CN1325076A - Comprehensive network Chinese information searcher - Google Patents
Comprehensive network Chinese information searcher Download PDFInfo
- Publication number
- CN1325076A CN1325076A CN 00115797 CN00115797A CN1325076A CN 1325076 A CN1325076 A CN 1325076A CN 00115797 CN00115797 CN 00115797 CN 00115797 A CN00115797 A CN 00115797A CN 1325076 A CN1325076 A CN 1325076A
- Authority
- CN
- China
- Prior art keywords
- engine
- degree
- correlation
- feedback
- user
- 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
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to a network Chinese information comprehensive searcher, its system includes: input end, noise filtering, interpretation end and automatic regulating adaptive treamtent for the engine of the partner, according to the established network analog mathematical modal analyzing and searching column information, and according to user's request editing and providing for user to use. Said analog mathematical model analysis adopts Wideway Search engine to make unified mathematical model analysis and subsumption for feedback pages of all engines, and unify degree of correlation for different feedback results. Suppose the condition is: original engine weight KL, page self correlation degree C, page feedback time T, one engine feedback result number N, using bell-shaped pulse mathermatical formula to make analysis and obtain unified degree of correlation.
Description
The present invention relates to the disposal system of computer network information, particularly a kind of network Chinese informix searcher.
The current Internet page whole world has 1,000,000,000 approximately, relates to the various aspects of daily life.Once people were also for the information on the Internet was few and worried, nowadays but were trapped in knowledge explosion, too many information makes people condition at a loss as to what to do.From being born of YAHOO search engine, to the appearance of increasing search engine today, demonstrate the attention of people invariably to information searching, how the information world in vastness finds own desired information, becomes the problem that people have to think.
The METAENGINE that occurred on INTERNET in recent years is that the database with each search engine puts together, for the user provides cover ratio wider, more accurate search, because a tame engine is difficult to all pages in the limit world, technically, all impossible in the operation.
Along with popularizing of Chinese INTERNET, Chinese number of netizens just doubled every the several months, and the quantity of information of Chinese is also increasing.Same information searching problem pendulum is in face of people, provide the engine of Chinese search mainly to be at present, Chinese YAHOO, SOHU Chinese, Sina's Chinese, day net, several families such as move about unhurriedly, the Chinese information that they include is limited, classification, arrangement to Chinese simultaneously do not have careful processing, and the information that people often find not is that they are desired.
So far go back the METAENGINE of neither one Chinese on the INTERNET, if the database of each Chinese engine is put together, the result to search chooses carefully simultaneously, allows Chinese also can enjoy the benefit that advanced inquiry brings so.
Purpose of the present invention is exactly to satisfy the demand of user to the query and search of Chinese information for the problem that solves prior art.Adopt the SERVLET technology, use the network environment of the up-to-date present domestic complexity of technical finesse.
System architecture of the present invention comprises a kind of network Chinese informix searcher, system architecture mainly comprises input end, noise filtering, explanation is held and the automatic adjustment adaptation of the other side's engine is handled, the information that searches according to set network analog mathematics model analysis, and ask according to the user, hand over the user to use after the layout, it is characterized in that described network analog mathematics model analysis is to adopt Wideway Search engine that the page of all engine feedback is made unified mathematics model analysis to sort out, and variant feedback result unified the degree of correlation, this degree of correlation analysis condition is: the weight KL of original engine, the degree of correlation C of the page itself, the time T of page feedback, the number of results N of an engine feedback must unify the degree of correlation behind bell pulse mathematics formula analysis.
Fig. 1 is a system architecture synoptic diagram of the present invention
Further specify embodiments of the invention below in conjunction with accompanying drawing
The user proposes retrieval request such as descriptor, keyword etc. at input end, enter page intellectual analysis engine, this engine has been included the most authoritative search engine of present Chinese such as Chinese YAHOO, Chinese EXITE, Sina's Chinese, Sohu's Chinese, Beijing University's sky net, has been moved about unhurriedly, flyings Chinese, Chinese network allusion quotation, Omron, searcher, Netease's Chinese.This search engine is also supported English, includes present world-technology forefront, five tame search engine: YAHOO, ALAVESTA, NORTHERNLIGHT, DIRECTHIT, GOOGLE that data is the most complete.Because the Chinese present network bandwidth is limited, the analysis of the page often will be depended on the result of original engine.In order from finite information, to obtain more information, this engine adopts intelligent inference to analyze the page, with the request of user input through noise filtering, to information classify automatically handle after, automatically find out relevant speech by explaining end from the page that feeds back, count in the degree of correlation, engine adjustment to the other side adapts to processing automatically then, user's request meanwhile enters parallel page request engine and searches for, and the information that searches returned feedback page intellectual analysis engine, the information that searches according to set network analog mathematics model analysis, and ask to hand over the user to use after the layout according to the user.
The network analog mathematics model analysis adopts the WIDEWAYSEARCH engine that the page of all engine feedback is made unified mathematics model analysis to sort out, and the result of different engine feedback is had the unified degree of correlation.
This degree of correlation analysis condition:
1, the weight KL of original engine: this is basic weight.Because there is difference in Chinese engine masses, so with the engine classification, for good engine, the weight of the record that feeds back to will be higher than the result of other engine feedback.
2, the degree of correlation C of the page itself: the result to every feedback makes intellectual analysis, and the degree of correlation of judged result is promptly done the full-text search analysis to brief introduction.
3, the time T of page feedback: because the degree of correlation and server and user's distance dependent, meet user's needs especially such as certain bar record of YAHOO, still will be linked to this page may change the user 10 minutes, and the validity of this record will fall under suspicion.Therefore monitor the feedback result of every record, offer the spended time of this record of user's reference links.And do not make the user do useless trial.
