CN102810117A - Method and equipment for supplying search result - Google Patents
Method and equipment for supplying search result Download PDFInfo
- Publication number
- CN102810117A CN102810117A CN201210226803XA CN201210226803A CN102810117A CN 102810117 A CN102810117 A CN 102810117A CN 201210226803X A CN201210226803X A CN 201210226803XA CN 201210226803 A CN201210226803 A CN 201210226803A CN 102810117 A CN102810117 A CN 102810117A
- Authority
- CN
- China
- Prior art keywords
- search results
- result
- search
- confirm
- order models
- 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.)
- Granted
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention aims to provide a method and equipment for supplying a search result. Computer equipment acquires initial search results corresponding to a query sequence input by a user; a first sequencing model is used for screening out preferable search results from the initial search results; a second sequencing model is used for screening out the optimal search results from the preferable search results; and the optimal search results are supplied to the user. Compared with the prior art, the method and the equipment have the advantages that the search results are screened according to different levels through a plurality of sequencing models, so that the optimal search results are supplied to the user; and therefore, the precision and the efficiency of the search results are considered, and the search efficiency and the search effect are optimized. Moreover, the sequencing models are determined by a machine learning method; an upper layer model is generated by a submodel through the machine learning; and therefore, the arrangement of the sequencing models is optimized; and the instantaneity, the understandability and the controllability of the sequencing models of the systems are guaranteed.
Description
Technical field
The present invention relates to computer realm, relate in particular to a kind of technology that is used to provide Search Results.
Background technology
Current, for the mode of most employing one minor sort of providing of Search Results,,, utilize the order models that presets through inquiry to background data base promptly according to user's query requests, the Search Results of respective user query requests is offered the user.This mode exists certain problem, and promptly a minor sort is not easy to reach the optimization on efficient and the effect.From the efficiency comes first angle, can utilize the lower inquiry mode of precision to carry out, but can't guarantee the pin-point accuracy of Query Result; From the preferential angle of effect, can utilize the higher inquiry mode of precision to carry out, but can't guarantee search efficiency simultaneously.
Summary of the invention
The purpose of this invention is to provide a kind of method and apparatus that is used to provide Search Results.
According to an aspect of the present invention, provide a kind of by the computer implemented method that is used to provide Search Results, this method may further comprise the steps:
A obtains the corresponding initial search result of search sequence with user's input;
B utilizes first order models, in said initial search result, filters out preferred Search Results;
C utilizes second order models, in said preferred Search Results, filters out the optimum search result;
D offers said user with said optimum search result.
According to a further aspect in the invention, also provide a kind of and be used to provide the result of Search Results that equipment is provided, this equipment comprises:
Deriving means as a result is used to obtain the corresponding initial search result of search sequence with user's input;
First screening plant is used to utilize first order models, in said initial search result, filters out preferred Search Results;
Second screening plant is used to utilize second order models, in said preferred Search Results, filters out the optimum search result;
Generator is used for said optimum search result is offered said user as a result.
In accordance with a further aspect of the present invention, a kind of search engine is provided also, has comprised like above-mentioned being used to providing the result of Search Results that equipment is provided.
In accordance with a further aspect of the present invention, a kind of search engine plug-in unit is provided also, has comprised like above-mentioned being used to providing the result of Search Results that equipment is provided.
In accordance with a further aspect of the present invention, a kind of browser is provided also, has comprised like above-mentioned being used to providing the result of Search Results that equipment is provided.
In accordance with a further aspect of the present invention, a kind of browser plug-in is provided also, has comprised like above-mentioned being used to providing the result of Search Results that equipment is provided.
Compared with prior art, the present invention carries out sizing screening to Search Results through a plurality of order models, has realized the optimum search result is offered the user, thereby has taken into account the precision and the efficient of Search Results, reaches the optimization of search efficiency and effect.Further, utilize the method for machine learning to confirm order models, and utilize submodel to generate layer model, thereby optimized the setting of order models, guaranteed real-time, intelligibility and the controllability of system's order models through machine learning.In addition; Current feature selecting for order models is that all low-level image features of various angles are put together mostly, and the problem of bringing like this is to have weakened real-time, intelligibility and controllability; The location that is unfavorable for problem, the algorithm accumulation is multiplexing before also being unfavorable for; Given this, the present invention also uses different character for different order models, and in the optimized while that guarantees search efficiency and effect, the feature selecting of optimization sorting model has further been improved the optimization of search efficiency and effect.
Description of drawings
Through reading the detailed description of doing with reference to following accompanying drawing that non-limiting example is done, it is more obvious that other features, objects and advantages of the present invention will become:
Fig. 1 illustrates according to a kind of of one aspect of the invention and is used to provide the result of Search Results that the equipment synoptic diagram is provided;
Fig. 2 illustrates in accordance with a preferred embodiment of the present invention a kind of and is used to provide the result of Search Results that the equipment synoptic diagram is provided;
Fig. 3 illustrates a kind of method flow diagram that is used to provide Search Results that provides equipment to realize by the result according to a further aspect of the present invention;
Fig. 4 illustrates a kind of method flow diagram that is used to provide Search Results that provides equipment to realize by the result in accordance with a preferred embodiment of the present invention.
Same or analogous Reference numeral is represented same or analogous parts in the accompanying drawing.
Embodiment
Below in conjunction with accompanying drawing the present invention is described in further detail.
Fig. 1 illustrates according to a kind of of one aspect of the invention and is used to provide the result of Search Results that the equipment synoptic diagram is provided; Wherein, this result provides equipment to comprise as a result deriving means 11, first screening plant 12, second screening plant 13, generator 14 as a result.The corresponding initial search result of search sequence that deriving means 11 obtains and the user imports as a result; First screening plant 12 utilizes first order models, in said initial search result, filters out preferred Search Results; Second screening plant 13 utilizes second order models, in said preferred Search Results, filters out the optimum search result; Generator 14 offers said user with said optimum search result as a result.Wherein, the result provides equipment not only can work alone, and can also be integrated in the network equipment, subscriber equipment or the network equipment and subscriber equipment through the mutually integrated equipment that constitutes of network.Wherein, the said network equipment its include but not limited to the cloud that computing machine, network host, single network server, a plurality of webserver collection or a plurality of server constitute; At this, cloud is by constituting based on the great amount of calculation machine of cloud computing (Cloud Computing) or the webserver, and wherein, cloud computing is a kind of of Distributed Calculation, a virtual supercomputer of being made up of the loosely-coupled computing machine collection of a group.It includes but not limited to any electronic product that can carry out man-machine interaction through keyboard, telepilot, touch pad or voice-operated device with the user, for example computing machine, smart mobile phone, PDA, game machine or IPTV etc. said subscriber equipment.Said network includes but not limited to internet, wide area network, Metropolitan Area Network (MAN), LAN, VPN network, 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, also should be included in the protection domain of the present invention, and be contained in this at this with way of reference.
Wherein, the corresponding initial search result of search sequence that deriving means 11 obtains and the user imports as a result.Particularly, deriving means 11 for example passes through page technology as a result, like page technology such as JSP, ASP, PHP; Perhaps, the application programming interfaces that equipment provided (API) of said search sequence can be provided through invoke user equipment or other, or the communication mode of other agreements such as http, https; Carry out alternately with the user, obtain the search sequence of user's input, and through carrying out participle such as the search sequence that the user is imported; And the mode of in Query Database, searching for to the search sequence behind the participle; Obtain the corresponding initial search result of search sequence with user's input, wherein, the user can through such as keyboard, touch-screen, speech input device and as a result deriving means 11 carry out alternately; Import the search sequence that it hopes inquiry, thereby initiate search; Perhaps; Deriving means 11 passes through based on various communication protocols (Communications Protocol) as a result; The transportation protocol that refers to compunication in this " communication protocol "; As: TCP/IP, UDP, FTP, ICMP, NetBEUI etc. also comprise other forms of communication that are present in the computing machine, for example: the communication between the object of OOP the inside simultaneously; Message transmission protocol in the operating system between distinct program or the computing machine disparate modules can provide the equipment of said initial search result with other, like search engine, carries out alternately to obtain the corresponding initial search result of importing with the user of search sequence.Preferably, deriving means 11 can also be at the Search Results of intercepting some in that obtained and the corresponding Search Results of search sequence user input as a result, with as said initial search result.For example; The user has proposed the search sequence of " the niciest Sichuan cuisine " through page technology to deriving means 11 as a result; Deriving means 11 carries out participle to " the niciest Sichuan cuisine " as a result, in database, respectively " the niciest " and " Sichuan cuisine " is retrieved, and has obtained 1000 initial search results.It will be understood by those skilled in the art that above-mentioned mode and several kinds of communication transport protocols of obtaining initial search result are merely for example; Other the existing or modes of obtaining initial search result that possibly occur from now on or communication transport protocols are as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.
At this; Utilize the priority of first order models calculating initial search result or the mode of sequencing information to include but not limited to: to utilize first order models; The proper vector that for example comprises one or more characteristic components and weight thereof; Confirm the assignment of pairing each characteristic component of initial search result; Thereby obtain the pairing proper vector of this initial search result, promptly this proper vector comprises the assignment and the weight thereof of these characteristic components, with priority or the sequencing information as this initial search result; Preferably, can also come weighting to confirm the assignment of this proper vector according to the assignment and the weight thereof of each included characteristic component of this proper vector, with priority or sequencing information as this initial search result.
At this; The mode that initial search result is arranged by its priority or sequencing information includes but not limited to: be not general; Priority or the sequencing information that can suppose initial search result comprise and the corresponding proper vector of this initial search result; This proper vector comprises the assignment and the weight thereof of one or more characteristic components, can confirm corresponding search result's ordering according to the assignment of each proper vector size of (as being confirmed by the assignment weighting of its characteristic component); Perhaps; Weight and assignment size (like the dictionary ordering) thereof according to each characteristic component of each proper vector; Confirm corresponding search result's ordering, for example at first sort, then for the identical initial search result of assignment of the highest characteristic component of its weight by the assignment of the highest characteristic component of weight; Can sort by the assignment of time high characteristic component of weight, until the ordering of accomplishing all initial search results.For example; The user has proposed the search sequence of " the niciest Sichuan cuisine " through page technology to deriving means 11 as a result, and deriving means 11 provides 1000 initial search results as a result, and the initial search result that first order models is defined as the user inquiring sequence carries out based on each two component of 50% of weight; The authoritative analysis and semantic analysis; Sort, first screening plant 12 obtains 1000 said initial search results that deriving means 11 is as a result provided, and utilizes authoritative method with semantic ordering that said 1000 are screened; Obtain preceding 100 results semantic and authoritative ordering, as said preferred Search Results.
At this; It will be understood by those skilled in the art that the result provides equipment can also comprise three screening device and even multistage screening device more; Thereby initial search result is carried out multi-stage sequencing; For example adopt the multi-stage sequencing model that initial search result is sorted step by step, to obtain the user's of giving to be supplied optimum search result.Preferably; The present invention can also confirm the rank of needed order models according to different application requirements; And according to the multi-stage sequencing model of appropriate level; Like secondary order models, three grades or multi-stage sequencing model more, initial search result is sorted step by step, to obtain the user's of giving to be supplied optimum search result.
Preferably, first screening plant 12 can also utilize said first order models, confirms the priority of said initial search result; First amount threshold according to predetermined based on the priority of said initial search result, filters out said preferred Search Results from said initial search result, wherein, the quantity of said preferred Search Results satisfies said first amount threshold.Particularly; First screening plant 12 utilizes said first order models, confirms the priority of said initial search result, for example the pairing proper vector of this initial search result or its assignment; Again according to the first predetermined amount threshold; Based on the priority of these initial search results, from said initial search result, filter out said preferred Search Results, wherein; The quantity of said preferred Search Results satisfies said first amount threshold; For example from these initial search results, directly filter out the higher initial search result of the priority of some as said preferred Search Results, perhaps earlier these initial search results carried out descending sort by its priority, then with the initial search result of the some of standing out as said preferred Search Results.For example, suppose that first amount threshold is 100, the user has proposed the search sequence of " the niciest Sichuan cuisine " through page technology to deriving means 11 as a result; Deriving means 11 provides 1000 initial search results as a result; The initial search result that first order models is defined as the user inquiring sequence carries out based on each two component of 50% of weight, and authoritative the analysis and semantic analysis sorted; First screening plant 12 obtains 1000 said initial search results that deriving means 11 is as a result provided; And utilize authoritative method to confirm the priority of these 1000 initial search results, and, in the sequence of these 1000 initial search results, screen by its priority descending sort according to first amount threshold 100 with semantic ordering; Obtain preceding 100 results semantic and authoritative ordering, as said preferred Search Results.
More preferably, the result provides equipment to comprise that also first threshold confirms the device (not shown), and wherein, first threshold confirms that device confirms rule according to the first predetermined quantity, confirms said first amount threshold; Wherein, said first quantity confirm rule comprise following at least each: based on the quantity of said initial search result, confirm said first amount threshold; The amount threshold that is used for definite said optimum search result based on predetermined is established rules then really, confirms said first amount threshold.Particularly, first threshold confirms that device confirms that according to the quantity of said initial search result or predetermined being used for said optimum search result's amount threshold establishes rules then really, comes dynamically to confirm said first amount threshold.For example, first quantity is confirmed to be provided with in the rule said first amount threshold and is directly proportional with the quantity of said initial search result, and then said initial search result quantity is many more, and said first amount threshold is big more; Perhaps; The amount threshold that first quantity confirms to be provided with in the rule said first amount threshold and the said optimum search result then positive correlation of establishing rules really; Like restriction according to interface display; Optimum search result's amount threshold limits and fixes, and then first amount threshold is directly proportional with said fixing optimum search result's amount threshold, for example represents quantity according to every page and confirms that first amount threshold is every page of integral multiple that represents quantity; Perhaps, confirm and the quantity of said initial search result and predetermined being used for of said first amount threshold confirmed that said optimum search result's amount threshold is established rules really and then formed the positive feedback relation, finally reach balance, thereby confirm said first amount threshold.
Preferably, second screening plant 13 can also utilize said second order models, confirms the priority of said preferred Search Results; Second amount threshold according to predetermined based on the priority of said preferred Search Results, filters out said optimum search result from said preferred Search Results, wherein, said optimum search result's quantity satisfies said second amount threshold.Particularly; Second screening plant 13 utilizes said second order models, confirms the priority of said preferred Search Results, for example the pairing proper vector of this preferred Search Results or its assignment; Again according to the second predetermined amount threshold; Based on the priority of these preferred Search Results, from said preferred Search Results, filter out said optimum search result, wherein; Said optimum search result's quantity satisfies said second amount threshold; For example from these preferred Search Results, directly filter out the higher preferred Search Results of the priority of some as said optimum search result, perhaps earlier these preferred Search Results carried out descending sort by its priority, then with the preferred Search Results of the some of standing out as said optimum search result.
More preferably, the result provides equipment to comprise that also second threshold value confirms the device (not shown), and wherein, second threshold value confirms that device confirms rule according to the second predetermined quantity, confirms said second amount threshold; Wherein, said second quantity confirm rule comprise following at least each: based on the terminal attribute of said user's subscriber equipment, confirm said second amount threshold; Based on the type information of said search sequence, confirm said second amount threshold; Based on the quantity of said preferred Search Results, confirm said second amount threshold.Particularly; Second threshold value confirms that device confirms rule according to the said second predetermined quantity, based on the terminal attribute of said user's subscriber equipment, or based on the type information of said search sequence; Or, confirm said second amount threshold based on the quantity of said preferred Search Results.For example, different according to the terminal attribute of subscriber equipment, the also corresponding difference of second amount threshold; Relatively large like PC end display screen, every page can present 10 results, and then second amount threshold is 10; The display screen of mobile device is less relatively, and then second amount threshold is 6; Or it is different according to the type information of said search sequence; The also corresponding difference of second amount threshold; Type like search sequence is a train number information, demonstrates then the most accurately that few results gets final product, and second amount threshold is less relatively; Type like search sequence is a catering information, then demonstrates the demand that more relatively result just possibly satisfy the user; Or be directly proportional with the quantity of said preferred Search Results, confirm said second amount threshold.
Fig. 2 illustrates in accordance with a preferred embodiment of the present invention a kind of and is used to provide the result of Search Results that the equipment synoptic diagram is provided; Wherein, this result provides that equipment comprises as a result deriving means 11 ', first screening plant 12 ', second screening plant 13 ', generator 14 ', model are confirmed device 15 ' as a result.Particularly, model confirms that device 15 ' is according to the first Search Results training data through the mark ordering, through the mode of machine learning; Confirm order models; Wherein, said order models comprise following at least each: said first order models, said second order models; The corresponding initial search result of search sequence that deriving means 11 ' obtains and the user imports as a result; First screening plant 12 ' utilizes first order models, in said initial search result, filters out preferred Search Results; Second screening plant 13 ' utilizes second order models, in said preferred Search Results, filters out the optimum search result; Generator 14 ' offers said user with said optimum search result as a result.Wherein, The result provides deriving means 11 ' as a result in the equipment, first screening plant 12 ', second screening plant 13 ', generator 14 ' is identical with corresponding intrument shown in Figure 1 respectively or basic identical as a result; So locate to repeat no more, and mode by reference is contained in this.
Be constant work between above-mentioned each device; At this; It will be understood by those skilled in the art that " continuing " is meant that above-mentioned each device requires to carry out screening, the optimum search result's of the obtaining of the confirming of order models, initial search result, preferred Search Results the providing etc. of screening and optimum search result respectively according to mode of operation that set or adjustment in real time, provide equipment to stop to obtain the corresponding initial search result of search sequence with user's input until the result.
Model confirms that device 15 ' according to the first Search Results training data through mark ordering, through the mode of machine learning, confirms order models, wherein, said order models comprise following at least each: said first order models, said second order models.Particularly; Model confirms that device 15 ' is according to marking the first Search Results training data that ordering is accomplished; From initial order models or the initial order models of optional characteristic component conduct, and according to the parameter in demand utilization linear model, nonlinear model or the continuous adjustment of its combination order models, through the mode of machine learning; Confirm order models, like said first order models or said second order models.Wherein, The said first Search Results training data includes but not limited to query string (query) and corresponding search result (url); Every query string is indicated and relevance of search results grade numeral, perhaps a plurality of Search Results that wherein have correlativity just to distinguish are indicated the height relation etc. of said a plurality of Search Results.Wherein, if said order models is a linear model, include but not limited in the then said order models that each characteristic component reaches and the pairing weights of said characteristic component; If said order models is a nonlinear model; Can comprise in the then said order models like the decision-making value corresponding with some threshold point of characteristic component; Whole order models is made up of some decision-making values; For example the decision-making value by a plurality of characteristic components constitutes a decision tree, constitutes order models by many decision trees then, Search Results is comprehensively given a mark being used for.For example the first Search Results training data constantly is updated to current order models; The middle order models that obtains like initial order models or study; Calculate the sequencing information of this training data; For example a plurality of weighted sums that have the characteristic component of weight information; Perhaps through constituting or many score values that decision tree marking obtains by a plurality of characteristic components that have decision-making value, and according to the difference of this sequencing information and its sequencing information that has marked, preceding order models is deserved in adjustment; As increasing and decreasing characteristic component that deserves preceding order models or weight information or the decision-making value of adjusting its characteristic component, for example adjust or adjust simultaneously the weight information or the decision-making value of a plurality of characteristic components in order.Those skilled in the art will be understood that at this, confirm order models through the machine learning mode, make that not only the error of order models on the first Search Results training data is as far as possible little, also have certain extensive popularization ability.
Preferably, include but not limited to search sequence in the said first Search Results training data, Search Results, and the mapping relations between search sequence and the Search Results, priority, the ordering under corresponding search sequence maybe must grade like this Search Results; Preferably, search sequence and the mapping relations between the Search Results also comprise the weight or the decision-making value of the characteristic component of this Search Results under corresponding search sequence.For example; For given model prototype, like a proper vector that comprises a plurality of characteristic components, but the parameter of not demarcating each characteristic component as yet; Weights or decision-making value like this characteristic component; Then utilize to comprise the training set that marks the mapping relations between its search sequence and the Search Results,, confirm to comprise the model parameter of each characteristic component corresponding parameters through machine learning algorithms such as genetic algorithm, neural network, decision tree, SVMs; Promptly obtain order models, like said first order models or said second order models.
Preferably; This result provides equipment to comprise that also submodel confirms device 16 '; Wherein, Submodel is confirmed device 16 ' according to the second Search Results training data through the mark ordering, through the mode of machine learning, confirms one or more ordering submodels that are used for confirming said order models characteristic component.Those skilled in the art will be understood that; Except the difference of ordering submodel and order models, submodel confirms that the implementation of device 16 ' confirms that with model device 15 ' is identical or similar basically, so for simplicity; Repeat no more, only be contained in this by reference.Wherein, include but not limited to search sequence in the said second Search Results training data, Search Results, and the mapping relations between search sequence and the Search Results, priority, the ordering under corresponding search sequence maybe must grade like this Search Results; Preferably, search sequence and the mapping relations between the Search Results also comprise the weight or the decision-making value of the characteristic component of this Search Results under corresponding search sequence.For example, for given submodel prototype, like a proper vector that comprises a plurality of characteristic components; But do not demarcate the parameter of each characteristic component as yet; Like the weights or the decision-making value of this characteristic component, then utilize to comprise the training set that marks the mapping relations between its search sequence and the Search Results, through machine learning algorithms such as genetic algorithm, neural network, decision tree, SVMs; Confirm to comprise the model parameter of each characteristic component corresponding parameters, promptly obtain sub-order models.
Preferably, model confirms that device 15 ' can also mark the first Search Results training data of ordering according to warp, through the mode of machine learning, confirm said second order models, and wherein, said second order models comprises the user behavior characteristic component.Particularly; The address information that the user behavior characteristic information includes but not limited to temporal information that the user searches for, confirm according to IP address, user through click, touch, swipe, the page residence time etc. generated about the operation information of Search Results etc.; Model confirms that device 15 ' can utilize the first Search Results training data through the mark ordering; Utilize machine learning algorithms such as genetic algorithm, neural network; Said user behavior characteristic information is carried out machine learning, confirm the weight or the decision-making value of each characteristic component under the different user behavior characteristic information, promptly obtain second order models.At this; The user behavior characteristic information can be contained in the said first Search Results training data; Also can be stored in via network and result provides in third party's equipment such as search engine that equipment is connected or search log database, and from these third party devices, obtains said user behavior characteristic information through the application programming interfaces (API) that these third party devices provided.
In another preferred embodiment; Can be used to provide the result of Search Results that equipment is provided with above-mentioned; Combine with existing search engine; Constitute a kind of new search engine, existing search engine can be the Google search engine of for example Google company, the baidu search engine of company of Baidu etc.
In another preferred embodiment; Can be used to provide the result of Search Results that equipment is provided with above-mentioned; Combine with existing search engine plug-in unit; Constitute a kind of new search engine plug-in unit, the MSN ToolBar of despot, Microsoft etc. searches in the Google ToolBar that existing search engine plug-in unit can be a for example Google company, the Baidu of company of Baidu.
In another preferred embodiment; Can equipment be provided with the above-mentioned result of Search Results that provides; Combine with existing browser; Constitute a kind of new browser, existing browser can be the Maxthon browser of the IE browser of for example Microsoft company, the netscape browser of Netscape company, the Firefox browser of Mozilla company, the Chrome browser of Google company, the company of roaming, the opera browser of Opera company, 360 browsers of 360 companies, the search dog browser of Sohu.com Inc., the TT of the Tengxun browser of company of Tengxun etc.
In another preferred embodiment; Can be used to provide the result of Search Results that equipment is provided with above-mentioned; Combine with existing browser plug-in; Constitute a kind of new browser plug-in, existing browser plug-in can be for example 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 that is used to provide Search Results that provides equipment to realize by the result according to a further aspect of the present invention; Particularly, the result provides equipment in step s1, obtains the corresponding initial search result of search sequence with user's input; In step s2, utilize first order models, in said initial search result, filter out preferred Search Results; In step s3, utilize second order models, in said preferred Search Results, filter out the optimum search result; In step s4, said optimum search result is offered said user.Wherein, the result provides equipment not only can work alone, and can also be integrated in the network equipment, subscriber equipment or the network equipment and subscriber equipment through the mutually integrated equipment that constitutes of network.Wherein, the said network equipment its include but not limited to the cloud that computing machine, network host, single network server, a plurality of webserver collection or a plurality of server constitute; At this, cloud is by constituting based on the great amount of calculation machine of cloud computing (Cloud Computing) or the webserver, and wherein, cloud computing is a kind of of Distributed Calculation, a virtual supercomputer of being made up of the loosely-coupled computing machine collection of a group.It includes but not limited to any electronic product that can carry out man-machine interaction through keyboard, telepilot, touch pad or voice-operated device with the user, for example computing machine, smart mobile phone, PDA, game machine or IPTV etc. said subscriber equipment.Said network includes but not limited to internet, wide area network, Metropolitan Area Network (MAN), LAN, VPN network, wireless self-organization network (Ad Hoc network) etc.Those skilled in the art will be understood that other result provides equipment to be equally applicable to the present invention, also should be included in the protection domain of the present invention, and be contained in this at this with way of reference.
Wherein, in step s1, the result provides equipment to obtain the corresponding initial search result of importing with the user of search sequence.Particularly, in step s1, the result provides equipment for example to pass through page technology; Like page technology such as JSP, ASP, PHP, perhaps, the application programming interfaces that equipment provided (API) of said search sequence can be provided through invoke user equipment or other; Or the communication mode of other agreements such as http, https, carry out alternately with the user, obtain the search sequence of user's input; And through carrying out participle such as the search sequence to user input, and the mode of in Query Database, searching for to the search sequence behind the participle, obtain the corresponding initial search result of search sequence with user's input; Wherein, The user can carry out alternately through equipment is provided such as keyboard, touch-screen, speech input device and result, imports the search sequence that it hopes inquiry, thereby initiates search; Perhaps; The result provides equipment to pass through based on various communication protocols (Communications Protocol); The transportation protocol that refers to compunication in this " communication protocol "; As: TCP/IP, UDP, FTP, ICMP, NetBEUI etc. also comprise other forms of communication that are present in the computing machine, for example: the communication between the object of OOP the inside simultaneously; Message transmission protocol in the operating system between distinct program or the computing machine disparate modules can provide the equipment of said initial search result with other, like search engine, carries out alternately to obtain the corresponding initial search result of importing with the user of search sequence.Preferably, in step s 1, the result provides the equipment can also be at the Search Results of intercepting some in that obtained and the corresponding Search Results of search sequence user input, with as said initial search result.For example; The user provides equipment to propose the search sequence of " the niciest Sichuan cuisine " through page technology to the result; The result provides equipment that " the niciest Sichuan cuisine " carried out participle, in database, respectively " the niciest " and " Sichuan cuisine " is retrieved, and has obtained 1000 initial search results.It will be understood by those skilled in the art that above-mentioned mode and several kinds of communication transport protocols of obtaining initial search result are merely for example; Other the existing or modes of obtaining initial search result that possibly occur from now on or communication transport protocols are as applicable to the present invention; Also should be included in the protection domain of the present invention, and be contained in this with way of reference at this.
In step s2, the result provides equipment utilization first order models, in said initial search result, filters out preferred Search Results.Particularly, in step s2, the result provides equipment in its said initial search result that in step s1, is provided, and utilizes said first order models, calculates the priority or the sequencing information of each initial search result; Again according to these priority or sequencing information; Said initial search result is screened; To obtain said preferred Search Results; As priority or sequencing information being satisfied initial search result that certain threshold value requires as preferred Search Results, perhaps with these initial search results by its priority or sequencing information descending sort, and will come the top n initial search result as preferred Search Results.Wherein, include but not limited to sort algorithm, sequencing feature vector etc. in said first order models.
At this; Utilize the priority of first order models calculating initial search result or the mode of sequencing information to include but not limited to: to utilize first order models; The proper vector that for example comprises one or more characteristic components and weight thereof; Confirm the assignment of pairing each characteristic component of initial search result; Thereby obtain the pairing proper vector of this initial search result, promptly this proper vector comprises the assignment and the weight thereof of these characteristic components, with priority or the sequencing information as this initial search result; Preferably, can also come weighting to confirm the assignment of this proper vector according to the assignment and the weight thereof of each included characteristic component of this proper vector, with priority or sequencing information as this initial search result.
At this; The mode that initial search result is arranged by its priority or sequencing information includes but not limited to: be not general; Priority or the sequencing information that can suppose initial search result comprise and the corresponding proper vector of this initial search result; This proper vector comprises the assignment and the weight thereof of one or more characteristic components, can confirm corresponding search result's ordering according to the assignment of each proper vector size of (as being confirmed by the assignment weighting of its characteristic component); Perhaps; Weight and assignment size (like the dictionary ordering) thereof according to each characteristic component of each proper vector; Confirm corresponding search result's ordering, for example at first sort, then for the identical initial search result of assignment of the highest characteristic component of its weight by the assignment of the highest characteristic component of weight; Can sort by the assignment of time high characteristic component of weight, until the ordering of accomplishing all initial search results.
For example; The search sequence that the user provides equipment to propose " the niciest Sichuan cuisine " through page technology to the result, result provide equipment in step s1, to obtain 1000 initial search results, and the initial search result that first order models is defined as the user inquiring sequence carries out based on each two component of 50% of weight; The authoritative analysis and semantic analysis; Sort, the result provides equipment according to 1000 said initial search results that in step s1, obtain, and utilizes authoritative method with semantic ordering that said 1000 are screened; Obtain preceding 100 results semantic and authoritative ordering, as said preferred Search Results.
In step s3, the result provides equipment utilization second order models, in said preferred Search Results, filters out the optimum search result.Particularly, in step s3, the result provides equipment in the said preferred Search Results that step s2 is provided, and utilizes said second order models that said user's search sequence is carried out further Screening Treatment, to obtain said optimum search result.Those skilled in the art will be understood that except the difference of second order models and first order models, the implementation of step s3 is identical with step s2 or similar basically, so for simplicity, repeat no more, and only are contained in this by reference.For example; The user provides equipment to propose the search sequence of " the niciest Sichuan cuisine " through page technology to the result; The result provides equipment that 1000 initial search results are provided in step s1; The result provides equipment that 100 preferred Search Results are provided in step s2; Second order models is defined as that the preferred result of user inquiring sequence is carried out based on weight respectively is four components of 25%: user requirements analysis, the analysis of user behavior statistics, authoritative analyze and semantic analysis is sorted, and in step s3, the result provides equipment to obtain 100 preferred Search Results that the result provides equipment in step s2, to provide; And utilize second order models; Said 100 preferred Search Results are carried out the integrated ordered of 4 components, obtain further ranking results, and preceding 10 result that will sort is as said optimum search result.
In step s4, the result provides equipment that said optimum search result is offered said user.Particularly; In step s4; The result provides equipment to obtain the optimum search result that the result provides equipment in step s3, to be filtered out; And utilization and user carry out alternately, perhaps according to the call format of the communication mode of application programming interfaces (API) that subscriber equipment provided or other agreements such as http, https, said optimum search result offered said user.For example; 10 said optimum search results that provide equipment in step s3, to be obtained the result present to the user as user search result's homepage; Perhaps 10 that are obtained among the step s3 said optimum search results are presented to the user as user search result's homepage, and will remain preferred Search Results and after homepage, present to the user as required successively.At this, those skilled in the art will be understood that the demonstration for initial search result, preferred Search Results or optimum search result, can carry out successively according to optimum search result, preferred Search Results, initial search result; Also can show from optimum search result, preferred Search Results, the optional one of which of initial search result; Also can be with both or triplicity in optimum search result, preferred Search Results, the initial search result, show 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 steps that are used for middle Search Results is screened and even multistage screening step more; Thereby initial search result is carried out multi-stage sequencing; For example adopt the multi-stage sequencing model that initial search result is sorted step by step, to obtain the user's of giving to be supplied optimum search result.Preferably; The present invention can also confirm the rank of needed order models according to different application requirements; And according to the multi-stage sequencing model of appropriate level; Like secondary order models, three grades or multi-stage sequencing model more, initial search result is sorted step by step, to obtain the user's of giving to be supplied optimum search result.
Preferably, in step s2, the result provides equipment can also utilize said first order models, confirms the priority of said initial search result; First amount threshold according to predetermined based on the priority of said initial search result, filters out said preferred Search Results from said initial search result, wherein, the quantity of said preferred Search Results satisfies said first amount threshold.Particularly, in step s2, the result provides equipment utilization said first order models; Confirm the priority of said initial search result; The for example pairing proper vector of this initial search result or its assignment are again according to the first predetermined amount threshold, based on the priority of these initial search results; From said initial search result, filter out said preferred Search Results; Wherein, the quantity of said preferred Search Results satisfies said first amount threshold, for example from these initial search results, directly filters out the higher initial search result of the priority of some as said preferred Search Results; Perhaps earlier these initial search results are carried out descending sort by its priority, then with the initial search result of the some of standing out as said preferred Search Results.For example; Suppose that first amount threshold is 100, the search sequence that the user provides equipment to propose " the niciest Sichuan cuisine " through page technology to the result, result provide equipment that 1000 initial search results are provided in step s1; The initial search result that first order models is defined as the user inquiring sequence carries out based on each two component of 50% of weight; The authoritative analysis and semantic analysis sorted, in step s2; The result provides equipment to obtain its 1000 said initial search results that in step s1, provided; And utilize authoritative method to confirm the priority of these 1000 initial search results, and, in the sequence of these 1000 initial search results, screen by its priority descending sort according to first amount threshold 100 with semantic ordering; Obtain preceding 100 results semantic and authoritative ordering, as said preferred Search Results.
More preferably, this embodiment also comprises step s7 (not shown), and wherein, in step s7, the result provides equipment to confirm rule according to the first predetermined quantity, confirms said first amount threshold; Wherein, said first quantity confirm rule comprise following at least each: based on the quantity of said initial search result, confirm said first amount threshold; The amount threshold that is used for definite said optimum search result based on predetermined is established rules then really, confirms said first amount threshold.Particularly, in step s7, the result provide equipment according to said initial search result quantity or predetermined be used for confirming that said optimum search result's amount threshold establishes rules then really, come dynamically to confirm said first amount threshold.For example, first quantity is confirmed to be provided with in the rule said first amount threshold and is directly proportional with the quantity of said initial search result, and then said initial search result quantity is many more, and said first amount threshold is big more; Perhaps; The amount threshold that first quantity confirms to be provided with in the rule said first amount threshold and the said optimum search result then positive correlation of establishing rules really; Like restriction according to interface display; Optimum search result's amount threshold limits and fixes, and then first amount threshold is directly proportional with said fixing optimum search result's amount threshold, for example represents quantity according to every page and confirms that first amount threshold is every page of integral multiple that represents quantity; Perhaps, confirm and the quantity of said initial search result and predetermined being used for of said first amount threshold confirmed that said optimum search result's amount threshold is established rules really and then formed the positive feedback relation, finally reach balance, thereby confirm said first amount threshold.
Preferably, in step s3, the result provides equipment can also utilize said second order models, confirms the priority of said preferred Search Results; Second amount threshold according to predetermined based on the priority of said preferred Search Results, filters out said optimum search result from said preferred Search Results, wherein, said optimum search result's quantity satisfies said second amount threshold.Particularly, in step s3, the result provides equipment utilization said second order models; Confirm the priority of said preferred Search Results; The for example pairing proper vector of this preferred Search Results or its assignment are again according to the second predetermined amount threshold, based on the priority of these preferred Search Results; From said preferred Search Results, filter out said optimum search result; Wherein, said optimum search result's quantity satisfies said second amount threshold, for example from these preferred Search Results, directly filters out the higher preferred Search Results of the priority of some as said optimum search result; Perhaps earlier these preferred Search Results are carried out descending sort by its priority, then with the preferred Search Results of the some of standing out as said optimum search result.
More preferably, this embodiment also comprises step s8 (not shown), and wherein, in step s8, the result provides equipment to confirm rule according to the second predetermined quantity, confirms said second amount threshold; Wherein, said second quantity confirm rule comprise following at least each: based on the terminal attribute of said user's subscriber equipment, confirm said second amount threshold; Based on the type information of said search sequence, confirm said second amount threshold; Based on the quantity of said preferred Search Results, confirm said second amount threshold.Particularly, in step s8, the result provides equipment to confirm rule according to the said second predetermined quantity; Terminal attribute based on said user's subscriber equipment; Or based on the type information of said search sequence, or, confirm said second amount threshold based on the quantity of said preferred Search Results.For example, different according to the terminal attribute of subscriber equipment, the also corresponding difference of second amount threshold; Relatively large like PC end display screen, every page can present 10 results, and then second amount threshold is 10; The display screen of mobile device is less relatively, and then second amount threshold is 6; Or it is different according to the type information of said search sequence; The also corresponding difference of second amount threshold; Type like search sequence is a train number information, demonstrates then the most accurately that few results gets final product, and second amount threshold is less relatively; Type like search sequence is a catering information, then demonstrates the demand that more relatively result just possibly satisfy the user; Or be directly proportional with the quantity of said preferred Search Results, confirm said second amount threshold.
Fig. 4 illustrates a kind of method flow diagram that is used to provide Search Results that provides equipment to realize by the result in accordance with a preferred embodiment of the present invention; Particularly, in step s5 ', the result provides equipment according to the first Search Results training data through the mark ordering; Through the mode of machine learning, confirm order models, wherein; Said order models comprise following at least each: said first order models, said second order models; In step s1 ', the result provides equipment to obtain the corresponding initial search result of importing with the user of search sequence; In step s2 ', the result provides equipment utilization first order models, in said initial search result, filters out preferred Search Results; In step s3 ', the result provides equipment utilization second order models, in said preferred Search Results, filters out the optimum search result; In step s4 ', the result provides equipment that said optimum search result is offered said user.Wherein, the result provides step s1 ', step s2 ', step s3 ' and the step s4 ' in the equipment identical or basic identical with corresponding step shown in Figure 3 respectively, so locate to repeat no more, and mode by reference is contained in this.
Between above-mentioned each step constant work; At this; It will be understood by those skilled in the art that " continuing " is meant that above-mentioned each step requires to carry out screening, the optimum search result's of the obtaining of the confirming of order models, initial search result, preferred Search Results the providing etc. of screening and optimum search result respectively according to mode of operation that set or adjustment in real time, provide equipment to stop to obtain the corresponding initial search result of search sequence with user's input until the result.
In step s5 ', the result provides equipment according to the first Search Results training data through the mark ordering, through the mode of machine learning; Confirm order models; Wherein, said order models comprise following at least each: said first order models, said second order models.Particularly; Step s5 ' is according to marking the first Search Results training data that ordering is accomplished; From initial order models or the initial order models of optional characteristic component conduct, and according to the parameter in demand utilization linear model, nonlinear model or the continuous adjustment of its combination order models, through the mode of machine learning; Confirm order models, like said first order models or said second order models.Wherein, The said first Search Results training data includes but not limited to query string (query) and corresponding search result (url); Every query string is indicated and relevance of search results grade numeral, perhaps a plurality of Search Results that wherein have correlativity just to distinguish are indicated the height relation etc. of said a plurality of Search Results.Wherein, if said order models is a linear model, include but not limited in the then said order models that each characteristic component reaches and the pairing weights of said characteristic component; If said order models is a nonlinear model; Can comprise in the then said order models like the decision-making value corresponding with some threshold point of characteristic component; Whole order models is made up of some decision-making values; For example the decision-making value by a plurality of characteristic components constitutes a decision tree, constitutes order models by many decision trees then, Search Results is comprehensively given a mark being used for.For example the first Search Results training data constantly is updated to current order models; The middle order models that obtains like initial order models or study; Calculate the sequencing information of this training data; For example a plurality of weighted sums that have the characteristic component of weight information; Perhaps through constituting or many score values that decision tree marking obtains by a plurality of characteristic components that have decision-making value, and according to the difference of this sequencing information and its sequencing information that has marked, preceding order models is deserved in adjustment; As increasing and decreasing characteristic component that deserves preceding order models or weight information or the decision-making value of adjusting its characteristic component, for example adjust or adjust simultaneously the weight information or the decision-making value of a plurality of characteristic components in order.Those skilled in the art will be understood that at this, confirm order models through the machine learning mode, make that not only the error of order models on the first Search Results training data is as far as possible little, also have certain extensive popularization ability.
Preferably, include but not limited to search sequence in the said first Search Results training data, Search Results, and the mapping relations between search sequence and the Search Results, priority, the ordering under corresponding search sequence maybe must grade like this Search Results; Preferably, search sequence and the mapping relations between the Search Results also comprise the weight or the decision-making value of the characteristic component of this Search Results under corresponding search sequence.For example; For given model prototype, like a proper vector that comprises a plurality of characteristic components, but the parameter of not demarcating each characteristic component as yet; Weights or decision-making value like this characteristic component; Then utilize to comprise the training set that marks the mapping relations between its search sequence and the Search Results,, confirm to comprise the model parameter of each characteristic component corresponding parameters through machine learning algorithms such as genetic algorithm, neural network, decision tree, SVMs; Promptly obtain order models, like said first order models or said second order models.
Preferably, this embodiment also comprises step s6 ', wherein; In step s6 '; The result provides equipment according to the second Search Results training data through the mark ordering, through the mode of machine learning, confirms one or more ordering submodels that are used for confirming said order models characteristic component.Those skilled in the art will be understood that the implementation of step s6 ' is identical with step s5 ' or similar basically, so for simplicity, repeats no more, and only is contained in this by reference except the difference of ordering submodel and order models.Wherein, include but not limited to search sequence in the said second Search Results training data, Search Results, and the mapping relations between search sequence and the Search Results, priority, the ordering under corresponding search sequence maybe must grade like this Search Results; Preferably, search sequence and the mapping relations between the Search Results also comprise the weight or the decision-making value of the characteristic component of this Search Results under corresponding search sequence.For example, for given submodel prototype, like a proper vector that comprises a plurality of characteristic components; But do not demarcate the parameter of each characteristic component as yet; Like the weights or the decision-making value of this characteristic component, then utilize to comprise the training set that marks the mapping relations between its search sequence and the Search Results, through machine learning algorithms such as genetic algorithm, neural network, decision tree, SVMs; Confirm to comprise the model parameter of each characteristic component corresponding parameters, promptly obtain sub-order models.
Preferably, in step s5 ', the result provides equipment through the mode of machine learning, to confirm said second order models according to the first Search Results training data through the mark ordering, and wherein, said second order models comprises the user behavior characteristic component.Particularly; The address information that the user behavior characteristic information includes but not limited to temporal information that the user searches for, confirm according to IP address, user through click, touch, swipe, the page residence time etc. generated about the operation information of Search Results etc.; The result provides equipment can utilize the first Search Results training data through the mark ordering; Utilize machine learning algorithms such as genetic algorithm, neural network; Said user behavior characteristic information is carried out machine learning, confirm the weight or the decision-making value of each characteristic component under the different user behavior characteristic information, promptly obtain second order models.At this; The user behavior characteristic information can be contained in the said first Search Results training data; Also can be stored in via network and result provides in third party's equipment such as search engine that equipment is connected or search log database, and from these third party devices, obtains said user behavior characteristic information through the application programming interfaces (API) that these third party devices provided.
To those skilled in the art, obviously the invention is not restricted to the details of above-mentioned example embodiment, and under the situation that does not deviate from spirit of the present invention or essential characteristic, can realize the present invention with other concrete form.Therefore; No matter from which point; All should regard embodiment as exemplary; And be nonrestrictive, scope of the present invention is limited accompanying claims rather than above-mentioned explanation, therefore is intended to the implication of the equivalents that drops on claim and all changes in the scope are included in the present invention.Should any Reference numeral in the claim be regarded as limit related claim.In addition, obviously other unit or step do not got rid of in " comprising " speech, and odd number is not got rid of plural number.A plurality of unit of stating in the device claim or device also can be realized through software or hardware by a unit or device.The first, the second word such as grade is used for representing title, and does not represent any specific order.
Claims (20)
1. one kind by the computer implemented method that is used to provide Search Results, and wherein, this method may further comprise the steps:
A obtains the corresponding initial search result of search sequence with user's input;
B utilizes first order models, in said initial search result, filters out preferred Search Results;
C utilizes second order models, in said preferred Search Results, filters out the optimum search result;
D offers said user with said optimum search result.
2. method according to claim 1, wherein, this method also comprises:
X through the mode of machine learning, confirms order models according to the first Search Results training data through the mark ordering;
Wherein, said order models comprise following at least each:
-said first order models;
-said second order models.
3. method according to claim 2, wherein, this method also comprises:
-according to the second Search Results training data,, confirm one or more ordering submodels that are used for confirming said order models characteristic component through the mode of machine learning through the mark ordering.
4. method according to claim 2, wherein, said step x comprises:
-according to the first Search Results training data,, confirm said second order models through the mode of machine learning through the mark ordering, wherein, said second order models comprises the user behavior characteristic component.
5. according to each described method in the claim 1 to 4, wherein, said step b comprises:
-utilize said first order models, confirm the priority of said initial search result;
-according to the first predetermined amount threshold,, from said initial search result, filter out said preferred Search Results based on the priority of said initial search result, wherein, the quantity of said preferred Search Results satisfies said first amount threshold.
6. method according to claim 5, wherein, this method also comprises:
-confirm rule according to the first predetermined quantity, confirm said first amount threshold;
Wherein, said first quantity confirm rule comprise following at least each:
-based on the quantity of said initial search result, confirm said first amount threshold;
-establish rules really then based on the predetermined amount threshold that is used for definite said optimum search result, confirm said first amount threshold.
7. according to each described method in the claim 1 to 6, wherein, said step c comprises:
-utilize said second order models, confirm the priority of said preferred Search Results;
-according to the second predetermined amount threshold,, from said preferred Search Results, filter out the optimum search result based on the priority of said preferred Search Results, wherein, said optimum search result's quantity satisfies said second amount threshold.
8. method according to claim 7, wherein, this method also comprises:
-confirm rule according to the second predetermined quantity, confirm said second amount threshold;
Wherein, said second quantity confirm rule comprise following at least each:
-based on the terminal attribute of said user's subscriber equipment, confirm said second amount threshold;
-based on the type information of said search sequence, confirm said second amount threshold;
-based on the quantity of said preferred Search Results, confirm said second amount threshold.
9. one kind is used to provide the result of Search Results that equipment is provided, and wherein, this equipment comprises:
Deriving means as a result is used to obtain the corresponding initial search result of search sequence with user's input;
First screening plant is used to utilize first order models, in said initial search result, filters out preferred Search Results;
Second screening plant is used to utilize second order models, in said preferred Search Results, filters out the optimum search result;
Generator is used for said optimum search result is offered said user as a result.
10. result according to claim 9 provides equipment, and wherein, this equipment also comprises:
Model is confirmed device, is used for through the mode of machine learning, confirming order models according to the first Search Results training data through the mark ordering;
Wherein, said order models comprise following at least each:
-said first order models;
-said second order models.
11. result according to claim 10 provides equipment, wherein, this equipment also comprises:
Submodel is confirmed device, is used for according to the second Search Results training data through the mark ordering, through the mode of machine learning, confirms one or more ordering submodels that are used for confirming said order models characteristic component.
12. result according to claim 10 provides equipment, wherein, said model confirms that device is used for:
-according to the first Search Results training data,, confirm said second order models through the mode of machine learning through the mark ordering, wherein, said second order models comprises the user behavior characteristic component.
13. according to each described result in the claim 9 to 12 equipment is provided, wherein, said first screening plant is used for:
-utilize said first order models, confirm the priority of said initial search result;
-according to the first predetermined amount threshold,, from said initial search result, filter out said preferred Search Results based on the priority of said initial search result, wherein, the quantity of said preferred Search Results satisfies said first amount threshold.
14. result according to claim 13 provides equipment, wherein, this equipment also comprises:
First threshold is confirmed device, is used for confirming rule according to the first predetermined quantity, confirms said first amount threshold;
Wherein, said first quantity confirm rule comprise following at least each:
-based on the quantity of said initial search result, confirm said first amount threshold;
-establish rules really then based on the predetermined amount threshold that is used for definite said optimum search result, confirm said first amount threshold.
15. according to each described result in the claim 9 to 14 equipment is provided, wherein, said second screening plant is used for:
-utilize said second order models, confirm the priority of said preferred Search Results;
-according to the second predetermined amount threshold,, from said preferred Search Results, filter out the optimum search result based on the priority of said preferred Search Results, wherein, said optimum search result's quantity satisfies said second amount threshold.
16. result according to claim 15 provides equipment, wherein, this equipment also comprises:
Second threshold value is confirmed device, is used for confirming rule according to the second predetermined quantity, confirms said second amount threshold;
Wherein, said second quantity confirm rule comprise following at least each:
-based on the terminal attribute of said user's subscriber equipment, confirm said second amount threshold;
-based on the type information of said search sequence, confirm said second amount threshold;
-based on the quantity of said preferred Search Results, confirm said second amount threshold.
17. a search engine comprises as each describedly is used to provide the result of Search Results that equipment is provided in the claim 9 to 16.
18. a search engine plug-in unit comprises as each describedly is used to provide the result of Search Results that equipment is provided in the claim 9 to 16.
19. a browser comprises as each describedly is used to provide the result of Search Results that equipment is provided in the claim 9 to 16.
20. a browser plug-in comprises as each describedly is used to provide the result of Search Results that equipment is provided in the claim 9 to 16.
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 true CN102810117A (en) | 2012-12-05 |
| CN102810117B 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) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103164536A (en) * | 2013-04-11 | 2013-06-19 | 苏州阔地网络科技有限公司 | Method and system for achieving cloud education platform data search |
| CN103164535A (en) * | 2013-04-11 | 2013-06-19 | 苏州阔地网络科技有限公司 | Realization method and system for searching data of Cloud education platform |
| 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 |
| CN104598611A (en) * | 2015-01-29 | 2015-05-06 | 百度在线网络技术（北京）有限公司 | Method and system for sequencing search entries |
| CN105701155A (en) * | 2015-12-30 | 2016-06-22 | 百度在线网络技术（北京）有限公司 | Information push method and the device |
| CN106055567A (en) * | 2015-04-09 | 2016-10-26 | 谷歌公司 | Providing app store search results |
| CN106897289A (en) * | 2015-12-18 | 2017-06-27 | 北京奇虎科技有限公司 | The optimization method and device of information search |
| 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 |
| 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 |
| 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 |
| CN107463704A (en) * | 2017-08-16 | 2017-12-12 | 北京百度网讯科技有限公司 | Searching method and device based on artificial intelligence |
| CN107977405A (en) * | 2017-11-16 | 2018-05-01 | 北京三快在线科技有限公司 | Data reordering method, data sorting device, electronic equipment and readable storage medium storing program for executing |
| CN108829757A (en) * | 2018-05-28 | 2018-11-16 | 广州麦优网络科技有限公司 | A kind of intelligent Service method, server and the storage medium of chat robots |
| CN109033140A (en) * | 2018-06-08 | 2018-12-18 | 北京百度网讯科技有限公司 | A kind of method, apparatus, equipment and the computer storage medium of 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 |
| CN109436834A (en) * | 2018-09-25 | 2019-03-08 | 北京金茂绿建科技有限公司 | A kind of method and device for choosing funnel |
| 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 |
| CN110413357A (en) * | 2013-06-08 | 2019-11-05 | 苹果公司 | For synchronizing the equipment, method and graphic user interface of two or more displays |
| CN110674401A (en) * | 2019-09-19 | 2020-01-10 | 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 |
| 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 |
| 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 |
| 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 |
| CN111209378A (en) * | 2019-12-26 | 2020-05-29 | 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 weight |
| 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 |
| CN113656574A (en) * | 2021-10-19 | 2021-11-16 | 北京欧应信息技术有限公司 | Method, computing device and storage medium for search result ranking |
| CN113761084A (en) * | 2020-06-03 | 2021-12-07 | 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 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090319507A1 (en) * | 2008-06-19 | 2009-12-24 | 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 |
| 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 |
-
2012
- 2012-06-29 CN CN201210226803.XA patent/CN102810117B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090319507A1 (en) * | 2008-06-19 | 2009-12-24 | Yahoo! Inc. | Methods and apparatuses for adapting a ranking function of a search engine for use with a specific domain |
| 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 |
| CN101770521A (en) * | 2010-03-11 | 2010-07-07 | 东华大学 | Focusing relevancy ordering method for vertical search engine |
| CN102043846A (en) * | 2010-12-16 | 2011-05-04 | 上海电机学院 | Search method and device based on genetic algorithm |
Cited By (41)
| 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 |
| CN103164535A (en) * | 2013-04-11 | 2013-06-19 | 苏州阔地网络科技有限公司 | Realization method and system for searching data of Cloud education platform |
| CN103164536A (en) * | 2013-04-11 | 2013-06-19 | 苏州阔地网络科技有限公司 | Method and system for achieving cloud education platform data search |
| 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 |
| US11692840B2 (en) | 2013-06-08 | 2023-07-04 | Apple Inc. | Device, method, and graphical user interface for synchronizing two or more displays |
| CN110413357A (en) * | 2013-06-08 | 2019-11-05 | 苹果公司 | For synchronizing the equipment, method and graphic user interface of 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 |
| CN104598611A (en) * | 2015-01-29 | 2015-05-06 | 百度在线网络技术（北京）有限公司 | Method and system for sequencing search entries |
| 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 |
| CN106055567A (en) * | 2015-04-09 | 2016-10-26 | 谷歌公司 | Providing app store search results |
| CN106897289A (en) * | 2015-12-18 | 2017-06-27 | 北京奇虎科技有限公司 | The optimization method and device of information search |
| CN106897289B (en) * | 2015-12-18 | 2020-07-10 | 北京奇虎科技有限公司 | Information search optimization method and device |
| CN105701155A (en) * | 2015-12-30 | 2016-06-22 | 百度在线网络技术（北京）有限公司 | Information push method and the 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 |
| US10657325B2 (en) | 2017-06-21 | 2020-05-19 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method for parsing query based on artificial intelligence and computer device |
| 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 |
| CN107463704A (en) * | 2017-08-16 | 2017-12-12 | 北京百度网讯科技有限公司 | Searching 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 |
| CN107977405A (en) * | 2017-11-16 | 2018-05-01 | 北京三快在线科技有限公司 | Data reordering method, data sorting device, electronic equipment and readable storage medium storing program for executing |
| 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 |
| CN108829757A (en) * | 2018-05-28 | 2018-11-16 | 广州麦优网络科技有限公司 | A kind of intelligent Service method, server and the storage medium of chat robots |
| 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 |
| US10896188B2 (en) | 2018-06-08 | 2021-01-19 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for determining search results, device and computer storage medium |
| CN109033140A (en) * | 2018-06-08 | 2018-12-18 | 北京百度网讯科技有限公司 | A kind of method, apparatus, equipment and the computer storage medium of 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 |
| CN109436834A (en) * | 2018-09-25 | 2019-03-08 | 北京金茂绿建科技有限公司 | A kind of method and device for choosing 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 |
| CN110674401A (en) * | 2019-09-19 | 2020-01-10 | 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 |
| 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 |
| CN111209378A (en) * | 2019-12-26 | 2020-05-29 | 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 weight |
| 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 |
| CN113761084A (en) * | 2020-06-03 | 2021-12-07 | 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 |
| 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 |
| CN113656574A (en) * | 2021-10-19 | 2021-11-16 | 北京欧应信息技术有限公司 | Method, computing device and storage medium for search result ranking |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102810117B (en) | 2016-02-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102810117A (en) | Method and equipment for supplying search result | |
| CN105159962B (en) | Position recommends method and apparatus, resume to recommend method and apparatus, recruitment platform | |
| CN103593373A (en) | Search result sorting method and search result sorting device | |
| CN107491547A (en) | Searching method and device based on artificial intelligence | |
| US9928282B2 (en) | Method and apparatus for providing search results on a mobile terminal | |
| CN102346778B (en) | Method and equipment for providing searching result | |
| CN107463704A (en) | Searching method and device based on artificial intelligence | |
| CN107683471A (en) | Webpage of the index with deep linking | |
| CN102968454A (en) | Method and equipment for obtaining search results of popularization object | |
| CN103226393A (en) | Input method and equipment | |
| CN102236710A (en) | Method and equipment for displaying news information in query result | |
| CN107145496A (en) | The method for being matched image with content item based on keyword | |
| CN103942712A (en) | Product similarity based e-commerce recommendation system and method thereof | |
| CN105224959A (en) | The training method of order models and device | |
| CN102682099A (en) | Method and device for providing target pictures in target areas | |
| CN103076893A (en) | Method and equipment for realizing voice input | |
| CN104992347A (en) | Video matching advertisement method and device | |
| CN102945263A (en) | Method for determining access correlation information among multiple access objects | |
| CN102945273B (en) | A kind of for providing the method and apparatus of Search Results | |
| CN102323937A (en) | Method and equipment for providing search result | |
| CN111194457A (en) | Patent evaluation determination method, patent evaluation determination device, and patent evaluation determination program | |
| CN102541946B (en) | Method and equipment for determining recommendation degree of hyperlink based on recommendation attribute of hyperlink | |
| US20170235835A1 (en) | Information identification and extraction | |
| CN102567417A (en) | Analysis equipment and method for determining reliability of anchor text of hyperlink | |
| CN113918794A (en) | Enterprise network public opinion benefit analysis method, system, electronic equipment and storage medium |
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 |