CN101114294A - Self-help intelligent uprightness searching method - Google Patents
Self-help intelligent uprightness searching method Download PDFInfo
- Publication number
- CN101114294A CN101114294A CNA2007100709770A CN200710070977A CN101114294A CN 101114294 A CN101114294 A CN 101114294A CN A2007100709770 A CNA2007100709770 A CN A2007100709770A CN 200710070977 A CN200710070977 A CN 200710070977A CN 101114294 A CN101114294 A CN 101114294A
- Authority
- CN
- China
- Prior art keywords
- user
- search
- information
- users
- search results
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract 7
- 238000013179 statistical model Methods 0.000 claims abstract 5
- 230000011218 segmentation Effects 0.000 claims abstract 3
- 235000014510 cooky Nutrition 0.000 claims abstract 2
- 230000006399 behavior Effects 0.000 claims 1
- 238000011156 evaluation Methods 0.000 claims 1
- 238000010801 machine learning Methods 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract 1
- 235000019640 taste Nutrition 0.000 abstract 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to a self-service intelligent vertical search method and includes the following steps: cookies files of users, registered information, historical search information and ordered attention module can be utilized to study preferences of users, and the preferences of the users are set as statistical models of the users which are real-timely stored to a database of a search engine dynamically. A final key word/words collection file can be obtained by studying high speed Chinese word segmentation and search habits of users. The search engine can search all the information that is relevant to the inertial key sentence/word through a network database. Meanwhile, the search results are matched with the statistical models of the users and the search results which fit for the preferences of the users can be returned to the users. The present invention has the beneficial effects that the users can find out the needed information from the huge information collection and study using preferences and habits of the users actively, and then the search results can cater to ''tastes'' of the users more and the users make judgment to the values of the search results totally.
Description
Technical field
The present invention relates to a kind of search field that is applied to digital network, particularly a kind of self-help intelligent uprightness searching method.
Background technology
At present, the widespread use of search engine technique makes the user can obtain to wish the information of acquisition easily, easily.But existing search engine and search technique also exist shortcoming and defect:
1, the magnanimity information of network existence, the Search Results quantity tool that also makes the search engine utilize the keyword search technology return is big, wherein very many information are no-good for the user in fact, and the user has to concentrate the information of seeking their needs in these huge information.
2, existing search technique can not be learnt user's use preference and custom on one's own initiative, thereby makes the result of search conform with user " taste " more, but judges the value of Search Results to it by user oneself fully.
3, search engine of today can not reflect the evaluation of user to Search Results, the evaluation of this subjectivity can not be incorporated in the search engine, thereby revises the process of searching for, and makes Search Results more accurate.
4, most function executing is undertaken by server, existing universal search engine can not effectively utilize the digital terminal hardware resource of user side self, make that the pressure of server is overweight, thereby can not carry out function program efficiently, cause great amount of investment to purchase server hardware.
Summary of the invention
Purpose of the present invention is just in order to overcome above-mentioned shortcoming, and provide a kind of self-help intelligent uprightness searching method, in particular, be a kind of user's pro-active intervention search mechanisms, vertically deepen self-help intelligent searching method, system and the computer program of Search Results.
The present invention solves the technical scheme that its technical matters adopts.This self-help intelligent uprightness searching method comprises the steps:
1.1), utilize the concern module of user cookies file, log-on message, historical search information and the customization be stored in the client and server end to carry out user preference study, and this user preference be established as user's statistical model in real time, dynamically store in the search engine database;
1.2), utilize to close linking verses/dictionary the search statement of user's input carried out the high speed Chinese word segmentation, and generate a critical sentence/word set file, this document has comprised all relevant, similar critical sentence and keywords that carry out after the semantic analysis, user's historical search information is carried out statistical learning, draw the critical sentence/speech relevant, similar in user's search custom with these critical sentence/word sets, by high speed Chinese word segmentation and user search behavior learning, draw a final critical sentence/word set file;
1.3), search engine is by all information relevant with these inertia critical sentence/speech of network data library searching, simultaneously, these Search Results and user's statistical model are mated, its critical sentence/lexicon of search share the information of family preference in these Search Results, finally, the Search Results that will meet user preference returns to the user.
This method can reflect the user to the evaluation of Search Results, revise the process of machine learning in view of the above, and revises user's statistical model simultaneously.
The user can customize interested content and information, form Search Results and instant messaging in this method.
The effect that the present invention is useful is:
1, the user can concentrate the information that they need of seeking from huge information.
2, this method can be learnt user's use preference and custom on one's own initiative, thereby makes the result of search conform with user " taste " more, but judges the value of Search Results to it by user oneself fully.
3, can reflect the evaluation of user, also the evaluation of this subjectivity can be incorporated in the search engine, thereby revise the process of searching for, make Search Results more accurate Search Results.
4, the hardware resource of Xu Yaoing is more than universal search engines such as google, Baidu, can effectively utilize the hardware resource of user self digital terminal, thereby alleviates the pressure of server end greatly.
Description of drawings
The system chart that the inventive method is achieved among Fig. 1 embodiment;
Realize among Fig. 2 embodiment that the user of this method estimates the system flowchart of mechanism and systematic learning mechanism etc.;
Conjunctive word database data structural drawing among Fig. 3 embodiment;
Patent information data structure diagram among Fig. 4 embodiment;
Business opportunity information data structure figure among Fig. 5 embodiment;
Company information data structure diagram among Fig. 6 embodiment;
User's statistical model data structure diagram among Fig. 7 embodiment;
The user interface sectional drawing of the patent information search of Fig. 8 webpage;
The user interface sectional drawing of the business opportunity information search of Fig. 9 webpage;
The user interface sectional drawing of the company information search of Figure 10 webpage.
Embodiment
The invention will be described further below in conjunction with drawings and Examples:
In order to set forth the mechanism of the inventive method and system better, at first description below done in following noun:
User: the user who uses system of the present invention with the purpose of certain search information.
User intervention: mean the user by certain intervention program module,, come the deviation that occurs in the update the system machine learning, can not only improve the accuracy of machine learning, make also that simultaneously the result of search is more accurate as user's appraisement system.
Vertical search: as the letter explanation, vertical search refers to in-depth, the precision of Search Results, and this in-depth refers to the search purpose and the preference of more being close to the users, so, its search basis is user's statistical model and a whole network data base, but not the preceding once result of search, this just makes present more well-known search engine such as the inventive method and system and Baidu, google etc. that difference be arranged.
Concerning this description, accompanying drawing any or a plurality of in quote under the situation of step with same numeral or feature, these steps or feature have substantially the same function or operation.
Shown in Fig. 1 is the system chart of self-help intelligent uprightness searching method in the exemplary embodiment.This system comprises client I 100, and client II 110, digital network 120, external data source 140, server-side system 130, data-base recording 150 and application program 160.Below in conjunction with Fig. 1 various piece is described in detail.
Client I 100 and client II 110 are two kinds of multi-form client, and client and server-side system can think it all is a kind of client machine system on function is formed.
Client machine system: client machine system of the present invention can be realized by digital termination system, is used to carry out the application program of processing procedure of the present invention, but is not limited in this.Client machine system can be digital terminal or the terminal that is connected to digital terminal.Usually, in order to realize the method for the invention and system, the digital terminal of indication needs to comprise display device, audio frequency input and output device, user input unit, storer and CPU at least in the present invention, and be considered to carry out the application program that can realize the method for the invention and system, as network browser program Internet Explorer.
Be appreciated that ground, this client machine system is not limited in digital termination system, also can be other equipment such as mobile phone, and the person skilled in art should be able to understand this point at an easy rate.
Display device: can be a monitor, as the CRT and the LED of routine, or other any devices that are arranged to the display message content.
Audio frequency input-output unit: can be the device that earphone, microphone, microphone or the like input or output voice data in computing machine.Certainly, audio frequency input and output device can combine together, as has the earphone of microphone.
User input unit: can be keyboard, mouse or the like, input block can be equipped with cursor control key, as left Arrow, right key, upwarding key and down Arrow.Certainly, display device and user input unit can combine together, as touch-screen.
Storer: this storer can be understood as storage and carries out the application program that can realize the method for the invention by CPU, also can store document, for example conventional random access storage device (RAM).
CPU: this CPU can be the general processor unit, in order to the document in the reference-to storage, to search for, also can be an independent communication unit, and as modulator-demodular unit, the effect of this communication unit is to obtain document from the outside.
What client I 100. client I represented is that a kind of accesses network 120 communicates movable client composition mode with server-side system 130.The purpose that it communicates is to server-side system 130 request search information.Client I has comprised cohort 1 and the cohort 2 that connects by local network 103, and cohort 1 is two different client machine systems with cohort 2 equally.Cohort 1 is with in cohort 2 can be distributed in same or different local networks.It is client I that client machine system 101, client machine system 102 connect by local network 103.
Cohort: can be the set of uniting, also can be represented as an industry, as financial circles, manufacturing industry by individual, department, commodity, subsidiary company, affiliate or other modes.
Local network 103: comprise the LAN (Local Area Network) LAN that is limited in limited geographic area, and the wide area network WAN and the Metropolitan Area Network (MAN) MAN that are not subject to limited geographic area.
Client II 110: different with client I is that what client II represented is another kind of as a client form that can communicate activity by network 120 and server-side system 130.What client I represented is an independent client machine system 110.
Be appreciated that ground, in another embodiment, may comprise wherein a kind of or whole client forms of client I and client II, but the array configuration of client do not influence the realization of the method for the invention.
Digital network 120: the transmission network of wired or wireless digital network information or signal is used for the information of transmission of digital network.Can be understood as but be not limited only to LAN (Local Area Network) LAN, wide area network WAN, Metropolitan Area Network (MAN) MAN, virtual private network and the Internet.Client I and client II and other network terminal entities can be connected to server-side system 130 by the network of any form, but they not necessarily are connected on the server-side system 130 by same network.
Server-side system 130: server-side system realizes by one or more servers, can be wherein one or more server associatings of database server 131, the webserver 132, apps server 133, also can be to have comprised the wherein function of one or more servers in the server.
Server: be used to respond the computer program operation that is stored on the server.
Database server 131: all electronic information of stored data base record 150 and execution are to the visit of data-base recording 150.
Data-base recording 150: all users that storage is relevant with server-side system 130 or the various information contents and the data of client machine system, as related dictionary 151, Search Results 152, user's statistical model 153.These information contents and data comprise the field that data-base recording comprised of Fig. 3, Fig. 4-1, Fig. 4-2, Fig. 4-3 and exemplary embodiment illustrated in fig. 5.
Fig. 3 has illustrated an example of related dictionary 151 structures, and it has comprised a plurality of fields.Wherein similar local sentence word set 310 has been represented the set of all similar sentence/speech of certain critical sentence/speech, and these similar sentence/speech are stored in the database server 131.Similar outside sentence word set 320 has been represented the set of all similar sentence/speech of this critical sentence/speech, these similar sentence/speech are to be stored in the external data source 140 that is connected on the digital network 120, and server-side system 130 when needed can be by digital network 120 to external data source 140 these critical sentence/speech of request and store in the local database server 131.Relevant local sentence word set 330 has been represented the set of all relevant sentence/speech of this critical sentence/speech, and these relevant sentence/speech are stored in the database server 131.Relevant outside sentence/word set 340 has been represented the set of all relevant sentence/speech of this critical sentence/speech, these relevant sentence/speech are stored in the external data source 140, and server-side system 130 can be asked these critical sentence/speech and stores in the local database server 131 to external data source 140 when needed by digital network 120.The historical critical sentence word set 350 of similar user is the similar sentence/word sets about this critical sentence/speech of certain user that get by user's historical search result statistics, these critical sentence/speech have specific user characteristics, for certain user proprietary, server-side system 130 draws this word set after certain user's historical search result and evaluation information are added up, when this User login system was searched for, server-side system 130 was called in the related dictionary that this word set joins this critical sentence/speech automatically.The historical critical sentence word set 360 of relevant user is the relevant sentence word sets about this critical sentence/speech of certain user that get by user's historical search result statistics, to the historical critical sentence word set 350 of above-mentioned similar user similarly, when certain User login system was searched for, server-side system 130 was called in the related dictionary that this word set joins this critical sentence/speech automatically.
Similar: in the present invention " similar " refers to such a case, a critical sentence/speech has a lot of other different critical sentence/speech to be close in meaning with it, for example, the similar keyword of " computer " has " computing machine ", " computer ", " Pc machine " etc., and wherein " computing machine " may be the similar keyword that draws according to user's historical search result statistics.
Relevant: in the present invention " being correlated with " refers to such a case, critical sentence/speech has a lot of other different critical sentence/speech to have closely with it to get in touch, this contact has specific epoch and history feature, change that can be in the movement and changing, for example, the associative key of " computer " has " notebook ", " keyboard ", " mouse ", " USB flash disk ", " MP3 " etc., wherein " MP3 " associative key of may be exactly drawing according to user's historical search result statistics.
Fig. 4, Fig. 5, Fig. 6 are three examples of the data structure of Search Results 152.In one exemplary embodiment of the present invention, system is primarily aimed at the search of three contents: patent, business opportunity and company.
What wherein Fig. 4 showed is the formation of patent information data, the patent information data constitute 410 comprise that patent number 411, patent describe 412, patent summary 413, full patent texts 414, issuing time 415, inventor 416, patent type 417 and affiliated company numbering 418.What patent number 411 was represented is the unique number of patent information, and the system of being convenient to retrieves and calls.Patent is described 412 titles that are this patent.Patent summary 413 is for server-side system and user, play the effect of an interface in fact, it is static data, the same with other information of patent, be stored in statically in the database server, whether the user can understand this patent by the summary of patent relevant to its useful search purpose with him; On the other hand, server-side system also is by the critical sentence/speech in the patent summary 413 but not critical sentence/the speech in the full patent texts 414 is searched for, mated and calls, so, the purpose that this also feasible result who searches for more is close to the users; And the demonstration by the Search Results that this approach obtained, be not the same with google as Baidu yet, just show the full text selected parts that comprise keyword, but the description of patent and other information, have only the user input unit of working as such as mouse to move to patent and describe on 412, just can show patent summary 413.Full patent texts 414 refers to whole supporting papers of patent.Issuing time 415 is that this patent is issued the i.e. time of storage in this website, but not the announcement time of patent.Inventor 416 is inventors of this patent.Patent type number 417 is represented the type under this patent, is divided into utility model patent, appearance patent and patent of invention, and is related with patent type attribute epiphase.Affiliated company numbering 418 is meant the unique number of the company that has this patent.
Similarly, what Fig. 5 showed is the formation of business opportunity information data, and the business opportunity of indication of the present invention is the abbreviation of commercial opportunity, and by business opportunity, the user can find the mode with other companies or individual's cooperation.The business opportunity information data constitute 420 comprise that business opportunity numbering 421, business opportunity describe 422, business opportunity summary 423, business opportunity specify 424, business opportunity type number 425, effective time 426 and affiliated company numbering 427.What business opportunity numbering 421 was represented is the unique number of business opportunity information, and the system of being convenient to retrieves and calls.Business opportunity is described 422 titles that are this business opportunity.With the patent summary similarly, business opportunity summary 423 plays the effect of an interface to server-side system and user, it has comprised the contact method of product information, company information, supply-demand information and company.On the one hand, whether the user can relevant to its useful search purpose with him by business opportunity summary 423 these business opportunities of understanding; On the other hand, server-side system also is to search for, mate and call by the critical sentence/speech in the business opportunity summary 423.Have only the user input unit of working as such as mouse to move to business opportunity and describe on 422, just can show business opportunity summary 423.Business opportunity specifies 424 and refers to specifying of business opportunity.Business opportunity type number 425 is sorted out the classification under the business opportunity, as wants to buy and sell, and is related with business opportunity type attribute epiphase.Refer to the time that this business opportunity can produce value effective time 426, in case surpass this time bar, this business opportunity has probably just disappeared.Affiliated company numbering 427 is meant the unique number of the company that has this business opportunity.
What similarly, Fig. 6 showed is the formation of company information data.The company information data constitute 430 and comprise company's numbering 431, company describes 432, company information summary 433, company introduction 434, company's specifying information 435, company's type number 436, the establishment time 437, registered capital 438, registration date 439, headcount 440, annual turnover 441, enterprise's form of ownership numbering 442, outlet rate 443, the foreigner invests ratio 444, company's network address 445, the email446 of company, want to buy classification numbering 447, sell classification numbering 448, the contact person 449, company's telephone number 450, fax number 451, company contact address 452 and postcode 453.What company's numbering 431 was represented is the unique number of company, and the system of being convenient to retrieves and calls.432 titles that are the said firm are described by company.With the patent summary similarly, company information summary 433 plays the effect of an interface to server-side system and user.On the one hand, whether the user can relevant to its useful search purpose with him by company information summary 433 these business opportunities of understanding; On the other hand, server-side system also is to search for, mate and call by the critical sentence/speech in the company information summary 433.Have only the user input unit of working as such as the mouse company of moving to describe on 432, just can show company information summary 433.Company introduction 434 is parts of company information summary 433, in order to introduce company's situation simply.Company's specifying information 435 refers to the specifying information of company.Company's type number 436 is related with company type attribute epiphase, the classification under the company is sorted out, as type of production, trade type, service type, government and other mechanisms etc.The establishment time 437 refers to the company of declaring when the said firm carries out the industrial and commercial registration and the tax registration and sets up the time.Total assets when registered capital 438 refers to company's this industrial and commercial registration in when registration.Registration date 439 is dates that the said firm is registered as system user.Headcount 440 is headcount of the said firm.Annual turnover 441 is meant the business sales in 1 year.The forms of ownership of enterprise's form of ownership 442 expression enterprises are as state-run, private, limited liability system.Outlet rate 443 is meant that the exported product of company accounts for the ratio of company's output aggregate quantity.The foreigner invests ratio 444 and is meant that the foreigner accounts for the ratio of corporate assets total value in the investment of company.Company's network address 445 is websites of the said firm.The email446 of company is meant the e-mail address of company's outside contact, makes things convenient for the external world to carry out business consultation.Want to buy classification numbering 447 and be meant that the said firm wants to buy the classification of product, be associated as digital terminal periphery etc. and product category attribute list.Sell classification numbering 448 and be meant the said firm's product sold classification, be associated as digital equipment etc. and product category attribute list.Contact person 449 is personnel's titles of the said firm's outside contact.Company's telephone number 450 is telephone numbers of the said firm.Fax number 451 is fax numbers of the said firm.Company contact address 452 is meant the contact address of the said firm.Postcode 453 is postcodes of the said firm contact address.
The data that Fig. 7 illustrates user's statistical model constitute.The data that user's statistical model comprises have: Customs Assigned Number 510, user name 520, other log-on messages 530, historical search information 540, historical critical sentence word set 550, concern module 560 and client cookies file.Customs Assigned Number 510 representative be that this user profile is stored in the unique number in the database, be convenient to the renewal of 130 pairs of user models of server-side system and call.User name is that the user submits to voluntarily, as the authentication data of User login search system.Other log-on messages 530 are user's other information except user name when being registered as the search system registered user, as land password, affiliated industry, Business Name etc.Historical search information 540 is the search statement searched for after registering of user and the set of Search Results, and server-side system 130 utilizes historical search information 540 to form historical critical sentence word set 550, is the important basis that system carries out preferential learning.Historical critical sentence word set 550 is formed by historical search information 540, representative is in user's search custom, which the critical sentence/speech similar, relevant with certain critical sentence/speech that the user thinks be, the integrated sentence word set of these critical sentence/vocabulary, form the critical sentence word set of certain critical sentence/speech of this user-specific, critical sentence/the word set of the critical sentence/speech of user search is stored in this user's the statistical model, forms user's historical critical sentence word set 550.Paying close attention to module and be by what the user customized voluntarily and interestedly want the content of searching for, can be specific industry, as financial circles, service sector, also can be the information specific language, as English, Japanese, can also be specific geographic area, as continent, Hong Kong, Macao and Taiwan.Paying close attention to module 560 is the important evidence of user preference study equally.Client cookies file 570 is some info webs that are stored in client, as user name and network address, when the user does not have login system and searches for, client cookies file 570 is unique foundations of user preference study, and when the User login system is searched for thereafter, other data of client cookies file 570 and user's statistical model together, as the foundation of user preference study.
In order to understand the data structure shown in Fig. 7 better, below client cookies file 570 is done a more detailed explanation.Cookies also claims cookie.Cookies is that a kind of Website server that can allow is stored into the hard disk or the internal memory of client to low volume data, or from a kind of technology of the hard disk reading of data of client.Cookies is when certain user browses certain website, places a very little text on user's hard disk by the webserver, the information such as time of the user name that it can recording user, password, the webpage of browsing, stop.When the user came this website once more, user's relevant information was learnt by reading cookies in the website, just can make corresponding action, as show the poster of user welcome at the page, perhaps allowed the user need not input the just directly login or the like of user name, password.In an embodiment of the present invention, not separately client cookies file 570 unique data as user's statistical model, and as the Another reason of the foundation of user preference study be other data in client cookies file 570 and the user's statistical model 153 together, the situation that the deletion of the shared digital terminal of many people, temporary folder may occur, make the cookies file accurately not report situations even to lose, so, must introduce user's statistical model 153 to other data, make that the process of preferential learning is more accurate.
Return Fig. 1 below.
The webserver 132: communicate with the client as client I 100 and client II 110, as sending information, reception information, and carry out being associated of task to client I 100 and client II110.
Apps server 133: according to exemplary embodiment, the computer program of apps server storage, execution such as application program 160.
External data source 140: can adopt the one or more servers that are similar to server-side system 130 to realize, its effect is the available third party's information source outside the querying server end system 130, and utilize related information content that these information sources provide by application program 160 visit and carry out and generate related information and return to client I 100 and client II110.
Application program 160: in this explanation, one or more computer programs that can realize the method for the invention and system are referred to as application program, certainly, some processing in the application program can realize by client I 100 and client II 110.Application program 160 has comprised following main program and mechanism: form program 161, user estimate mechanism 162, user preference study mechanism 163, concurrent reptile robot program 164 and instant messaging program 165.
Form program 161: refer to such program, it is with the content structureization of Search Results 152, resolve into the field that display page needs, as a patent information is resolved into patent name, the inventor, fields such as patent summary, and these are decomposed good field deposit in correspondingly in correspondingly the list cell, system calls out this form then, is shown as the page that the user sees.
The user estimates mechanism 162: the user is by the evaluation to Search Results, it is thought that relatively meeting the Search Results of searching for purpose and preference picks out, system is according to the critical sentence/speech in the summary info of these Search Results of choosing, further search for, thereby reach the purpose that in-depth is searched for; On the other hand, the user passes through the evaluation to Search Results, the mistake of update the system preferential learning and deviation, thus corrected user's statistical model 153.
User preference study mechanism 163: server-side system 130 is by being stored in the user's statistical model 153 and related dictionary 151 in the database server 131, by the data in Search Results and the user's statistical model being carried out degree of association coupling, promptly the critical sentence word set according to user preference and custom is searched for once more in these results, the high more expression of degree of association user is to this Search Results preference more, and promptly interest is big more.According to the degree of association, system deletes to Search Results and sorts that it is just forward more that the Search Results that the degree of association is high shows.
Concurrent reptile robot program 164: system responses user's searching request, and an information relevant with all critical sentence/speech in the critical sentence word set grasps needed data and information from each external data source 140, and the program of a kind of like this method of realization is called concurrent reptile robot program.The Search Results that gets by this programmed acquisition deposits in the database, and upgrades user's statistical model with this by analysis.In the exemplary embodiment of this explanation, concurrent reptile robot program 164 has used correlation techniques such as http protocol, socket technology, cookie thread pool, dom4j, XML, regular expression.
Http protocol: http protocol (Hypertext Transfer Protocol, HTML (Hypertext Markup Language)) is the transportation protocol that is used for from www server transmission hypertext to local browser.It can make browser more efficient, and Network Transmission is reduced.It guarantees that not only digital terminal correctly transmits hypertext document apace, also determines which part in the transferring documents, and which partial content at first shows (as text prior to figure) etc.
Socket: so-called socket is also referred to as " socket " usually, is used to describe IP address and port, is the handle of a communication chain.Application program is sent to network by " socket " usually and is asked or reply network requests.
Cookie: as among Fig. 7 to the explanation of client cookies file 570, cookie is a document files, can only be read and call by specific website.
Dom4j:dom4j is the XML API of a Java, is similar to jdom, is used for reading and writing the XML file.Dom4j is a very outstanding Java XML API, has the characteristics of excellent performance, powerful and extremely easy-to-use use.
XML:XML represents Extensible Markup Language (abbreviation of eXtensible Markup Language means extendible SGML).XML is the rule of a cover definition semantic marker, and these marks are divided into many parts with document and these parts are labelled.It also is the meta-tag language, has promptly defined to be used to define other relevant with specific area, syntax-languages semantic, structurized SGML.XML has defined the first sentence structure of a cover, if an application program is appreciated that this monobasic sentence structure, it also just automatically can understand the language that all meta-language is thus set up so.What XML described is structure and semanteme, rather than format.
Regular expression: regular expression (regular expression) has been described a kind of pattern of string matching, can be used for checking whether a string contains certain substring, the substring of coupling be done replaced or take out the substring that meets certain condition etc. from certain string.Regular expression mates certain character pattern and the character string of being searched for as a template.
Instant messaging program 165: in Search Results 152, patent information 410, business opportunity information 420, company information 430 has all related to the telephone number of company, instant messaging 165 is such programs, the user is by the user input apparatus of client I 100 or client II110, as mouse, certain company in system request and Search Results carries out communication, system start-up instant messaging application program, the fixed telephone terminal or the network telephone terminal of this user and this company are connected, pick up the telephone machine microphone or start network telephone terminal of the personnel of the said firm, promptly represent the communication successful connection, the user utilizes the audio frequency input-output unit, just can immediately seek advice from as earphone and microphone, and called associate also can utilize the consulting of fixed telephone or earphone and microphone answer to interested company.Like this, the user need not utilize communication apparatus call peers such as landline telephone when having a question, but directly finishes consulting on the net.
Should be appreciated that Fig. 1 just illustrates wherein a kind of demonstration system in order to be illustrated more clearly in the present invention, but do not represent the present invention just to be confined to this scope.
Fig. 2 below.Fig. 2 illustrates the processing procedure of exemplary embodiment.Wherein the frame of broken lines among the figure partly is step or the sightless step of user that carry out on the system backstage.At first the user logs on system website by client I 100 or client II 110, promptly sends information request by digital network 120 to server-side system 130, and server-side system 130 returns to the user with initial page information 200.Initial page 200 comprises following components:
Search statement input frame 201: in search statement input frame 201, the user can import one and have the complete statement of searching for purpose, as " the hard disk price in Hangzhou August how? " Also can import keyword, as " computer Hangzhou ".
Pay close attention to module customization button 202: be used for starting custom program, after the user clicked this button, system turned customized web page automatically, and by this mechanism, the user can customize own interested content, as specific industry and specific geographic position etc.Certainly, the effective prerequisite of this button is that this user has been registered user and login system, and this prerequisite also has similar description in following step.
User login/registration button 203: the user can be registered as the registered user of this system by this button, also can log on this system by this button, so that system start-up user statistical model 153 makes Search Results more accurate.
In the step 210, the user is by the user input unit among client I 100 or the client II 110, as keyboard, problem statement or the keyword searched for wanted in input in search statement input frame 201, as " the hard disk price in Hangzhou August how? ", " computer Hangzhou " etc.
Server-side system 130 receives searching request, at first execution in step 211, problem statement or keyword to user input carry out the high speed Chinese word segmentation, will " the hard disk price in Hangzhou August how? " this complete statement semantics is decomposed into " Hangzhou ", " hard disk price ", " August " these several critical sentence/speech.
Follow step 212, server-side system 130 externally checks whether this user's related dictionary 151 comprises the relevant critical sentence/speech of these critical sentences/speech phase Sihe in the data source 140 with concurrent reptile robot program 164 in local database server 131 and by network 120.
Follow step 213, system with these similar and relevant critical sentence/speech add after semantic the decomposition critical sentence/speech together, extract from this user's related dictionary 151, generate a new critical sentence word set, this critical sentence word set has comprised all above-mentioned critical sentence/speech.
Then in step 214, server-side system 130 access local database servers 131 and the information that comprises these critical sentence/speech by network 120 and concurrent reptile robot program 164 from external data source 140 request search.
Step 218 subsequently, system start-up user preference study mechanism 163, utilize user's statistical model 153 of related dictionary 151 and specific user to carry out user's preferential learning, that draw which critical sentence/speech and be user preference or meet the user search custom, judge according to these critical sentence/speech whether the next result of search is useful for this user, the degree of association is higher, and continues execution in step 219 according to this thinking.
Step 219 is utilized the result of user preference study, and system deletes, sorts Search Results, incoherent information is deleted from Search Results, the degree of association higher be arranged in before.
In step 220 subsequently, the form program 161 in the system call apps server 133 is write the Search Results that has sorted in the form of webpage with structured way, makes every content corresponding one by one, in order succinct.Then system shows the user with the Search Results 230 of formization.And while execution in step 221, step 222 and step 223.In the step 221, system utilizes the user's statistical model 153 in the Search Results update service device end system 130, and stores in the database server 131.In the step 222, system utilizes the cookies file among Search Results renewal client I 100 or the client II 110.Step 223 kind, system utilizes the critical sentence/speech of Search Results to upgrade this user's related dictionary.Show user's Search Results 230 to comprise following information at last:
The user estimates check box 231: a check box is all arranged before Search Results describes 234, the reader can choose this check box to represent that attention rate to this Search Results is than other unchecked Search Results height, thereby make server-side system 130 further to search for, and upgrade user's statistical model simultaneously according to this according to these Search Results of choosing.
Search Results describes 234: represent a Search Results briefly, but it should be noted that Search Results is described sometimes can not be fully or correctly reflect the content of Search Results.
Instant messaging button 235: this button excites instant messaging program 165, and purpose is momentarily to obtain voice contact with the opposing party, so that obtain the most accurate up-to-date information.
In step 236, the user chooses by input block such as mouse, keyboard etc. and estimates check box 231, and expression is comparatively satisfied, interested to this Search Results.In the later step 237, the user clicks search button once more, the system start-up user estimates mechanism 162, the Search Results of choosing is carried out the high speed Chinese word segmentation again, exciting step 211 and step subsequently once more, purpose be again in whole network data but not in primary Search Results the search information relevant with choosing Search Results, searching for the information that gets once more may be more, abundanter than the information that search for the first time gets, rather than search fewer and fewer, so also make the needs that Search Results is more accurate, more be close to the users.This process also can be upgraded user's statistical model 153, thereby makes that the learning process of user preference study mechanism 163 is more accurate.
Can be alternatively, user's execution in step 238, the user moves to the result with mouse and describes on 234.At this moment exciting step 239, and system shows the user with the summary of this object information, and the user can judge clearly by this outline information whether this information is useful to it.
Can be alternatively, user's execution in step 240 is with click instant messaging button 235.After system received user's request, step 241, system judged that the user whether with audio frequency input-output device, is connected on the computing machine as earphone, microphone apparatus.
If system can detect these equipment in running, then execution in step 244, and the prompting user puts on headset and guarantees the microphone unlatching.At this moment enter step 245 after the other side hangs on, success has been set up in the expression communication.
And if system monitoring is not connected to earphone and microphone on the computing machine to the user, then execution in step 242, and the system prompt user connects equipment such as earphone and microphone and computing machine.
Treat that the user connects communication apparatus, promptly after the step 243, system continues execution in step 244 and subsequent step thereof.
Can be alternatively, user's execution in step 246 is clicked Search Results with mouse or keyboard and is described 234.Subsequently, system's execution in step 247, the search result web page of link is shown to the user, and continues execution in step 221, step 222 and step 223, update service device end subscriber statistical model 153, client cookies file and this user's related dictionary 151.
In the alternative steps 250 of step 210, the user can customize own interested content, and as specific industry and geographic position, but this function is only open to the registered user.After the user clicks and pays close attention to module customization button 202, system's actuating logic determining step 251 judges whether the user has landed the website, if the user lands, then this user must be the registered user of system, and then system continues execution in step 221, step 222 and step 223.In the step 221, system utilizes the user's statistical model 153 in the Search Results update service device end system 130, and stores in the database server 131.In the step 222, system utilizes the cookies file among Search Results renewal client I 100 or the client II 110.In the step 223, system utilizes the critical sentence/speech of Search Results to upgrade related dictionary 151.
If the result of the logic determines step 251 of system is a "No", promptly the user does not land this system, and then system's execution in step 252, explicit user registration/landing frame.
Then step 253, if this user be the registered user of this system, then the user can select execution in step 254, input username and password or be password logs on this system website then.
Can be alternatively, if this user has not yet registered, then the user can be by the information of submitting to registration to need, promptly step 255 is registered as the registered user of this system.Subsequently, to utilize log-on message automatically be the newly-built user's statistical model 153 of this user and be stored in the database server 131 for step 256, server-side system 130.Simultaneously, server-side system 130 execution in step 222 are upgraded the cookies file among client I 100 or the client II 110.
Certainly, can find out at an easy rate, this flow process is not necessarily to carry out in proper order as described above, but process that constantly circulates repeatedly, the difference of sequence of steps does not influence the system that realizes method of the present invention, so the present invention is not subject to the process flow diagram that this exemplary embodiment is drawn yet.
Fig. 8, Fig. 9, Figure 10 below, what these three figure showed respectively is the searched page sectional drawing of three contents among the embodiment: patent, business opportunity and company.
Fig. 8 illustrates the user interface sectional drawing of the patent information search and webpage of exemplary embodiment establishment and generation.Wherein search statement input frame 610 is corresponding to the search statement input frame 201 of Fig. 2, both are two different number in the figure differences, but the function of carrying out is identical, for example the user is in 610 inputs " mobile phone " of search statement input frame, then the patent information that system is relevant with mobile phone is shown to the user, be among Fig. 2, system's execution in step 230, the information of returning comprise that patent describes 613, affiliated Business Name 614, the email615 of company, telephone number 616 and contact address 618.Wherein, the user has chosen two patents to describe 613 preceding evaluation check boxes 612, represents that these two patent information are user's needs.After the user clicked once more search button 611, the step 237 in system's meeting execution graph 2 was searched for relevant information again at whole network data base.What estimate that check box 612 back shows the user is that patent describes 613, and Fig. 8 shows, and to be the user move on to patent with cursor describes situation on 613, and at this moment, the step 239 in system's execution graph 2 is shown to the user with the summary info 619 of this patent.After each telephone number 616, an instant messaging button 617 is all arranged, if user's interesting or query to this patent can be clicked this button and be connected to the other side and carry out voice call.Certainly, the function that this system also provides the general search engine to provide, i.e. filter information in the result is in native system, the user can screen as email, phone, address according to contact method 620, also can screen as Beijing, Zhejiang, Shanghai, Hubei according to special key words 621.In addition, system is when being shown to the user to the patent information of user search, relevant business opportunity information 622 and associated companies information 623 also are provided, have been convenient to the user and search, and these business opportunities are to be undertaken related by the numbering of the company among Fig. 6 451 with company information.
Fig. 9 illustrates the user interface sectional drawing of the business opportunity information search webpage of exemplary embodiment establishment and generation.Wherein search statement input frame 630 is corresponding to the search statement input frame 201 of Fig. 2 and the search statement input frame 610 of Fig. 8, the three is in different number in the figure differences, but the function of carrying out is identical, for example the user is in 630 inputs " computer " of search statement input frame, then the business opportunity information that system is relevant with computer is shown to the user, be among Fig. 2, system's execution in step 230, the information of returning comprise that business opportunity describes 633, business opportunity type 634, affiliated Business Name 635, telephone number 636, contact address 638.Wherein, the user has chosen five business opportunities to describe 633 preceding evaluation check boxes 632, represents that these five business opportunity information are user's needs.After the user clicked once more search button 631, the step 237 in system's meeting execution graph 2 was searched for relevant information again at whole network data base.What estimate that check box 632 back shows the user is that business opportunity describes 633, and Fig. 9 shows, and to be the user move on to business opportunity with cursor describes situation on 633, and at this moment, the step 239 in system's execution graph 2 is shown to the user with the summary info 639 of this business opportunity.Different with patent information is that what business opportunity type 634 was represented is that this business opportunity is sale information or wants to buy information.After each telephone number 636, an instant messaging button 637 is all arranged, if user's interesting or query to this business opportunity can be clicked this button and be connected to the other side and carry out voice call.Certainly, the function that this system also provides the general search engine to provide, i.e. filter information in the result, in native system, the user can as sell according to business opportunity type 640, want to buy and screen, also can screen as email, phone, address, can also screen as Beijing, Zhejiang, Shanghai, Hubei according to special key words 643 according to contact method 641.In addition, system is when being shown to the user to the business opportunity information of user search, relevant patent information 643 and associated companies information 644 also are provided, have been convenient to the user and search, and these patents are to be undertaken related by the numbering of the company among Fig. 6 451 with company information.
Figure 10 illustrates the user interface sectional drawing of the company information search and webpage of exemplary embodiment establishment and generation.Wherein search statement input frame 650 is corresponding to the search statement input frame 201 of Fig. 2, the search statement input frame 610 of Fig. 8 and the search statement input frame 630 of Fig. 9, four in different number in the figure differences, but the function of carrying out is identical, for example the user is in 650 inputs " computer " of search statement input frame, then the company information that system is relevant with computer is shown to the user, be among Fig. 2, system's execution in step 230, the information of returning comprise that company describes 653, company's type 654, registered capital 655, telephone number 656, contact address 658, postcode 659.Wherein, the user has chosen three companies to describe 653 preceding evaluation check boxes 652, represents that these three company informations are user's needs.After the user clicked once more search button 651, the step 237 in system's meeting execution graph 2 was searched for relevant information again at whole network data base.Different with business opportunity information with patent information is that what company's type 654 was represented is that the said firm is trade type, type of production, service type or government or other mechanism.After each telephone number 656, an instant messaging button 657 is all arranged, if user's interesting or query to this company can be clicked this button and be connected to the other side and carry out voice call.Certainly, the function that this system also provides the general search engine to provide, i.e. filter information in the result, in native system, the user can screen as trade type, type of production, service type, government or other mechanism according to company management pattern 661, also can screen as email, phone, address, can also screen as Beijing, Zhejiang, Shanghai, Hubei according to special key words 663 according to contact method 662.In addition, system is when being shown to the user to the company information of user search, relevant patent information 664 and relevant business opportunity information 665 also are provided, have been convenient to the user and search, and these patents are to be undertaken related by the numbering of the company among Fig. 6 451 with business opportunity information.Figure 10 shows, and to be the user move on to situation on relevant business opportunity information 665 clauses and subclauses with cursor, and similarly, the step 239 in system's execution graph 2 is shown to the user with the summary info 660 of this business opportunity.
More than by to reference to the accompanying drawings detailed description; the person skilled in art can understand the realization principle and the mechanism of the method for the invention and system at an easy rate; drafting with reference to the accompanying drawings is just in order to illustrate method and system of the present invention better; rather than the scope of regulation protection, protection scope of the present invention is defined by appended claims.In addition to the implementation, the present invention can also have other embodiments.All employings are equal to the technical scheme of replacement or equivalent transformation formation, all drop on the protection domain of requirement of the present invention.
Claims (3)
1. self-help intelligent uprightness searching method, it is characterized in that: this method comprises the steps:
1.1), utilize the concern module of user cookies file, log-on message, historical search information and the customization be stored in the client and server end to carry out user preference study, and this user preference be established as user's statistical model in real time, dynamically store in the search engine database;
1.2), utilize to close linking verses/dictionary the search statement of user's input carried out the high speed Chinese word segmentation, and generate a critical sentence/word set file, this document has comprised all relevant, similar critical sentence and keywords that carry out after the semantic analysis, user's historical search information is carried out statistical learning, draw the critical sentence/speech relevant, similar in user's search custom with these critical sentence/word sets, by high speed Chinese word segmentation and user search behavior learning, draw a final critical sentence/word set file;
1.3), search engine is by all information relevant with these inertia critical sentence/speech of network data library searching, simultaneously, these Search Results and user's statistical model are mated, its critical sentence/lexicon of search share the information of family preference in these Search Results, finally, the Search Results that will meet user preference returns to the user.
2. self-help intelligent uprightness searching method according to claim 1 is characterized in that: the user is reflected the evaluation of Search Results, revise the process of machine learning in view of the above, and revise user's statistical model simultaneously.
3. self-help intelligent uprightness searching method according to claim 1 is characterized in that: the user can customize interested content and information, form Search Results and instant messaging.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007100709770A CN101114294A (en) | 2007-08-22 | 2007-08-22 | Self-help intelligent uprightness searching method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007100709770A CN101114294A (en) | 2007-08-22 | 2007-08-22 | Self-help intelligent uprightness searching method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101114294A true CN101114294A (en) | 2008-01-30 |
Family
ID=39022640
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2007100709770A Pending CN101114294A (en) | 2007-08-22 | 2007-08-22 | Self-help intelligent uprightness searching method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101114294A (en) |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010025653A1 (en) * | 2008-09-08 | 2010-03-11 | 华为技术有限公司 | Searching information method, system, device, and vertical search engine register method |
| CN101930457A (en) * | 2010-08-13 | 2010-12-29 | 百度在线网络技术（北京）有限公司 | Quick object selecting and searching method, equipment and system for user |
| CN101604324B (en) * | 2009-07-15 | 2011-11-23 | 中国科学技术大学 | Method and system for searching video service website based on meta search |
| WO2012025040A1 (en) * | 2010-08-27 | 2012-03-01 | Huang Bin | Visualized search engine system and implementation method and application thereof |
| CN102385636A (en) * | 2011-12-22 | 2012-03-21 | 陈伟 | Intelligent searching method and device |
| CN102624675A (en) * | 2011-01-27 | 2012-08-01 | 腾讯科技（深圳）有限公司 | Self-service customer service system and method |
| CN102654868A (en) * | 2011-03-02 | 2012-09-05 | 联想(北京)有限公司 | Keyword-based search method and device and server |
| CN103425656A (en) * | 2012-05-15 | 2013-12-04 | 阿里巴巴集团控股有限公司 | Commodity information searching method, server and terminal |
| CN103425659A (en) * | 2012-05-15 | 2013-12-04 | 阿里巴巴集团控股有限公司 | Method and server for searching information on basis of geographical locations |
| CN103617266A (en) * | 2013-12-03 | 2014-03-05 | 北京奇虎科技有限公司 | Personalized extension search method, device and system |
| CN104090757A (en) * | 2012-05-04 | 2014-10-08 | 北京奇虎科技有限公司 | Method and device for displaying rich media information in browser |
| CN104090923A (en) * | 2012-05-04 | 2014-10-08 | 北京奇虎科技有限公司 | Method and device for displaying rich media information in browser |
| CN104462552A (en) * | 2014-12-25 | 2015-03-25 | 北京奇虎科技有限公司 | Question and answer page core word extracting method and device |
| CN103310663B (en) * | 2013-05-09 | 2015-08-12 | 北京网梯科技发展有限公司 | A kind of Intelligent point-reading method, equipment and system |
| CN104965919A (en) * | 2015-07-06 | 2015-10-07 | 无锡天脉聚源传媒科技有限公司 | Search processing method and apparatus |
| CN105045854A (en) * | 2015-07-07 | 2015-11-11 | 国家电网公司 | Nutch based vertical search engine and method |
| CN105975506A (en) * | 2016-04-28 | 2016-09-28 | 百度在线网络技术（北京）有限公司 | Service search method and device |
| CN106202105A (en) * | 2015-05-06 | 2016-12-07 | 阿里巴巴集团控股有限公司 | A kind of e-commerce website air navigation aid and device |
| CN106371843A (en) * | 2016-08-31 | 2017-02-01 | 天脉聚源（北京）科技有限公司 | Method and device for displaying login information |
| CN106603683A (en) * | 2016-12-23 | 2017-04-26 | 安徽维智知识产权代理有限公司 | Enterprise cooperation platform based on IP address |
| WO2017084362A1 (en) * | 2015-11-18 | 2017-05-26 | 百度在线网络技术（北京）有限公司 | Model generation method, recommendation method and corresponding apparatuses, device and storage medium |
| CN106997343A (en) * | 2017-03-28 | 2017-08-01 | 联想(北京)有限公司 | Information processing method and equipment |
| CN107577726A (en) * | 2017-08-22 | 2018-01-12 | 努比亚技术有限公司 | A kind of searching method, server and computer-readable recording medium |
| CN108701014A (en) * | 2016-03-09 | 2018-10-23 | 电子湾有限公司 | Inquiry database for tail portion inquiry |
| CN109118330A (en) * | 2018-08-09 | 2019-01-01 | 珠海格力电器股份有限公司 | Household appliance recommendation method and device, storage medium and server |
| CN110445744A (en) * | 2018-05-02 | 2019-11-12 | 阿里巴巴集团控股有限公司 | A kind of data processing method and device |
| CN110516044A (en) * | 2019-08-30 | 2019-11-29 | 北京地厚云图科技有限公司 | Application method of the natural language intelligent search in construction management task |
| CN111598595A (en) * | 2019-02-21 | 2020-08-28 | 阿里巴巴集团控股有限公司 | Information stream data display method and device and terminal equipment |
| CN111723283A (en) * | 2019-03-22 | 2020-09-29 | 佳能株式会社 | Information processing apparatus, method, and computer-readable storage medium |
| CN113343068A (en) * | 2021-06-09 | 2021-09-03 | 聘聘云(上海)智能科技有限公司 | Data search method and device, storage medium and electronic device |
| US11593855B2 (en) | 2015-12-30 | 2023-02-28 | Ebay Inc. | System and method for computing features that apply to infrequent queries |
-
2007
- 2007-08-22 CN CNA2007100709770A patent/CN101114294A/en active Pending
Cited By (42)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8417684B2 (en) | 2008-09-08 | 2013-04-09 | Huawei Technologies Co., Ltd. | Method, system, and device for searching for information and method for registering vertical search engine |
| WO2010025653A1 (en) * | 2008-09-08 | 2010-03-11 | 华为技术有限公司 | Searching information method, system, device, and vertical search engine register method |
| CN101673272B (en) * | 2008-09-08 | 2012-12-19 | 华为技术有限公司 | Method, system and device for searching information and method for registering vertical search engine |
| CN101604324B (en) * | 2009-07-15 | 2011-11-23 | 中国科学技术大学 | Method and system for searching video service website based on meta search |
| CN101930457A (en) * | 2010-08-13 | 2010-12-29 | 百度在线网络技术（北京）有限公司 | Quick object selecting and searching method, equipment and system for user |
| WO2012025040A1 (en) * | 2010-08-27 | 2012-03-01 | Huang Bin | Visualized search engine system and implementation method and application thereof |
| CN102624675B (en) * | 2011-01-27 | 2014-08-06 | 腾讯科技（深圳）有限公司 | Self-service customer service system and method |
| CN102624675A (en) * | 2011-01-27 | 2012-08-01 | 腾讯科技（深圳）有限公司 | Self-service customer service system and method |
| CN102654868A (en) * | 2011-03-02 | 2012-09-05 | 联想(北京)有限公司 | Keyword-based search method and device and server |
| CN102654868B (en) * | 2011-03-02 | 2015-11-25 | 联想(北京)有限公司 | A kind of searching method based on key word, searcher and server |
| CN102385636A (en) * | 2011-12-22 | 2012-03-21 | 陈伟 | Intelligent searching method and device |
| CN104090757B (en) * | 2012-05-04 | 2018-10-12 | 北京奇虎科技有限公司 | For the rich media information methods of exhibiting of browser |
| CN104090923A (en) * | 2012-05-04 | 2014-10-08 | 北京奇虎科技有限公司 | Method and device for displaying rich media information in browser |
| CN104090757A (en) * | 2012-05-04 | 2014-10-08 | 北京奇虎科技有限公司 | Method and device for displaying rich media information in browser |
| US9390103B2 (en) | 2012-05-15 | 2016-07-12 | Alibaba Group Holding Limited | Information searching method and system based on geographic location |
| CN103425656B (en) * | 2012-05-15 | 2017-05-31 | 阿里巴巴集团控股有限公司 | The searching method of merchandise news, server and terminal |
| CN103425659A (en) * | 2012-05-15 | 2013-12-04 | 阿里巴巴集团控股有限公司 | Method and server for searching information on basis of geographical locations |
| CN103425656A (en) * | 2012-05-15 | 2013-12-04 | 阿里巴巴集团控股有限公司 | Commodity information searching method, server and terminal |
| CN103425659B (en) * | 2012-05-15 | 2017-06-09 | 阿里巴巴集团控股有限公司 | Information search method and server based on geographical position |
| CN103310663B (en) * | 2013-05-09 | 2015-08-12 | 北京网梯科技发展有限公司 | A kind of Intelligent point-reading method, equipment and system |
| CN103617266A (en) * | 2013-12-03 | 2014-03-05 | 北京奇虎科技有限公司 | Personalized extension search method, device and system |
| CN104462552B (en) * | 2014-12-25 | 2018-07-17 | 北京奇虎科技有限公司 | Question and answer page core word extracting method and device |
| CN104462552A (en) * | 2014-12-25 | 2015-03-25 | 北京奇虎科技有限公司 | Question and answer page core word extracting method and device |
| CN106202105A (en) * | 2015-05-06 | 2016-12-07 | 阿里巴巴集团控股有限公司 | A kind of e-commerce website air navigation aid and device |
| CN104965919A (en) * | 2015-07-06 | 2015-10-07 | 无锡天脉聚源传媒科技有限公司 | Search processing method and apparatus |
| CN105045854A (en) * | 2015-07-07 | 2015-11-11 | 国家电网公司 | Nutch based vertical search engine and method |
| WO2017084362A1 (en) * | 2015-11-18 | 2017-05-26 | 百度在线网络技术（北京）有限公司 | Model generation method, recommendation method and corresponding apparatuses, device and storage medium |
| US11593855B2 (en) | 2015-12-30 | 2023-02-28 | Ebay Inc. | System and method for computing features that apply to infrequent queries |
| CN108701014A (en) * | 2016-03-09 | 2018-10-23 | 电子湾有限公司 | Inquiry database for tail portion inquiry |
| CN105975506A (en) * | 2016-04-28 | 2016-09-28 | 百度在线网络技术（北京）有限公司 | Service search method and device |
| CN106371843A (en) * | 2016-08-31 | 2017-02-01 | 天脉聚源（北京）科技有限公司 | Method and device for displaying login information |
| CN106603683A (en) * | 2016-12-23 | 2017-04-26 | 安徽维智知识产权代理有限公司 | Enterprise cooperation platform based on IP address |
| CN106997343A (en) * | 2017-03-28 | 2017-08-01 | 联想(北京)有限公司 | Information processing method and equipment |
| CN107577726A (en) * | 2017-08-22 | 2018-01-12 | 努比亚技术有限公司 | A kind of searching method, server and computer-readable recording medium |
| CN110445744A (en) * | 2018-05-02 | 2019-11-12 | 阿里巴巴集团控股有限公司 | A kind of data processing method and device |
| CN110445744B (en) * | 2018-05-02 | 2022-06-28 | 阿里巴巴集团控股有限公司 | Data processing method and device |
| CN109118330A (en) * | 2018-08-09 | 2019-01-01 | 珠海格力电器股份有限公司 | Household appliance recommendation method and device, storage medium and server |
| CN111598595A (en) * | 2019-02-21 | 2020-08-28 | 阿里巴巴集团控股有限公司 | Information stream data display method and device and terminal equipment |
| CN111598595B (en) * | 2019-02-21 | 2024-03-29 | 阿里巴巴集团控股有限公司 | Information stream data display method and device and terminal equipment |
| CN111723283A (en) * | 2019-03-22 | 2020-09-29 | 佳能株式会社 | Information processing apparatus, method, and computer-readable storage medium |
| CN110516044A (en) * | 2019-08-30 | 2019-11-29 | 北京地厚云图科技有限公司 | Application method of the natural language intelligent search in construction management task |
| CN113343068A (en) * | 2021-06-09 | 2021-09-03 | 聘聘云(上海)智能科技有限公司 | Data search method and device, storage medium and electronic device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101114294A (en) | Self-help intelligent uprightness searching method | |
| US8935277B2 (en) | Context-aware question answering system | |
| US7756807B1 (en) | System and method for facts extraction and domain knowledge repository creation from unstructured and semi-structured documents | |
| US7730008B2 (en) | Database interface and database analysis system | |
| US7272595B2 (en) | Information search support system, application server, information search method, and program product | |
| US8060513B2 (en) | Information processing with integrated semantic contexts | |
| CA2647584C (en) | Search-based application development framework | |
| US7289985B2 (en) | Enhanced document retrieval | |
| US9727628B2 (en) | System and method of applying globally unique identifiers to relate distributed data sources | |
| CN102073726B (en) | Structured data import method and device for search engine system | |
| US20190286676A1 (en) | Contextual content collection, filtering, enrichment, curation and distribution | |
| US8473473B2 (en) | Object oriented data and metadata based search | |
| EP1587009A2 (en) | Content propagation for enhanced document retrieval | |
| US20020042789A1 (en) | Internet search engine with interactive search criteria construction | |
| US20020152202A1 (en) | Method and system for retrieving information using natural language queries | |
| US20100005087A1 (en) | Facilitating collaborative searching using semantic contexts associated with information | |
| TW201108007A (en) | Semantic trading floor | |
| CN102073725A (en) | Method for searching structured data and search engine system for implementing same | |
| CN103425714A (en) | Query method and system | |
| CA3088695A1 (en) | Method and system for decoding user intent from natural language queries | |
| Agrawal et al. | Qnamaker: Data to bot in 2 minutes | |
| US20050080774A1 (en) | Ranking of business objects for search engines | |
| EP1505520A2 (en) | Ranking of business objects for search engines | |
| Färber | Linked crunchbase: a linked data api and rdf data set about innovative companies | |
| CN201087865Y (en) | Personalized intelligent vertical searching system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20080130 |