4, the number of results N of an engine feedback: it will be for referencial use by the final degree of correlation, and we notice that the number of results of feedback is relevant with the database size of this engine in fact, and the cover ratio of this engine obviously can judge that the authority of this record exerts an influence to the user.
Behind bell pulse mathematics formula analysis, must unify the degree of correlation according to above degree of correlation analysis condition.
Advantage of the present invention is to adopt many classification to process, in the network environment of present domestic complexity The user can classify according to the engine of the degree of correlation, time, domain name, selection, allows the user The SERVLET technique construction is adopted in easier convenient navigation in information, and very big stretching arranged Special load has been done from suitable to service routine simultaneously to deal with the request of a large number of users in the space Should process, can offer the many and meticulous information of user, and can greatly reduce that the user waits for Time.
Claims (1)
1, a kind of network Chinese informix searcher, system architecture mainly comprises input end, noise filtering, explanation is held and the automatic adjustment adaptation of the other side's engine is handled, the information that searches according to set network analog mathematics model analysis, and ask according to the user, hand over the user to use after the layout, it is characterized in that described network analog mathematics model analysis is to adopt the WidewaySearch engine that the page of all engine feedback is made unified mathematics model analysis to sort out, and variant feedback result unified the degree of correlation, this degree of correlation analysis condition is: the weight KL of original engine, the degree of correlation C of the page itself, the time T of page feedback, the number of results N of an engine feedback must unify the degree of correlation behind bell pulse mathematics formula analysis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 00115797 CN1325076A (en) | 2000-05-23 | 2000-05-23 | Comprehensive network Chinese information searcher |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 00115797 CN1325076A (en) | 2000-05-23 | 2000-05-23 | Comprehensive network Chinese information searcher |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1325076A true CN1325076A (en) | 2001-12-05 |
Family
ID=4585239
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 00115797 Pending CN1325076A (en) | 2000-05-23 | 2000-05-23 | Comprehensive network Chinese information searcher |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1325076A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100371932C (en) * | 2004-03-23 | 2008-02-27 | 南京大学 | Expandable and customizable theme centralized universile-web net reptile setup method |
| CN101454782A (en) * | 2006-03-29 | 2009-06-10 | 甲骨文国际公司 | Contextual search of a collaborative environment |
| CN1648902B (en) * | 2004-01-26 | 2010-12-08 | 微软公司 | System and method for a unified and blended search |
| CN103970816A (en) * | 2013-01-24 | 2014-08-06 | 国际商业机器公司 | Simulating Accesses For Archived Content |
-
2000
- 2000-05-23 CN CN 00115797 patent/CN1325076A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1648902B (en) * | 2004-01-26 | 2010-12-08 | 微软公司 | System and method for a unified and blended search |
| CN100371932C (en) * | 2004-03-23 | 2008-02-27 | 南京大学 | Expandable and customizable theme centralized universile-web net reptile setup method |
| CN101454782A (en) * | 2006-03-29 | 2009-06-10 | 甲骨文国际公司 | Contextual search of a collaborative environment |
| CN101454782B (en) * | 2006-03-29 | 2014-01-29 | 甲骨文国际公司 | Contextual search of a collaborative environment |
| CN103970816A (en) * | 2013-01-24 | 2014-08-06 | 国际商业机器公司 | Simulating Accesses For Archived Content |
| CN103970816B (en) * | 2013-01-24 | 2017-04-05 | 国际商业机器公司 | The method and system of mark content to be issued |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | Query selection techniques for efficient crawling of structured web sources | |
| Tanudjaja et al. | Persona: A contextualized and personalized web search | |
| JP5114380B2 (en) | Reranking and enhancing the relevance of search results | |
| CN100440224C (en) | Automatization processing method of rating of merit of search engine | |
| US20020091661A1 (en) | Method and apparatus for automatic construction of faceted terminological feedback for document retrieval | |
| CN1389811A (en) | Intelligent search method of search engine | |
| CN101908071A (en) | Method and device thereof for improving search efficiency of search engine | |
| WO2008109485A1 (en) | Personalized shopping recommendation based on search units | |
| WO2010065345A1 (en) | System and methods for automatic clustering of ranked and categorized search objects | |
| KR20040029895A (en) | Search system | |
| US20070271228A1 (en) | Documentary search procedure in a distributed system | |
| Sharma et al. | The anatomy of web crawlers | |
| KR20030069640A (en) | System and method for geting information on hierarchical and conceptual clustering | |
| Shekhar et al. | An architectural framework of a crawler for retrieving highly relevant web documents by filtering replicated web collections | |
| Jin et al. | Tise: A temporal search engine for web contents | |
| CN1325076A (en) | Comprehensive network Chinese information searcher | |
| US20030018617A1 (en) | Information retrieval using enhanced document vectors | |
| Yan et al. | An improved PageRank method based on genetic algorithm for web search | |
| Jadidoleslamy | Introduction to metasearch engines and result merging strategies: a survey | |
| US7490082B2 (en) | System and method for searching internet domains | |
| Satokar et al. | Web search result personalization using web mining | |
| Sugiyama et al. | A method of improving feature vector for web pages reflecting the contents of their out-linked pages | |
| Yu et al. | The design and realization of open-source search engine based on Nutch | |
| US20100076964A1 (en) | Instance-Class-Attribute Matching Web Page Ranking | |
| Pardakhe et al. | Enhancement of web search engine results using keyword frequency based ranking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication |