determining characteristic component in described order models.Except those skilled in the art will be understood that the difference except sequence submodel and order models, the implementation of step s6 ' or basic simlarity identical with step s5 ', therefore for simplicity, repeat no more, be only contained in this by reference.Wherein, in described second Search Results training data, include but not limited to search sequence, Search Results, and the mapping relations between search sequence and Search Results, as the priority of this Search Results under corresponding search sequence, sequence maybe must grade; Preferably, the mapping relations between search sequence with Search Results also comprise weight or the decision-making value of the characteristic component of this Search Results under corresponding search sequence.Such as, for given submodel prototype, as one comprises the proper vector of multiple characteristic component, but not yet demarcate the parameter of each characteristic component, as weights or the decision-making value of this characteristic component, then utilize the training set comprising the mapping relations marked between its search sequence and Search Results, by machine learning algorithms such as genetic algorithm, neural network, decision tree, support vector machine, determine the model parameter comprising parameter corresponding to each characteristic component, namely obtain sub-order models.
Preferably, in step s5 ', result provide equipment can also according to through mark sequence the first Search Results training data, by the mode of machine learning, determine described second order models, wherein, described second order models comprises user behavior characteristic component.Particularly, user behavior characteristic information includes but not limited to that user carries out the temporal information of searching for, according to the address information that IP address confirms, user is by clicking, touch, swipe, the operation information etc. about Search Results that the page residence time etc. generates, result provides equipment can utilize the first Search Results training data sorted through mark, utilize genetic algorithm, the machine learning algorithms such as neural network, machine learning is carried out to described user behavior characteristic information, determine weight or the decision-making value of each characteristic component under different user behavior characteristic information, namely the second order models is obtained.At this, user behavior characteristic information can be contained in described first Search Results training data, also can be stored in the search engine providing equipment to be connected via network and result or search in the third party devices such as log database, and from these third party devices, obtaining described user behavior characteristic information by the application programming interfaces (API) that these third party devices provide.
To those skilled in the art, obviously the invention is not restricted to the details of above-mentioned one exemplary embodiment, and when not deviating from spirit of the present invention or essential characteristic, the present invention can be realized in other specific forms.Therefore, no matter from which point, all should embodiment be regarded as exemplary, and be nonrestrictive, scope of the present invention is limited by claims instead of above-mentioned explanation, and all changes be therefore intended in the implication of the equivalency by dropping on claim and scope are included in the present invention.Any Reference numeral in claim should be considered as the claim involved by limiting.In addition, obviously " comprising " one word do not get rid of other unit or step, odd number does not get rid of plural number.Multiple unit of stating in device claim or device also can be realized by software or hardware by a unit or device.First, second word such as grade is used for representing title, and does not represent any specific order.
Claims (18)
1. by computer implemented for providing a method for Search Results, wherein, the method comprises the following steps:
A obtains the initial search result corresponding with the search sequence that user inputs;
B utilizes the first order models, in described initial search result, filter out optimal search result;
C utilizes the second order models, filters out optimum search result in described optimal search result, and wherein, described second order models comprises the user behavior characteristic component corresponding with described user;
Described optimum search result is supplied to described user by d.
2. method according to claim 1, wherein, the method also comprises:
X, according to the first Search Results training data through mark sequence, by the mode of machine learning, determines order models;
Wherein, described order models comprises following at least any one:
-described first order models;
-described second order models.
3. method according to claim 2, wherein, the method also comprises:
-according to the second Search Results training data through mark sequence, by the mode of machine learning, determine one or more sequence submodel for determining characteristic component in described order models.
4. according to the method in any one of claims 1 to 3, wherein, described step b comprises:
-utilize described first order models, determine the priority of described initial search result;
-according to the first predetermined amount threshold, based on the priority of described initial search result, from described initial search result, filter out described optimal search result, wherein, the quantity of described optimal search result meets described first amount threshold.
5. method according to claim 4, wherein, the method also comprises:
-determine rule according to the first predetermined quantity, determine described first amount threshold;
Wherein, described first quantity determines that rule comprises following at least any one:
-based on the quantity of described initial search result, determine described first amount threshold;
-based on predetermined for determining that the amount threshold of described optimum search result is established rules then really, determine described first amount threshold.
6. according to the method in any one of claims 1 to 3, wherein, described step c comprises:
-utilize described second order models, determine the priority of described optimal search result;
-according to the second predetermined amount threshold, based on the priority of described optimal search result, filter out optimum search result from described optimal search result, wherein, the quantity of described optimum search result meets described second amount threshold.
7. method according to claim 6, wherein, the method also comprises:
-determine rule according to the second predetermined quantity, determine described second amount threshold;
Wherein, described second quantity determines that rule comprises following at least any one:
-based on the terminal attribute of the subscriber equipment of described user, determine described second amount threshold;
-based on the type information of described search sequence, determine described second amount threshold;
-based on the quantity of described optimal search result, determine described second amount threshold.
8., for providing the result of Search Results to provide an equipment, wherein, this equipment comprises:
Result acquisition device, for the initial search result that the search sequence obtained with user inputs is corresponding;
First screening plant, for utilizing the first order models, filters out optimal search result in described initial search result;
Second screening plant, for utilizing the second order models, filters out optimum search result in described optimal search result, and wherein, described second order models comprises the user behavior characteristic component corresponding with described user;
Result generator, for being supplied to described user by described optimum search result.
9. result according to claim 8 provides equipment, and wherein, this equipment also comprises:
Model determining device, for according to the first Search Results training data through mark sequence, by the mode of machine learning, determines order models;
Wherein, described order models comprises following at least any one:
-described first order models;
-described second order models.
10. result according to claim 9 provides equipment, and wherein, this equipment also comprises:
Submodel determining device, for according to the second Search Results training data through mark sequence, by the mode of machine learning, determines one or more sequence submodel for determining characteristic component in described order models.
Result according to any one of 11. according to Claim 8 to 10 provides equipment, and wherein, described first screening plant is used for:
-utilize described first order models, determine the priority of described initial search result;
-according to the first predetermined amount threshold, based on the priority of described initial search result, from described initial search result, filter out described optimal search result, wherein, the quantity of described optimal search result meets described first amount threshold.
12. results according to claim 11 provide equipment, and wherein, this equipment also comprises:
First threshold determining device, for determining rule according to the first predetermined quantity, determines described first amount threshold;
Wherein, described first quantity determines that rule comprises following at least any one:
-based on the quantity of described initial search result, determine described first amount threshold;
-based on predetermined for determining that the amount threshold of described optimum search result is established rules then really, determine described first amount threshold.
Result according to any one of 13. according to Claim 8 to 10 provides equipment, and wherein, described second screening plant is used for:
-utilize described second order models, determine the priority of described optimal search result;
-according to the second predetermined amount threshold, based on the priority of described optimal search result, filter out optimum search result from described optimal search result, wherein, the quantity of described optimum search result meets described second amount threshold.
14. results according to claim 13 provide equipment, and wherein, this equipment also comprises:
Second Threshold determining device, for determining rule according to the second predetermined quantity, determines described second amount threshold;
Wherein, described second quantity determines that rule comprises following at least any one:
-based on the terminal attribute of the subscriber equipment of described user, determine described second amount threshold;
-based on the type information of described search sequence, determine described second amount threshold;
-based on the quantity of described optimal search result, determine described second amount threshold.
15. 1 kinds of search engines, comprise according to any one of claim 8 to 14 for providing the result of Search Results to provide equipment.
16. 1 kinds of search engine plug-in units, comprise according to any one of claim 8 to 14 for providing the result of Search Results to provide equipment.
17. 1 kinds of browsers, comprise according to any one of claim 8 to 14 for providing the result of Search Results to provide equipment.
18. 1 kinds of browser plug-ins, comprise according to any one of claim 8 to 14 for providing the result of Search Results to provide equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210226803.XA CN102810117B (en) | 2012-06-29 | 2012-06-29 | A kind of for providing the method and apparatus of Search Results |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210226803.XA CN102810117B (en) | 2012-06-29 | 2012-06-29 | A kind of for providing the method and apparatus of Search Results |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102810117A CN102810117A (en) | 2012-12-05 |
| CN102810117B true CN102810117B (en) | 2016-02-24 |
Family
ID=47233823
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210226803.XA Active CN102810117B (en) | 2012-06-29 | 2012-06-29 | A kind of for providing the method and apparatus of Search Results |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102810117B (en) |
Families Citing this family (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103164535B (en) * | 2013-04-11 | 2016-08-03 | 阔地教育科技有限公司 | A kind of data search realization method and system of cloud teaching platform |
| CN103164536B (en) * | 2013-04-11 | 2016-08-03 | 阔地教育科技有限公司 | A kind of method and system realizing cloud education platform data search |
| KR101947229B1 (en) * | 2013-06-08 | 2019-02-12 | 애플 인크. | Device, method, and graphical user interface for synchronizing two or more displays |
| CN104462293A (en) * | 2014-11-27 | 2015-03-25 | 百度在线网络技术（北京）有限公司 | Search processing method and method and device for generating search result ranking model |
| CN104598611B (en) * | 2015-01-29 | 2018-03-23 | 百度在线网络技术（北京）有限公司 | The method and system being ranked up to search entry |
| US10635725B2 (en) * | 2015-04-09 | 2020-04-28 | Google Llc | Providing app store search results |
| CN106897289B (en) * | 2015-12-18 | 2020-07-10 | 北京奇虎科技有限公司 | Information search optimization method and device |
| CN105701155B (en) * | 2015-12-30 | 2019-05-31 | 百度在线网络技术（北京）有限公司 | Information-pushing method and device |
| CN107169573A (en) * | 2017-05-05 | 2017-09-15 | 第四范式（北京）技术有限公司 | Using composite machine learning model come the method and system of perform prediction |
| CN107169574A (en) * | 2017-05-05 | 2017-09-15 | 第四范式（北京）技术有限公司 | Using nested machine learning model come the method and system of perform prediction |
| CN107301227A (en) | 2017-06-21 | 2017-10-27 | 北京百度网讯科技有限公司 | Search information analysis method and device based on artificial intelligence |
| CN107463704B (en) * | 2017-08-16 | 2021-05-07 | 北京百度网讯科技有限公司 | Search method and device based on artificial intelligence |
| 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 |
| CN108829757B (en) * | 2018-05-28 | 2022-01-28 | 广州麦优网络科技有限公司 | Intelligent service method, server and storage medium for chat robot |
| CN109033140B (en) * | 2018-06-08 | 2020-05-29 | 北京百度网讯科技有限公司 | Method, device, equipment and computer storage medium for determining search result |
| CN109410090A (en) * | 2018-09-13 | 2019-03-01 | 张继升 | A kind of Intelligent life service platform of function high integrity |
| CN109436834B (en) * | 2018-09-25 | 2021-07-06 | 北京金茂绿建科技有限公司 | Method and device for selecting funnel |
| CN111177585A (en) * | 2018-11-13 | 2020-05-19 | 北京四维图新科技股份有限公司 | Map POI feedback method and device |
| CN109522345A (en) * | 2018-11-19 | 2019-03-26 | 百度在线网络技术（北京）有限公司 | For showing the method and device of information |
| CN110674401B (en) * | 2019-09-19 | 2022-04-15 | 北京字节跳动网络技术有限公司 | Method and device for determining sequence of search items and electronic equipment |
| CN110717263A (en) * | 2019-09-27 | 2020-01-21 | 中国人民解放军海军大连舰艇学院 | Combat model management system |
| CN110688458A (en) * | 2019-09-27 | 2020-01-14 | 中国人民解放军海军大连舰艇学院 | Combat model retrieval method |
| CN111209378B (en) * | 2019-12-26 | 2024-03-12 | 航天信息股份有限公司企业服务分公司 | Ordered hierarchical ordering method based on business dictionary weights |
| CN113761084B (en) * | 2020-06-03 | 2023-08-08 | 北京四维图新科技股份有限公司 | POI search ranking model training method, ranking device, method and medium |
| CN113360796A (en) * | 2021-05-20 | 2021-09-07 | 北京三快在线科技有限公司 | Data sorting method and device, and data sorting model training method and device |
| CN113656574B (en) * | 2021-10-19 | 2022-02-08 | 北京欧应信息技术有限公司 | Method, computing device and storage medium for search result ranking |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101930438A (en) * | 2009-06-19 | 2010-12-29 | 阿里巴巴集团控股有限公司 | Search result generating method and information search system |
| CN102043846A (en) * | 2010-12-16 | 2011-05-04 | 上海电机学院 | Search method and device based on genetic algorithm |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8126839B2 (en) * | 2008-06-19 | 2012-02-28 | Yahoo! Inc. | Methods and apparatuses for adapting a ranking function of a search engine for use with a specific domain |
| CN101770521A (en) * | 2010-03-11 | 2010-07-07 | 东华大学 | Focusing relevancy ordering method for vertical search engine |
-
2012
- 2012-06-29 CN CN201210226803.XA patent/CN102810117B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101930438A (en) * | 2009-06-19 | 2010-12-29 | 阿里巴巴集团控股有限公司 | Search result generating method and information search system |
| CN102043846A (en) * | 2010-12-16 | 2011-05-04 | 上海电机学院 | Search method and device based on genetic algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102810117A (en) | 2012-12-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102810117B (en) | A kind of for providing the method and apparatus of Search Results | |
| CN103593373A (en) | Search result sorting method and search result sorting device | |
| CN101986306B (en) | Method and equipment for acquiring yellow page information based on query sequence | |
| CN102609473B (en) | Method and system for website accessing | |
| US9928282B2 (en) | Method and apparatus for providing search results on a mobile terminal | |
| CN102682099B (en) | Method and device for providing target pictures in target areas | |
| CN105677765B (en) | Recommend the method and system of desired functional sequence to user | |
| CN109902708A (en) | A kind of recommended models training method and relevant apparatus | |
| CN107463704A (en) | Searching method and device based on artificial intelligence | |
| CN103226393A (en) | Input method and equipment | |
| CN110704739A (en) | Resource recommendation method and device and computer storage medium | |
| CN102968454B (en) | A kind of for obtaining the method and apparatus promoting object search results | |
| CN103189836A (en) | Method for classification of objects in a graph data stream | |
| KR20170080645A (en) | Method and apparatus for determining quality information about to-be-commented item | |
| CN102779136A (en) | Method and device for information search | |
| CN103514191A (en) | Method and device for determining keyword matching mode of target popularization information | |
| CN103577415B (en) | The method and apparatus of corresponding search configuration is applied for updating mobile search | |
| CN102236710A (en) | Method and equipment for displaying news information in query result | |
| CN102855261A (en) | Method and device for determining video authority value | |
| CN103902597A (en) | Method and device for determining search relevant categories corresponding to target keywords | |
| CN103810184A (en) | Method for determining web page address velocity, optimization method and device of methods | |
| CN111639253B (en) | Data weight judging method, device, equipment and storage medium | |
| CN104778283B (en) | A kind of user's occupational classification method and system based on microblogging | |
| CN102945273B (en) | A kind of for providing the method and apparatus of Search Results | |
| CN107679119A (en) | The method and apparatus for generating brand derivative words |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |