WO2015024522A1 - 搜索方法、系统、搜索引擎和客户端 - Google Patents

搜索方法、系统、搜索引擎和客户端 Download PDF

Info

Publication number
WO2015024522A1
WO2015024522A1 PCT/CN2014/084940 CN2014084940W WO2015024522A1 WO 2015024522 A1 WO2015024522 A1 WO 2015024522A1 CN 2014084940 W CN2014084940 W CN 2014084940W WO 2015024522 A1 WO2015024522 A1 WO 2015024522A1
Authority
WO
WIPO (PCT)
Prior art keywords
search
original
search engine
web pages
client
Prior art date
Application number
PCT/CN2014/084940
Other languages
English (en)
French (fr)
Inventor
巫国忠
刘洋
Original Assignee
百度在线网络技术(北京)有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 百度在线网络技术(北京)有限公司 filed Critical 百度在线网络技术(北京)有限公司
Priority to EP14837976.1A priority Critical patent/EP3037985A4/en
Priority to US14/913,368 priority patent/US20160203222A1/en
Publication of WO2015024522A1 publication Critical patent/WO2015024522A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9538Presentation of query results
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/951Indexing; Web crawling techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24Querying
    • G06F16/248Presentation of query results
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/28Databases characterised by their database models, e.g. relational or object models
    • G06F16/284Relational databases
    • G06F16/285Clustering or classification
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00Computing arrangements based on specific mathematical models
    • G06N7/01Probabilistic graphical models, e.g. probabilistic networks

Definitions

  • the present invention relates to the field of search technologies, and in particular, to a search method, a system, a search engine, and a client. Background technique
  • the present invention aims to solve at least one of the above technical problems.
  • a first object of the present invention is to propose a search method.
  • the method provides a more convenient way to obtain the original webpage, so that the client can easily access the original webpage, increase the probability that the original webpage is clicked, and improve the degree of attention of the client user to the original webpage.
  • a second object of the present invention is to propose a search system.
  • a third object of the present invention is to propose a search engine.
  • a fourth object of the present invention is to propose a client.
  • a fifth object of the present invention is to propose an application.
  • a sixth object of the present invention is to provide a storage medium.
  • the search method of the first aspect of the present invention includes the following steps: the search engine acquires a search term sent by the client; the search engine acquires multiple search results according to the search term, and acquires the respectively a plurality of the plurality of search results corresponding to the original webpage corresponding to the search result; the search engine provides the original webpage corresponding to the plurality of search results and the at least part of the search results to the client, and presents the client through the client .
  • the search engine when the search engine provides the search result to the client, the search result is associated with the original webpage, and the original webpage of the search result is also displayed.
  • the search result if the search result is ranked higher, it has The original page can also be displayed in the top position when the original page is provided, which provides a more convenient way to get the original page, which makes the client can easily access the original page, increase the probability that the original page is clicked, and improve the client.
  • the user's emphasis on the original webpage is convenient without modifying the overall search result ranking. Users also emphasize the search engine's respect for the original page and support for action.
  • users read the original page they often find more relevant content and knowledge. For example, for blogs, the original author's blog site will contain more. Multi-domain related knowledge and articles to enhance the user experience.
  • the search system of the second aspect of the present invention includes: a client and a search engine, wherein the client is configured to send a search term to the search engine and return the search engine The content is displayed; the search engine is configured to acquire the search term, and obtain a plurality of search results according to the search term, and respectively obtain original webpages corresponding to at least some of the plurality of search results, and The plurality of search results and the original webpage corresponding to the at least part of the search results are provided to the client.
  • the search engine when the search engine provides the search result to the client, the search result is associated with the original webpage, and the original webpage of the search result is also displayed.
  • the search result if the search result is ranked higher, it has The original page can also be displayed in the top position when the original page is provided, which provides a more convenient way to get the original page, which makes the client can easily access the original page, increase the probability that the original page is clicked, and improve the client.
  • the user's emphasis on the original webpage without modifying the overall search result ranking, is convenient for the user to emphasize the search engine's respect for the original webpage and the support of the action.
  • the user often finds when reading the original webpage. More relevant content and knowledge, for example, for blogs, the original author's blog site will contain more domain-related knowledge and articles to enhance the user experience.
  • the search engine of the third aspect of the present invention includes: a search term obtaining module, configured to acquire a search term sent by a client; and a search result obtaining module, configured to acquire multiple search results according to the search term. And acquiring, respectively, an original webpage corresponding to at least part of the plurality of search results; and a search result sending module, configured to provide the original webpage corresponding to the plurality of search results and the at least part of the search result to the client End, for the client to display.
  • the search engine of the embodiment of the present invention associates the search result with the original webpage when the search result is provided to the client, so that the client also displays the original webpage of the search result, on the one hand, if the search result is ranked higher.
  • the original webpage can also be displayed in the top position, which provides a more convenient way to obtain the original webpage, so that the client can easily access the original webpage, increasing the probability that the original webpage is clicked, and improving the probability.
  • the degree of emphasis on the original page by the client user without modifying the overall search result ordering, is convenient for the user to emphasize the respect and action support of the search engine for the original page.
  • the user reads the original page, More relevant content and knowledge can be found. For example, for blogs, the original author's blog site will contain more domain-related knowledge and articles to enhance the user experience.
  • the client of the fourth aspect of the present invention includes: a search term sending module, configured to send a search term to a search engine; and a search result obtaining module, configured to acquire, according to the search term, the search engine Multiple search results and original pages corresponding to at least some of the plurality of search results; and search results a presentation module, configured to display the plurality of search results and the original webpage corresponding to the at least part of the search results.
  • the client of the embodiment of the present invention can obtain the search result sent by the search engine and the original webpage, and display the original webpage of the search result.
  • the original webpage can also be used when the original webpage has the original webpage. Displayed in the top position, providing a more convenient way to get the original page, so that the client can easily access the original page, increase the probability of the original page being clicked, and improve the importance of the client user to the original page.
  • users read the original webpage they often find more relevant content and knowledge. For example, for blogs, the original author's blog site will contain more domain-related knowledge and articles to enhance the user experience.
  • an application of the fifth aspect of the present invention is for performing a search method of the first embodiment of the present invention at runtime.
  • the application program of the embodiment of the present invention associates the search result with the original webpage when the search engine provides the search result to the client, and also displays the original webpage of the search result, on the one hand, if the search result ranks high.
  • the original webpage can also be displayed in the top position, which provides a more convenient way to obtain the original webpage, which makes the client can easily access the original webpage, increases the probability that the original webpage is clicked, and improves the customer.
  • the degree of emphasis on the original page by the end user, without modifying the overall search result ordering is convenient for the user to emphasize the respect and action support of the search engine for the original page.
  • the user often reads the original page. Discover more relevant content and knowledge. For example, for blogs, the original author's blog site will contain more domain-related knowledge and articles to enhance the user experience.
  • a storage medium for storing an application for executing a search method of the first aspect of the present invention at runtime.
  • the storage medium of the embodiment of the present invention associates the search result with the original webpage when the search engine provides the search result to the client, and also displays the original webpage of the search result.
  • the original webpage can also be displayed in the top position, which provides a more convenient way to obtain the original webpage, which makes the client can easily access the original webpage, increases the probability that the original webpage is clicked, and improves the customer.
  • the degree of emphasis on the original page by the end user, without modifying the overall search result ordering, is convenient for the user to emphasize the respect and action support of the search engine for the original page.
  • the user often reads the original page. Discover more relevant content and knowledge. For example, for blogs, the original author's blog site will contain more domain-related knowledge and articles to enhance the user experience.
  • FIG. 1 is a flow chart of a search method in accordance with one embodiment of the present invention.
  • FIG. 2(a) is a flowchart of a search method according to another embodiment of the present invention.
  • FIG. 2(b) is a schematic diagram showing a client displaying a search result according to an embodiment of the present invention
  • FIG. 3 is a flowchart of obtaining a correspondence between a webpage and an original webpage according to an embodiment of the present invention
  • FIG. 4 is a flowchart of obtaining a correspondence between a webpage and an original webpage according to another embodiment of the present invention.
  • FIG. 5 is a flowchart of obtaining a correspondence between a webpage and an original webpage according to still another embodiment of the present invention.
  • FIG. 6 is a flow chart of a search method according to still another embodiment of the present invention.
  • FIG. 7 is a schematic structural diagram of a search system according to an embodiment of the present invention.
  • FIG. 8 is a schematic structural diagram of a search engine according to an embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of a search engine according to another embodiment of the present invention.
  • FIG. 10 is a schematic structural diagram of a search engine according to still another embodiment of the present invention.
  • FIG. 11 is a schematic structural diagram of a client according to an embodiment of the present invention. detailed description
  • the present invention provides a search method, a system, a search engine, and a client, which can display the associated original webpage in the search result, and provide the user with convenient access to the original page without modifying the sorting of the search result.
  • the way, that is, the user is also convenient to emphasize the search engine's respect for the original page and support for action.
  • a search method includes the following steps: a search engine acquires a search term sent by a client; a search engine obtains a plurality of search results according to the search term, and respectively obtains original webpages corresponding to at least some of the plurality of search results; the search engine The original webpage corresponding to the plurality of search results and at least part of the search results is provided to the client and presented through the client.
  • FIG. 1 is a flow chart of a search method in accordance with one embodiment of the present invention.
  • the search method includes the following steps:
  • the search engine obtains the search term sent by the client.
  • the client is a PC (personal computer) or a mobile terminal, wherein the mobile terminal may be a hardware with various operating systems such as a smart phone, a tablet, a personal digital assistant, an e-book, and the like. device.
  • the client can access the search engine website through the browser application and send the search word to the search engine, or send the search word to the search engine through other shortcuts (such as a search bar, etc.), which is not limited herein.
  • the search engine obtains multiple search results according to the search term, and obtains original webpages corresponding to at least some of the plurality of search results.
  • the search result is obtained according to the search term, and the corresponding original webpage is obtained according to each search result, and it should be understood that each search result does not necessarily have the original webpage, if the search If the result has the original webpage, the corresponding original webpage is obtained. If the search result does not have the original webpage, the search result may be the original webpage, or the search engine cannot obtain the corresponding original webpage.
  • the corresponding relationship between the webpage and the original webpage may be calculated according to a preset algorithm, including whether the webpage is an original webpage, etc., and the calculation result is permanently saved, for example, may be stored in a database or
  • the storage device is not limited here, and the storage method can also be diversified.
  • the typical storage format is [webpage A, original webpage 1, original webpage 2, ⁇ ], [webpage B, original webpage 3, the original page 4, ...], etc., do not limit the storage format here.
  • the advantage of storage is that the information is quickly and directly obtained. After the search engine obtains multiple search results, the corresponding original webpage can be directly read.
  • the correspondence between the webpage and the original webpage according to the preset algorithm will be described in detail in the subsequent embodiments.
  • the search result has a original webpage, it corresponds to an original webpage, but in the actual calculation process, there may be more than one original webpage.
  • the possible reason for this is that the webpage content is calculated.
  • the relevance and the original page match there are multiple pages that match the threshold range defined by the algorithm, so Matched multiple original pages. In reality, there will only be one in the original page.
  • the reprinting or piracy method is more elaborate, which interferes with the search engine algorithm system to some extent, but such a situation is allowed to happen, just like the existence of SEO.
  • the (Search Engine Optimization, Search Engine Optimization) method to improve page rank is the same as in the search engine ecosystem.
  • the search engine provides the original webpage corresponding to the plurality of search results and at least part of the search result to the client, and displays the same through the client.
  • the search result when the search engine provides the search result to the client, the search result is associated with the original webpage, and the original webpage of the search result is also displayed.
  • the search result if the search result is ranked higher, it has The original page can also be displayed in the top position when the original page is provided, which provides a more convenient way to get the original page, which makes the client can easily access the original page, increase the probability that the original page is clicked, and improve the client.
  • the user's emphasis on the original webpage without modifying the overall search result ranking, is convenient for the user to emphasize the search engine's respect for the original webpage and the support of the action.
  • the user often finds when reading the original webpage. More relevant content and knowledge, for example, for blogs, the original author's blog site will contain more domain-related knowledge and articles to enhance the user experience.
  • the client can specifically design different presentation effects according to the client configuration.
  • FIG. 2(a) is a flow chart of a search method in accordance with another embodiment of the present invention.
  • the search method includes the following steps:
  • the search engine obtains a search term sent by the client.
  • the search engine obtains multiple search results according to the search term, and obtains original webpages corresponding to at least some of the plurality of search results.
  • the search engine provides the original webpage corresponding to the plurality of search results and at least part of the search result to the client.
  • the client receives multiple search results and presents them to the user.
  • the client adds a prompt identifier to at least part of the search result.
  • the client may parse which search results have the original webpage and which do not have the original webpage, and the client adds the appropriate search result (ie, at least part of the search result) of the original webpage.
  • the prompt mark for example, can be on its left, right, top or bottom.
  • FIG. 2(b) is an exemplary schematic diagram of a client presenting search results according to an embodiment of the present invention, as shown in FIG. 2(b), for each search result having an original webpage, on its right
  • the side passes an icon 100 (and can be other ways, such as bubbles, etc.) to prompt the user to have the original page.
  • the client displays the original webpage of the search result corresponding to the prompt identifier that is triggered to the user.
  • the triggering manner of the prompting flag is various, for example, using a mouse click or performing a gesture click on the screen, or the mouse swiping the prompt mark or the gesture to cross the prompt mark.
  • FIG. 2(b) when the user moves the mouse over the icon, the pop-up box 200 on the right side appears, and the information of the original webpage is presented to the user, and the user is guided to further click, and the user can click to browse. The user can move the mouse off to close the popup box 200.
  • the user can also participate in protecting the original webpage, and can display the operation portal of "reporting non-original version" in the original webpage of the popup box 200, and the user can provide more auxiliary information to the search engine through the participation of the user, and help the pseudo The original page is found out, and the search engine needs to effectively identify whether the user's report is authentic and reliable, dig out valid information, and remove malicious attacks.
  • the original webpage presented to the user may contain more than one, and the user can actually distinguish which is the original original webpage.
  • the original webpage is prompted by the prompt identifier, which is convenient for the user to read, and further improves the user experience.
  • the most important part of the present invention is to obtain the correspondence between the web page and the original web page.
  • the following describes in detail the process of obtaining the correspondence between the web page and the original web page.
  • the search engine crawls the webpages on the Internet in order to calculate the original webpages associated with each webpage, ideally, for each webpage, the content is compared one by one according to the content of the webpage, and the webpage is almost identical in content. , can be considered as mutual reprinting or piracy, and save these same web pages to a list, and then calculate the time of the webpages in the list to be included by the search engine crawler. For example, a simple rule is the earliest included. The highest probability of a web page is the original page, because in theory the search engine crawler can quickly crawl the newly created web content.
  • FIG. 3 is a flowchart of obtaining a correspondence between a webpage and an original webpage according to an embodiment of the present invention. Specifically, as shown in FIG. 3, the method includes the following steps:
  • the search engine crawls multiple web pages.
  • a search engine can obtain multiple web pages on the Internet through a web crawler.
  • the search engine separately extracts feature values of multiple webpages according to a classification algorithm.
  • the search engine classifies the plurality of webpages according to the feature values of the plurality of webpages to establish a plurality of classification sets. Specifically, after all the web pages are classified by the classification algorithm, the larger the number of the classification sets, the better, and the fewer the number of web pages in each classification set, the better. Ideally, only pages with very similar web content will be classified as one Class collection.
  • the classification algorithm can use the more mature machine learning algorithm
  • the machine learning algorithm is a general algorithm, and there are many different deformation algorithms, for example, KNN (k-Nearest Neighbor algorithm, K nearest neighbor node algorithm),
  • KNN k-Nearest Neighbor algorithm, K nearest neighbor node algorithm
  • the basic idea of the algorithm is: After giving a new text, consider the piece of text that is closest (most similar) to the new text in the training text set, and determine the category to which the new text belongs according to the category to which the piece of text belongs; For example, support vector machine and neural network algorithms are widely used in text classification systems. A basic feature of such algorithms is to vectorize the calculation results of web content, and then mathematically compare the differences between the two vectors.
  • webpage content it often contains a variety of information, which may include pictures, videos, etc., so it is necessary to comprehensively consider various factors when comparing content similarity between webpages.
  • the search engine separately analyzes the webpages in each of the classification sets to obtain the original webpages in each of the classification collections.
  • FIG. 4 is a flowchart of obtaining a correspondence between a webpage and an original webpage according to another embodiment of the present invention.
  • the search engine crawls multiple web pages.
  • the search engine separately extracts feature values of multiple webpages according to a classification algorithm.
  • the search engine classifies the plurality of web pages according to the feature values of the plurality of web pages to establish a plurality of classification sets. S404.
  • the search engine separately acquires a search engine crawl time of the webpage in each category collection.
  • the search engine obtains the original webpage in each category collection according to the search engine crawl time.
  • the webpage with the oldest crawl time is usually used as the original webpage.
  • FIG. 5 is a flowchart of obtaining a correspondence between a webpage and an original webpage according to still another embodiment of the present invention.
  • the search engine crawls multiple web pages.
  • the search engine separately extracts feature values of the plurality of web pages according to the classification algorithm.
  • the search engine classifies the plurality of web pages according to the feature values of the plurality of web pages to establish a plurality of classification sets.
  • the search engine obtains the search engine crawl time of the webpage in each category collection.
  • the search engine acquires reliability information of the webpage in each category collection.
  • the search engine obtains the original webpage according to the search engine crawl time and reliability information.
  • the reliability information includes one or more of source information of the webpage, quality information of the webpage, and anti-cheat judgment information.
  • the webpage of the regular website is generally more reliable and of high quality.
  • the reliability of the webpage is higher, and the reliability of the webpage for judging cheating is low. Therefore, comprehensively considering the reliability information and the crawling time can increase the accuracy of the original webpage.
  • the search engine After analyzing the webpage and obtaining the corresponding original webpage, in an embodiment of the present invention, the search engine establishes an association relationship between the webpage in the classification set and the original webpage, to obtain at least part of the search result according to at least part of the search result and the association relationship.
  • the original page corresponding to the search result is established.
  • Figure 6 is a flow chart of a search method in accordance with yet another embodiment of the present invention.
  • the search method includes the following steps:
  • the search engine crawls multiple web pages.
  • the search engine separately extracts feature values of multiple webpages according to a classification algorithm.
  • the search engine classifies the plurality of webpages according to the feature values of the plurality of webpages to establish a plurality of classification sets.
  • the search engine analyzes the web pages in each of the classified collections to obtain the original web pages in each of the classified collections.
  • the search engine establishes a relationship between a webpage in each category collection and an original webpage in the category collection.
  • the search engine obtains a search term sent by the client.
  • the search engine obtains multiple search results according to the search term, and obtains at least a part of the original webpage corresponding to the search result according to at least some of the plurality of search results and the association relationship.
  • the search engine provides the original webpage corresponding to the plurality of search results and at least part of the search result to the client, and displays the same through the client.
  • the search method of the embodiment of the present invention obtains the original webpage corresponding to the search result according to the established association relationship, which is fast and convenient.
  • an embodiment of the present invention also proposes a search system.
  • a search system comprising: a client and a search engine, wherein: a client is configured to send a search term to a search engine; a search engine is configured to obtain a search term, and obtain a plurality of search results according to the search term, and obtain multiple search results respectively
  • the original webpage corresponding to at least some of the search results, and the original webpage corresponding to the plurality of search results and at least some of the search results are provided to the client, wherein the client displays the content sent by the search engine.
  • Figure 7 is a block diagram showing the structure of a search system in accordance with one embodiment of the present invention.
  • the search system includes: a client 10 and a search engine 20.
  • the client 10 is configured to send a search term to the search engine 20.
  • the customer The terminal 10 is a PC or a mobile terminal, wherein the mobile terminal can be a hardware device having various operating systems, such as a smart phone, a tablet computer, a personal digital assistant, an e-book, etc., and the client 10 can access the search engine 20 through a browser application.
  • the website sends the search term to the search engine 20, and can also send the search term to the search engine 20 through other shortcuts (such as a search bar, etc.), which is not limited herein.
  • the search engine 20 is configured to acquire a search term, and obtain a plurality of search results according to the search term, and respectively obtain an original webpage corresponding to at least some of the plurality of search results, and an original version that corresponds the plurality of search results to at least part of the search results.
  • the web page is provided to the client 10, wherein the client 10 presents the content sent by the search engine 20.
  • each search result does not necessarily have an original webpage. If the search result has the original webpage, the corresponding original webpage is obtained. If the search result does not have the original webpage, the search result may be the original webpage, or the search engine may not obtain the corresponding original webpage.
  • the corresponding relationship between the webpage and the original webpage may be calculated according to a preset algorithm, including whether the webpage is an original webpage, etc., and the calculation result is permanently saved, for example, may be stored in a database.
  • the storage device is not limited here, and the storage method can also be diversified.
  • the typical storage format is [webpage A, original webpage 1, original webpage 2, ⁇ ], [webpage B, original edition Page 3, original page 4, -], etc., the storage format is not limited here.
  • the advantage of storage is that the information is quickly and directly obtained.
  • the search engine 20 obtains multiple search results the corresponding original webpage can be directly read.
  • the correspondence between the webpage and the original webpage according to the preset algorithm will be described in detail in the subsequent embodiments.
  • the search result has a original webpage, it corresponds to an original webpage, but in the actual calculation process, there may be more than one original webpage.
  • the possible reason for this is that the webpage content is calculated.
  • the relevance and the original page match there are multiple pages that match the threshold range defined by the algorithm, so multiple original pages are matched. In reality, there will only be one in the original page. In this case, the reprinting or piracy method is more elaborate, which interferes with the algorithm system of the search engine 20 to a certain extent, but such a situation is allowed.
  • the search result when the search engine provides the search result to the client, the search result is associated with the original webpage, and the original webpage of the search result is also displayed.
  • the search result if the search result is ranked higher, it has The original page can also be displayed in the top position when the original page is provided, which provides a more convenient way to get the original page, which makes the client can easily access the original page, increase the probability that the original page is clicked, and improve the client.
  • the user's emphasis on the original webpage without modifying the overall search result ranking, is convenient for the user to emphasize the search engine's respect for the original webpage and the support of the action.
  • the user often finds when reading the original webpage.
  • the client 10 can specifically design different presentation effects according to the configuration of the client 10, and a simple presentation manner is briefly described below.
  • the client 10 is further configured to: receive a plurality of search results and display the same to the user, and add a prompt identifier to at least part of the search results, and when the prompt identifier is triggered, The triggered web page indicates that the original web page of the corresponding search result is displayed to the user.
  • FIG. 2(b) is an exemplary schematic diagram of a client presenting search results according to an embodiment of the present invention, as shown in FIG. 2(b), for each search result having a webpage of the original, on its right
  • the side passes an icon 100 (and can be other ways, such as bubbles, etc.) to prompt the user to have the original page.
  • the triggering manner of the prompting flag is various, for example, using a mouse click or a gesture click on the screen, or a mouse over the prompt mark or a gesture to cross the prompt mark.
  • a mouse click or a gesture click on the screen or a mouse over the prompt mark or a gesture to cross the prompt mark.
  • FIG. 2(b) when the user moves the mouse over the icon, the pop-up box 200 on the right side appears, and the information of the original webpage is presented to the user, and the user is guided to further click, and the user can click to browse. The user can move the mouse off to close the popup box 200.
  • the user can also participate in the protection of the original webpage, and the operation portal of "reporting non-original" can be displayed in the original webpage of the popup box 200, and the auxiliary information can be provided to the search engine through the participation of the user.
  • the original webpage presented to the user may contain more than one, and the user can actually distinguish which is the original original webpage.
  • the original webpage is prompted by the prompt identifier, which is convenient for the user to read, and further improves the user experience.
  • the search engine 20 is further configured to: crawl a plurality of web pages, respectively extract feature values of the plurality of web pages according to the classification algorithm, and classify the plurality of web pages according to the feature values of the plurality of web pages.
  • a plurality of classification sets are created, and web pages in each classification set are separately analyzed to obtain original web pages in each classification set.
  • the classification algorithm can use the more mature machine learning algorithm.
  • the machine learning algorithm is a general algorithm, and there are many different deformation algorithms.
  • KNN the basic idea of the algorithm is: After given new text, Consider the K-text of the closest (most similar) distance to the new text in the training text set, and determine the category to which the new text belongs according to the category to which the K-text belongs; for example, the support vector machine and the neural network algorithm in the text classification system It is also widely used.
  • a basic feature of such algorithms is to vectorize the calculation results of web content and then from mathematics. Compare the differences between the two vectors.
  • webpage content it often contains a variety of information, which may include pictures, videos, etc., so it is necessary to comprehensively consider various factors when comparing content similarity between webpages.
  • the search engine 20 is further configured to: separately acquire the search engine 20 crawling time of the webpage in each of the classified sets, and obtain the original webpage in each of the classified sets according to the crawling time of the search engine 20.
  • the webpage with the oldest crawl time is usually used as the original webpage.
  • the search engine 20 is further configured to: obtain reliability information of the webpage in each of the classification sets to obtain the original webpage according to the search engine 20 crawling time and reliability information.
  • the reliability information includes one or more of source information of the web page, quality information of the web page, and anti-cheat judgment information.
  • the webpage of the regular website is generally more reliable and of high quality. The reliability of the webpage is higher, and the reliability of the webpage for judging cheating is low. Therefore, comprehensively considering the reliability information and the crawling time can increase the accuracy of the original webpage.
  • the search engine 20 is further configured to: after acquiring the original webpages in the plurality of classification sets, establish an association relationship between the webpages in each of the classification collections and the original webpages in the classification collection, The original webpage corresponding to at least part of the search result is obtained according to at least part of the search result and the association relationship. Therefore, obtaining the original webpage corresponding to the search result according to the established association relationship is fast and convenient.
  • an embodiment of the present invention also proposes a search engine.
  • a search engine comprising: a search word obtaining module, configured to obtain a search word sent by a client; a search result obtaining module, configured to obtain a plurality of search results according to the search term, and obtain at least some search results of the plurality of search results respectively Corresponding original webpage; a search result sending module, configured to provide a plurality of search results and at least part of the search results corresponding to the original webpage to the client, and display through the client.
  • FIG. 8 is a schematic structural diagram of a search engine according to an embodiment of the present invention.
  • the search engine includes: a search term acquisition module 210, a search result acquisition module 220, and a search result transmission module 230.
  • the search term acquisition module 210 is configured to acquire a search term sent by the client.
  • the search result obtaining module 220 is configured to obtain a plurality of search results according to the search term, and respectively obtain original webpages corresponding to at least some of the plurality of search results. More specifically, after the search term acquisition module 210 obtains the search term, the search result obtaining module 220 obtains a plurality of corresponding search results according to the search term, and respectively obtains corresponding original webpages according to each search result, and if the search result has an original webpage, Then, the corresponding original webpage is obtained. If the search result does not have the original webpage, the search result may be the original webpage, or the search engine may not obtain the corresponding original webpage.
  • the search result has a original webpage, it corresponds to an original webpage, but in the actual calculation process, there may be more than one original webpage.
  • the possible reason for this is that the webpage content relevance and the original version are calculated.
  • the page matches there are multiple pages that match the threshold range defined by the algorithm, thus matching multiple original pages.
  • the reprinting or piracy method is more elaborate, which interferes with the search engine algorithm system to some extent, but such a situation is allowed.
  • the search result sending module 230 is configured to provide a plurality of search results and at least a part of the search results corresponding to the original webpage to the client for presentation by the client.
  • the search engine of the embodiment of the present invention associates the search result with the original webpage when the search result is provided to the client, so that the client also displays the original webpage of the search result, on the one hand, if the search result is ranked higher.
  • the original webpage can also be displayed in the top position, which provides a more convenient way to obtain the original webpage, so that the client can easily access the original webpage, increasing the probability that the original webpage is clicked, and improving the probability.
  • the degree of emphasis on the original page by the client user without modifying the overall search result ordering, is convenient for the user to emphasize the respect and action support of the search engine for the original page.
  • the user reads the original page, More relevant content and knowledge can be found. For example, for blogs, the original author's blog site will contain more domain-related knowledge and articles to enhance the user experience.
  • FIG. 9 is a schematic structural diagram of a search engine according to another embodiment of the present invention.
  • the search engine includes: a search term acquisition module 210, a search result acquisition module 220, a search result sending module 230, a webpage crawling module 240, a webpage classification module 250, and a webpage analysis module 260.
  • the webpage crawling module 240 is configured to capture a plurality of webpages, for example, a plurality of webpages on the Internet can be obtained through a web crawler.
  • the webpage classification module 250 is configured to respectively extract feature values of a plurality of webpages according to a classification algorithm, and classify the plurality of webpages according to the feature values of the plurality of webpages to establish a plurality of classification sets. More specifically, after all the web pages are classified by the classification algorithm, the larger the number of classification sets, the better, and the fewer the number of web pages in each classification set, the better. Ideally, only pages that have very similar web content are grouped into a single collection. Among them, the classification algorithm can use the more mature machine learning algorithm.
  • the machine learning algorithm is a general-purpose algorithm, and there are many different deformation algorithms.
  • KNN the basic idea of the algorithm is: After given new text, Consider the K-text of the closest (most similar) distance to the new text in the training text set, and determine the category to which the new text belongs according to the category to which the K-text belongs; for example, the support vector machine and the neural network algorithm in the text classification system It is also widely used.
  • a basic feature of such algorithms is to vectorize the calculation results of web content and then mathematically compare the differences between the two vectors. It should be understood that for web content, it often contains a variety of information, including pictures, videos, etc., so it is necessary to consider various factors when comparing the similarity between web pages.
  • the original webpage analysis module 260 is configured to separately analyze webpages in each sorting set to obtain each category. The original page in the collection.
  • the original webpage analysis 260 module is further configured to: separately acquire a search engine crawling time of the webpage in each of the classified collections, and obtain an original webpage in each of the classified collections according to the search engine crawling time. .
  • the webpage with the oldest crawl time is usually used as the original webpage.
  • the original webpage analysis module 260 is further configured to: separately obtain reliability information of the webpage in each of the classification sets, to acquire each classification set according to the search engine crawling time and reliability information.
  • Original page the reliability information includes one or more of source information of the web page, quality information of the web page, and anti-cheat judgment information.
  • the reliability information includes one or more of source information of the web page, quality information of the web page, and anti-cheat judgment information.
  • the webpage of the regular website is generally more reliable and of high quality.
  • the reliability of the webpage is higher, and the reliability of the webpage for judging cheating is low. Therefore, comprehensively considering the reliability information and the crawling time can increase the accuracy of the original webpage.
  • FIG. 10 is a schematic structural diagram of a search engine according to still another embodiment of the present invention.
  • the search engine includes: a search term acquisition module 210, a search result acquisition module 220, a search result sending module 230, a webpage crawling module 240, a webpage classification module 250, a original webpage analysis module 260, and an association relationship establishing module 270. .
  • the association relationship establishing module 270 is configured to: after the original webpage analysis module 260 acquires the original webpages in the plurality of classification sets, establish an association relationship between the webpages in each of the classification collections and the original webpages in the classification collection, according to At least some of the search results and associations obtain the original webpage corresponding to at least part of the search results.
  • the association relationship is permanently saved.
  • it can be stored in a database or in a file system.
  • the storage device is not limited herein, and the storage manner can also be diversified.
  • a typical storage format is [Webpage A, original version. Page 1, original page 2, ...:
  • the advantage of storage is that the information is quickly and directly obtained. After the search engine obtains multiple search results, the corresponding original webpage can be directly read.
  • an embodiment of the present invention also proposes a client.
  • a client comprising: a search term sending module, configured to send a search term to a search engine; a search result obtaining module, configured to obtain a plurality of search results sent by the search engine according to the search term, and at least some search results of the plurality of search results Corresponding original webpage; and a search result presentation module, configured to display a plurality of search results and an original webpage corresponding to at least part of the search results.
  • FIG. 11 is a schematic structural diagram of a client according to an embodiment of the present invention.
  • the client includes: a search term sending module 110, a search result obtaining module 120, and a search result presenting module 130.
  • the client is a PC or a mobile terminal
  • the mobile terminal may be a hardware device having various operating systems such as a smart phone, a tablet computer, a personal digital assistant, an e-book, and the like.
  • the search term sending module 110 is configured to send a search term to the search engine. More specifically, the search term sending module 110 may access the search engine website through the browser application and send the search term to the search engine, or may use other shortcuts. Ways (such as search bar, etc.) send search terms to search engines, which are not limited here.
  • the search result obtaining module 120 is configured to obtain a plurality of search results sent by the search engine according to the search term and an original webpage corresponding to at least part of the plurality of search results.
  • the search result presentation module 130 is configured to present a plurality of search results and an original web page corresponding to at least part of the search results.
  • the client of the embodiment of the present invention can obtain the search result sent by the search engine and the original webpage, and display the original webpage of the search result.
  • the original webpage can also be used when the original webpage has the original webpage. Displayed in the top position, providing a more convenient way to get the original page, so that the client can easily access the original page, increase the probability of the original page being clicked, and improve the importance of the client user to the original page.
  • users read the original webpage they often find more relevant content and knowledge. For example, for blogs, the original author's blog site will contain more domain-related knowledge and articles to enhance the user experience.
  • the client may specifically design different presentation effects according to the client configuration.
  • the search result presentation module 130 is further configured to: receive multiple search results and display them to The user adds a prompt identifier to at least part of the search result, and when the prompt identifier is triggered, displays the original webpage of the search result corresponding to the triggered prompt identifier to the user.
  • the search result presentation module 130 can parse which search results have the original webpage and which do not have the original webpage, and add the prompting flag at the appropriate position of the search result (ie, at least part of the search result) of the original webpage, for example, Its left side, right side, top or bottom.
  • the prompting flag is various, for example, using a mouse click or a gesture click on the screen, or a mouse crossing the prompt mark or a gesture to cross the prompt mark. For example, as shown in FIG.
  • the pop-up box 200 on the right side appears, and the information of the original webpage is presented to the user, and the user is guided to further click, and the user can click to browse.
  • the user can move the mouse off to close the popup box 200.
  • the user can also participate in protecting the original webpage, and can display the operation portal of "reporting non-original version" in the original webpage of the popup box 200, and the user can provide more auxiliary information to the search engine through the participation of the user, and help the pseudo The original page is found out, and the search engine needs to effectively identify whether the user's report is authentic and reliable, dig out valid information, and remove malicious attacks.
  • the original webpage presented to the user may contain more than one, and the user can actually distinguish which is the original original webpage.
  • the original webpage is prompted by the prompt identifier to facilitate the user to read, thereby further improving the user experience.
  • the present invention also proposes an application.
  • the application is for performing a search method as described in an embodiment of the present invention at runtime.
  • the application program associates the search result with the original webpage when the search engine provides the search result to the client, and also displays the original webpage of the search result, on the one hand, if the search result is ranked higher.
  • the original webpage can also be displayed in the top position, which provides a more convenient way to obtain the original webpage, so that the client can easily access the original webpage, increasing the probability that the original webpage is clicked, and improving the probability.
  • the degree of emphasis on the original page by the client user, without modifying the overall search result ordering, is convenient for the user to emphasize the respect and action support of the search engine for the original page.
  • the user reads the original page More relevant content and knowledge can be found. For example, for blogs, the original author's blog site will contain more domain-related knowledge and articles to enhance the user experience.
  • the present invention also provides a storage medium for storing an application for executing a search method as described in the embodiment of the present invention at runtime.
  • the search engine when the search engine provides the search result to the client, the search result is associated with the original webpage, and the original webpage of the search result is also displayed.
  • the original webpage can also be displayed in the top position, which provides a more convenient way to obtain the original webpage, so that the client can easily access the original webpage, increasing the probability that the original webpage is clicked, and improving the probability.
  • the degree of emphasis on the original page by the client user without modifying the overall search result ordering, is convenient for the user to emphasize the respect and action support of the search engine for the original page.
  • the user reads the original page, More relevant content and knowledge can be found. For example, for blogs, the original author's blog site will contain more domain-related knowledge and articles to enhance the user experience.
  • portions of the invention may be implemented in hardware, software, firmware or a combination thereof.
  • multiple steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system.
  • a suitable instruction execution system For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or combination of the following techniques well known in the art: having logic gates for implementing logic functions on data signals Discrete logic circuits, application specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.
  • the description of the terms “one embodiment”, “some embodiments”, “example”, “specific example”, or “some examples” and the like means a specific feature described in connection with the embodiment or example.
  • a structure, material or feature is included in at least one embodiment or example of the invention.
  • the schematic representation of the above terms does not necessarily mean the same embodiment or example.
  • the particular features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computing Systems (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Algebra (AREA)
  • Probability & Statistics with Applications (AREA)
  • Information Transfer Between Computers (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本发明提出一种搜索方法、系统、搜索引擎和客户端。其中,搜索方法包括:搜索引擎获取客户端发送的搜索词;搜索引擎根据搜索词获取多个搜索结果,并分别获取多个搜索结果中至少部分搜索结果对应的原版网页;搜索引擎将多个搜索结果和至少部分搜索结果对应的原版网页提供至客户端,并通过客户端展现。本发明实施例的方法,提供了一种更为便捷获得原版网页的方式,使得客户端可以轻松访问原版网页,增加了原版网页被点击的概率,提高了客户端用户对于原版网页的重视程度。

Description

搜索方法、 系统、 搜索引擎和客户端
技术领域
本发明涉及搜索技术领域, 尤其涉及一种搜索方法、 系统、 搜索引擎和客户端。 背景技术
目前, 搜索引擎返回的搜索结果页面中的多个搜索结果是经过了复杂的算法处理后排 序所得, 需综合考虑多种因素决定排序的优先级。 目前存在的问题是: 作为原版网页内容, 由于其网站本身的搜索排名等综合因素可能不够好, 导致原版网页无法在搜索结果的前几 条中出现, 反而一些转载的页面却排到了前面几条, 同时现在很多的盗版和转载的网页并 没有注明内容的原出处, 原版网页的访问途径减少, 被点击的概率低。 发明内容
本发明旨在至少解决上述技术问题之一。
为此,本发明的第一个目的在于提出一种搜索方法。该方法提供了一种更为便捷获得 原版网页的方式, 使得客户端可以轻松访问原版网页, 增加了原版网页被点击的概率, 提高了客户端用户对于原版网页的重视程度。
本发明的第二个目的在于提出一种搜索系统。
本发明的第三个目的在于提出一种搜索引擎。
本发明的第四个目的在于提出一种客户端。
本发明的第五个目的在于提出一种应用程序。
本发明的第六个目的在于提出一种存储介质。
为了实现上述目的, 本发明第一方面实施例的搜索方法, 包括以下步骤: 搜索引擎获 取客户端发送的搜索词; 所述搜索引擎根据所述搜索词获取多个搜索结果, 并分别获取所 述多个搜索结果中至少部分搜索结果对应的原版网页; 所述搜索引擎将所述多个搜索结果 和所述至少部分搜索结果对应的原版网页提供至所述客户端, 并通过所述客户端展现。
本发明实施例的搜索方法, 搜索引擎向客户端提供搜索结果时, 将搜索结果和其原版 网页关联, 对搜索结果的原版网页也进行展现, 一方面如果搜索结果的排名靠前, 在其具 有原版网页时原版网页也可展现在排名靠前的位置, 提供了一种更为便捷获得原版网页 的方式, 使得客户端可以轻松访问原版网页, 增加了原版网页被点击的概率, 提高了 客户端用户对于原版网页的重视程度, 在不修改整体搜索结果排序的情况下, 即方便了 用户又强调了搜索引擎对于原版网页的尊重和行动上的支持, 另一方面用户阅读原版网页 时往往可发现更多相关的内容和知识, 例如, 对于博客, 原作者的博客站点中会包含更多 领域相关的知识和文章, 提升用户体验。
为了实现上述目的, 本发明第二方面实施例的搜索系统, 包括: 客户端和搜索引擎, 其中, 所述客户端, 用于向所述搜索引擎发送搜索词, 并对所述搜索引擎返回的内容进行 展示; 所述搜索引擎, 用于获取所述搜索词, 并根据所述搜索词获取多个搜索结果, 并分 别获取所述多个搜索结果中至少部分搜索结果对应的原版网页, 以及将所述多个搜索结果 和所述至少部分搜索结果对应的原版网页提供至所述客户端。
本发明实施例的搜索系统, 搜索引擎向客户端提供搜索结果时, 将搜索结果和其原版 网页关联, 对搜索结果的原版网页也进行展现, 一方面如果搜索结果的排名靠前, 在其具 有原版网页时原版网页也可展现在排名靠前的位置, 提供了一种更为便捷获得原版网页 的方式, 使得客户端可以轻松访问原版网页, 增加了原版网页被点击的概率, 提高了 客户端用户对于原版网页的重视程度, 在不修改整体搜索结果排序的情况下, 即方便了 用户又强调了搜索引擎对于原版网页的尊重和行动上的支持, 另一方面用户阅读原版网页 时往往可发现更多相关的内容和知识, 例如, 对于博客, 原作者的博客站点中会包含更多 领域相关的知识和文章, 提升用户体验。
为了实现上述目的, 本发明第三方面实施例的搜索引擎, 包括: 搜索词获取模块, 用 于获取客户端发送的搜索词; 搜索结果获取模块, 用于根据所述搜索词获取多个搜索结果, 并分别获取所述多个搜索结果中至少部分搜索结果对应的原版网页; 搜索结果发送模块, 用于将所述多个搜索结果和所述至少部分搜索结果对应的原版网页提供至所述客户端, 以 供所述客户端进行展现。
本发明实施例的搜索引擎, 在向客户端提供搜索结果时, 将搜索结果和其原版网页关 联, 使得客户端对搜索结果的原版网页也进行展现, 一方面如果搜索结果的排名靠前, 在 其具有原版网页时原版网页也可展现在排名靠前的位置,提供了一种更为便捷获得原版网 页的方式, 使得客户端可以轻松访问原版网页, 增加了原版网页被点击的概率, 提高 了客户端用户对于原版网页的重视程度, 在不修改整体搜索结果排序的情况下, 即方便 了用户又强调了搜索引擎对于原版网页的尊重和行动上的支持, 另一方面用户阅读原版网 页时往往可发现更多相关的内容和知识, 例如, 对于博客, 原作者的博客站点中会包含更 多领域相关的知识和文章, 提升用户体验。
为了实现上述目的, 本发明第四方面实施例的客户端, 包括: 搜索词发送模块, 用于 向搜索引擎发送搜索词; 搜索结果获取模块, 用于获取所述搜索引擎根据所述搜索词发送 的多个搜索结果和所述多个搜索结果中至少部分搜索结果对应的原版网页; 以及搜索结果 展现模块, 用于展现所述多个搜索结果和所述至少部分搜索结果对应的原版网页。
本发明实施例的客户端, 可获取搜索引擎发送的搜索结果和其原版网页, 对搜索结果 的原版网页进行展现, 一方面如果搜索结果的排名靠前, 在其具有原版网页时原版网页也 可展现在排名靠前的位置, 提供了一种更为便捷获得原版网页的方式, 使得客户端可以 轻松访问原版网页, 增加了原版网页被点击的概率, 提高了客户端用户对于原版网页 的重视程度, 另一方面用户阅读原版网页时往往可发现更多相关的内容和知识, 例如, 对 于博客, 原作者的博客站点中会包含更多领域相关的知识和文章, 提升用户体验。
为了实现上述目的, 本发明第五方面实施例的应用程序, 用于在运行时执行本发明第 一方面实施例的搜索方法。
本发明实施例的应用程序, 在搜索引擎向客户端提供搜索结果时, 将搜索结果和其原 版网页关联, 对搜索结果的原版网页也进行展现, 一方面如果搜索结果的排名靠前, 在其 具有原版网页时原版网页也可展现在排名靠前的位置, 提供了一种更为便捷获得原版网 页的方式, 使得客户端可以轻松访问原版网页, 增加了原版网页被点击的概率, 提高 了客户端用户对于原版网页的重视程度, 在不修改整体搜索结果排序的情况下, 即方便 了用户又强调了搜索引擎对于原版网页的尊重和行动上的支持, 另一方面用户阅读原版网 页时往往可发现更多相关的内容和知识, 例如, 对于博客, 原作者的博客站点中会包含更 多领域相关的知识和文章, 提升用户体验。
为了实现上述实施例, 本发明第六方面实施例的存储介质, 用于存储应用程序, 所述 应用程序用于在运行时执行本发明第一方面实施例的搜索方法。
本发明实施例的存储介质, 在搜索引擎向客户端提供搜索结果时, 将搜索结果和其原 版网页关联, 对搜索结果的原版网页也进行展现, 一方面如果搜索结果的排名靠前, 在其 具有原版网页时原版网页也可展现在排名靠前的位置, 提供了一种更为便捷获得原版网 页的方式, 使得客户端可以轻松访问原版网页, 增加了原版网页被点击的概率, 提高 了客户端用户对于原版网页的重视程度, 在不修改整体搜索结果排序的情况下, 即方便 了用户又强调了搜索引擎对于原版网页的尊重和行动上的支持, 另一方面用户阅读原版网 页时往往可发现更多相关的内容和知识, 例如, 对于博客, 原作者的博客站点中会包含更 多领域相关的知识和文章, 提升用户体验。
本发明附加的方面和优点将在下面的描述中部分给出, 部分将从下面的描述中变得明 显, 或通过本发明的实践了解到。 附图说明
本发明上述的和 /或附加的方面和优点从下面结合附图对实施例的描述中将变得明显和 容易理解, 其中,
图 1是根据本发明一个实施例的搜索方法的流程图;
图 2 (a) 是根据本发明另一个实施例的搜索方法的流程图;
图 2 (b) 是根据本发明实施例的客户端展现搜索结果的一种示例性示意图; 图 3是本发明实施例的获取网页与原版网页对应关系的流程图;
图 4是本发明另一实施例的获取网页与原版网页对应关系的流程图;
图 5是本发明又一实施例的获取网页与原版网页对应关系的流程图;
图 6是本发明又一个实施例的搜索方法的流程图;
图 7是根据本发明一个实施例的搜索系统的结构示意图;
图 8是根据本发明一个实施例的搜索引擎的结构示意图;
图 9是根据本发明另一个实施例的搜索引擎的结构示意图;
图 10是根据本发明又一个实施例的搜索引擎的结构示意图;
图 11是根据本发明一个实施例的客户端的结构示意图。 具体实施方式
下面详细描述本发明的实施例, 所述实施例的示例在附图中示出, 其中自始至终相同 或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。 下面通过参考附图描 述的实施例是示例性的, 仅用于解释本发明, 而不能理解为对本发明的限制。 相反, 本发 明的实施例包括落入所附加权利要求书的精神和内涵范围内的所有变化、 修改和等同物。
在本发明的描述中, 需要理解的是, 术语"第一"、 "第二"等仅用于描述目的, 而不能理 解为指示或暗示相对重要性。 在本发明的描述中, 需要说明的是, 除非另有明确的规定和 限定, 术语 "相连"、 "连接 "应做广义理解, 例如, 可以是固定连接, 也可以是可拆卸连接, 或一体地连接; 可以是机械连接, 也可以是电连接; 可以是直接相连, 也可以通过中间媒 介间接相连。 对于本领域的普通技术人员而言, 可以具体情况理解上述术语在本发明中的 具体含义。 此外, 在本发明的描述中, 除非另有说明, "多个"的含义是两个或两个以上。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为, 表示包括一个 或更多个用于实现特定逻辑功能或过程的步骤的可执行指令的代码的模块、 片段或部分, 并且本发明的优选实施方式的范围包括另外的实现, 其中可以不按所示出或讨论的顺序, 包括根据所涉及的功能按基本同时的方式或按相反的顺序, 来执行功能, 这应被本发明的 实施例所属技术领域的技术人员所理解。
为了解决原版网页在搜索结果中的排序问题, 也为了增加原版网页的访问途径、 点击 概率, 同时为了方便用户查看原版内容, 鼓励原版内容的创作, 例如对原版博客内容的支 持, 帮助产生更多优质的互联网内容。 为此, 本发明提供了一种搜索方法、 系统、 搜索引 擎和客户端, 可在搜索结果中展现关联的原版网页, 在不修改搜索结果排序的情况下, 为 用户提供了便捷的获取原版页面的途径, 即方便了用户又强调了搜索引擎对于原版网页的 尊重和行动上的支持。
一种搜索方法, 包括以下步骤: 搜索引擎获取客户端发送的搜索词; 搜索引擎根据搜 索词获取多个搜索结果, 并分别获取多个搜索结果中至少部分搜索结果对应的原版网页; 搜索引擎将多个搜索结果和至少部分搜索结果对应的原版网页提供至客户端, 并通过客户 端展现。
图 1是根据本发明一个实施例的搜索方法的流程图。
如图 1所示, 搜索方法包括以下步骤:
5101 , 搜索引擎获取客户端发送的搜索词。
在本发明的一个实施例中, 客户端为 PC (personal computer, 个人计算机) 或移动终 端, 其中, 移动终端可以是智能手机、 平板电脑、 个人数字助理、 电子书等具有各种操作 系统的硬件设备。 具体地, 客户端可通过浏览器类应用程序访问搜索引擎网站并向搜索引 擎发送搜索词, 也可以通过其他快捷途径 (例如搜索栏等) 向搜索引擎发送搜索词, 在此 不做限定。
5102, 搜索引擎根据搜索词获取多个搜索结果, 并分别获取多个搜索结果中至少部分 搜索结果对应的原版网页。
具体地, 搜索引擎获取搜索词之后, 根据搜索词获取对应的多个搜索结果, 再分别根 据每个搜索结果获取对应的原版网页, 还应当理解, 未必每个搜索结果都有原版网页, 如 果搜索结果具有原版网页, 则获取对应的原版网页, 如果搜索结果没有原版网页, 则有可 能该搜索结果即为原版网页, 或者搜索引擎无法获取到对应的原版网页。
其中, 搜索引擎通过网络爬虫抓取网页之后, 可根据预设算法计算网页与原版网页的 对应关系, 包括网页是否为原版网页等, 并对计算结果进行永久保存, 例如, 可存储在数 据库中或文件系统中, 在此不对存储装置做限定, 存储的方式也可多样化, 例如, 典型的 存储格式为 [网页 A, 原版网页 1, 原版网页 2, ···], [网页 B, 原版网页 3, 原版网页 4,…] 等, 在此不对存储格式做限定。 存储的好处是方便信息的快速、 直接获取, 搜索引擎获取 多个搜索结果之后, 可以直接读取对应的原版网页。 根据预设算法计算网页与原版网页的 对应关系将在后续实施例中详细说明。
还应当理解, 理想情况下, 如果搜索结果具有原版网页, 则对应一个原版网页, 但是 在实际的计算过程中, 可能存在一个以上的原版网页, 这种情况出现的可能原因是, 在计 算网页内容相关性和原版网页匹配程度时出现多个符合算法定义的阀值范围的网页, 因此 匹配了多个原版网页。 现实情况中, 原版网页理论上只会有一个, 出现这个情况说明转载 或盗版手段做的比较精细, 一定程度上干扰了搜索引擎的算法系统, 但是这样的情况是允 许发生的, 就像存在 SEO ( Search Engine Optimization, 搜索引擎优化)方法来提高网页排 名一样, 是搜索引擎生态系统中的正常情况。
S 103 , 搜索引擎将多个搜索结果和至少部分搜索结果对应的原版网页提供至客户端, 并通过客户端展现。
本发明实施例的搜索方法, 搜索引擎向客户端提供搜索结果时, 将搜索结果和其原版 网页关联, 对搜索结果的原版网页也进行展现, 一方面如果搜索结果的排名靠前, 在其具 有原版网页时原版网页也可展现在排名靠前的位置, 提供了一种更为便捷获得原版网页的 方式, 使得客户端可以轻松访问原版网页, 增加了原版网页被点击的概率, 提高了客户端 用户对于原版网页的重视程度, 在不修改整体搜索结果排序的情况下, 即方便了用户又强 调了搜索引擎对于原版网页的尊重和行动上的支持, 另一方面用户阅读原版网页时往往可 发现更多相关的内容和知识, 例如, 对于博客, 原作者的博客站点中会包含更多领域相关 的知识和文章, 提升用户体验。
为了更好地展现原版网页, 在客户端可具体地根据客户端配置设计不同的展现效果, 下面简单介绍一种展现方式。
图 2 ( a) 是根据本发明另一个实施例的搜索方法的流程图。
如图 2 ( a) 所示, 搜索方法包括以下步骤:
S201 , 搜索引擎获取客户端发送的搜索词。
S202, 搜索引擎根据搜索词获取多个搜索结果, 并分别获取多个搜索结果中至少部分 搜索结果对应的原版网页。
5203 , 搜索引擎将多个搜索结果和至少部分搜索结果对应的原版网页提供至客户端。
5204, 客户端接收多个搜索结果并展示给用户。
5205, 客户端分别在至少部分搜索结果处添加提示标识。
具体地, 客户端获取搜索引擎发送的搜索结果之后, 可以解析出哪些搜索结果具有原 版网页, 哪些不具有原版网页, 客户端在具有原版网页的搜索结果(即至少部分搜索结果) 的适当位置添加提示标志, 例如, 可以在其左侧、 右侧、 上方或下方。 举例来说, 图 2 (b ) 是根据本发明实施例的客户端展现搜索结果的一种示例性示意图, 如图 2 (b ) 所示, 对于 具有原版网页的每个搜索结果, 在其右侧通过一个图标 100 (还可以是其他的方式, 例如, 气泡等) 提示用户具有原版网页。
5206, 当提示标识被触发时, 客户端将被触发的提示标识对应的搜索结果的原版网页 展示给用户。 具体地, 提示标志的触发方式多式多样, 例如, 使用鼠标点击或者在屏幕上进行手势 点击, 或者鼠标划过该提示标志或者手势划过该提示标志等。 举例来说, 如图 2 ( b), 用 户将鼠标移动到图标上方即出现右侧的弹出式框 200, 将原版网页的信息展现给用户, 引 导用户进行进一步的点击,用户可以点击进行浏览,用户将鼠标移开即可关闭弹出式框 200。 另外, 用户还可以参与到保护原版网页中, 可以在弹出式框 200的原版网页中显示 "举报 非原版" 的操作入口, 通过用户的参与可以给搜索引擎提供更多的辅助信息, 帮助将伪原 版的网页找出来, 搜索引擎还需根据策略来有效识别用户的举报是否真实可靠, 挖掘出有 效的信息, 去除恶意攻击的信息等。
应当理解, 展现给用户的原版网页可能包含多个, 用户一般情况下其实是可以分辨出 哪个才是真正的原版网页。
本发明实施例的搜索方法, 将原版网页通过提示标识进行提示, 方便用户阅读, 进 一步提升用户体验。
为了达到上述实施例所述的目的, 本发明的最重要的部分在于获取网页与原版网页的 对应关系。 下面详细介绍网页与原版网页对应关系的获取过程。
搜索引擎抓取互联网中的网页之后, 为了计算每个网页关联的原版网页, 理想情况下, 对于每个网页, 根据其内容和其它所有网页的内容逐个进行比较, 对于内容上几乎完全一 致的网页, 可以首先认为是相互的转载或盗版, 并且将这些相同的网页都保存到一个列表 中, 然后对列表中的网页计算其被搜索引擎爬虫收录的时间, 例如, 一个简单的规则就是 最早被收录的网页具有最高的概率是原版网页, 因为理论上搜索引擎爬虫可以很快的抓取 新出现的网页内容。
但是, 这种理想情况下的算法计算量巨大, 假设网页数量是 N, 则每个网页计算其关 联原版网页的计算次数就是 N*N, 则将全部网页计算完一遍的计算次数就是 N*N*N,现在 的搜索引擎普遍收集 500亿以上的网页数量, 因此这个巨大的计算量很难完成, 对资源和 时间的消耗也非常巨大。
为了解决这个问题, 使得技术方案具有可实现性, 图 3是本发明实施例的获取网页与 原版网页对应关系的流程图, 具体地, 如图 3所示, 包括以下步骤:
5301 , 搜索引擎抓取多个网页。
例如, 搜索引擎可通过网络爬虫获取互联网上的多个网页。
5302, 搜索引擎根据分类算法分别提取多个网页的特征值。
S303 , 搜索引擎根据多个网页的特征值对多个网页进行分类以建立多个分类集合。 具体地, 通过分类算法对所有网页分类后, 产生分类集合的数量越大越好, 同时每个 分类集合中的网页数量越少越好。 理想情况是, 仅把网页内容非常相近的网页归为一个分 类集合。
其中, 分类算法可运用目前较成熟的机器学习算法, 机器学习算法是一种通用算法, 可有较多不同的变形算法,例如, KNN (k-Nearest Neighbor algorithm, K最邻近结点算法), 该算法的基本思路是: 在给定新文本后, 考虑在训练文本集中与该新文本距离最近 (最相 似) 的 Κ篇文本, 根据这 Κ篇文本所属的类别判定新文本所属的类别; 又如, 支持向量机 和神经网络算法在文本分类系统中应用得也较为广泛, 此类算法一个基本的特点就是把对 网页内容的计算结果向量化, 然后从数学上比较两个向量的差异。
应当理解, 对于网页内容来说, 其往往包含丰富的各种信息, 可能包括图片、 视频等, 因此比较网页之间内容相似度时需要综合考虑各种因素。
S304, 搜索引擎分别对每个分类集合中的网页进行分析, 以获取每个分类集合中的原 版网页。
由此, 可极大地减少计算量, 分类集合中的网页个数越少, 计算量就越少。 还需要说 明的是, S304中的限定并不是说每个分类集合中一定存在原版网页, 也有不存在原版网页 的可能性存在。
图 4是本发明另一实施例的获取网页与原版网页对应关系的流程图。
如图 4所示, 包括以下步骤:
5401 , 搜索引擎抓取多个网页。
5402, 搜索引擎根据分类算法分别提取多个网页的特征值。
5403, 搜索引擎根据多个网页的特征值对多个网页进行分类以建立多个分类集合。 S404, 搜索引擎分别获取每个分类集合中的网页的搜索引擎抓取时间。
S405, 搜索引擎根据搜索引擎抓取时间获取每个分类集合中的原版网页。
例如, 通常将抓取时间最早的网页作为原版网页。
为了增加原版网页的可靠性, 还需要增加可靠性信息。 图 5是本发明又一实施例的获 取网页与原版网页对应关系的流程图。
如图 5所示, 包括以下步骤:
5501 , 搜索引擎抓取多个网页。
5502, 搜索引擎根据分类算法分别提取多个网页的特征值。
5503, 搜索引擎根据多个网页的特征值对多个网页进行分类以建立多个分类集合。
5504, 搜索引擎获取每个分类集合中的网页的搜索引擎抓取时间。
S505, 搜索引擎获取每个分类集合中的网页的可靠性信息。
其中, S505和 S504的执行不分先后顺序。
S506, 搜索引擎根据搜索引擎抓取时间和可靠性信息获取原版网页。 在本发明的一个实施例中, 可靠性信息包括网页的来源信息、 网页的质量信息和反作 弊判断信息中的一种或多种。 具体地, 获取原版网页时, 还需要考虑网页的来源信息、 网 页的质量信息和反作弊判断信息等多种反映网页可靠性的信息, 例如, 通常正规网站的网 页可靠性更高, 质量高的网页可靠性更高, 判断作弊的网页的可靠性低等, 由此, 综合考 虑可靠性信息和抓取时间可增加原版网页的准确性。
在对网页进行分析并获取对应的原版网页之后, 在本发明的一个实施例中, 搜索引擎 建立分类集合中网页与原版网页之间的关联关系, 以根据至少部分搜索结果和关联关系获 取至少部分搜索结果对应的原版网页。
图 6是本发明又一个实施例的搜索方法的流程图。
如图 6所示, 搜索方法包括以下步骤:
5601 , 搜索引擎抓取多个网页。
5602, 搜索引擎根据分类算法分别提取多个网页的特征值。
5603 , 搜索引擎根据多个网页的特征值对多个网页进行分类以建立多个分类集合。
5604, 搜索引擎对每个分类集合中的网页进行分析, 以获取每个分类集合中的原版网 页。
5605, 搜索引擎建立每个分类集合中的网页与该分类集合中的原版网页之间的关联关 系。
5606, 搜索引擎获取客户端发送的搜索词。
5607, 搜索引擎根据搜索词获取多个搜索结果, 并分别根据多个搜索结果中至少部分 搜索结果和关联关系获取至少部分搜索结果对应的原版网页。
5608, 搜索引擎将多个搜索结果和至少部分搜索结果对应的原版网页提供至客户端, 并通过客户端展现。
本发明实施例的搜索方法, 根据建立的关联关系获取搜索结果对应的原版网页, 快捷、 方便。
为了实现上述实施例, 本发明的实施例还提出一种搜索系统。
一种搜索系统, 包括: 客户端和搜索引擎, 其中, 客户端, 用于向搜索引擎发送搜索 词; 搜索引擎, 用于获取搜索词, 并根据搜索词获取多个搜索结果, 并分别获取多个搜索 结果中至少部分搜索结果对应的原版网页, 以及将多个搜索结果和至少部分搜索结果对应 的原版网页提供至客户端, 其中, 客户端对搜索引擎发送的内容进行展现。
图 7是根据本发明一个实施例的搜索系统的结构示意图。
如图 7所示, 搜索系统包括: 客户端 10和搜索引擎 20。
具体地, 客户端 10用于向搜索引擎 20发送搜索词。 在本发明的一个实施例中, 客户 端 10为 PC或移动终端, 其中, 移动终端可以是智能手机、 平板电脑、 个人数字助理、 电 子书等具有各种操作系统的硬件设备, 客户端 10可通过浏览器类应用程序访问搜索引擎 20 网站并向搜索引擎 20发送搜索词, 也可以通过其他快捷途径 (例如搜索栏等) 向搜索 引擎 20发送搜索词, 在此不做限定。
搜索引擎 20用于获取搜索词, 并根据搜索词获取多个搜索结果, 并分别获取多个搜索 结果中至少部分搜索结果对应的原版网页, 以及将多个搜索结果和至少部分搜索结果对应 的原版网页提供至客户端 10, 其中, 客户端 10对搜索引擎 20发送的内容进行展现。
更具体地, 搜索引擎 20获取搜索词之后, 根据搜索词获取对应的多个搜索结果, 再分 别根据每个搜索结果获取对应的原版网页, 还应当理解, 未必每个搜索结果都有原版网页, 如果搜索结果具有原版网页, 则获取对应的原版网页, 如果搜索结果没有原版网页, 则有 可能该搜索结果即为原版网页, 或者搜索引擎无法获取到对应的原版网页。
其中, 搜索引擎 20通过网络爬虫抓取网页之后, 可根据预设算法计算网页与原版网页 的对应关系, 包括网页是否为原版网页等, 并对计算结果进行永久保存, 例如, 可存储在 数据库中或文件系统中, 在此不对存储装置做限定, 存储的方式也可多样化, 例如, 典型 的存储格式为 [网页 A,原版网页 1,原版网页 2, ···], [网页 B,原版网页 3,原版网页 4, -] 等, 在此不对存储格式做限定。 存储的好处是方便信息的快速、 直接获取, 搜索引擎 20获 取多个搜索结果之后, 可以直接读取对应的原版网页。 根据预设算法计算网页与原版网页 的对应关系将在后续实施例中详细说明。
还应当理解, 理想情况下, 如果搜索结果具有原版网页, 则对应一个原版网页, 但是 在实际的计算过程中, 可能存在一个以上的原版网页, 这种情况出现的可能原因是, 在计 算网页内容相关性和原版网页匹配程度时出现多个符合算法定义的阀值范围的网页, 因此 匹配了多个原版网页。 现实情况中, 原版网页理论上只会有一个, 出现这个情况说明转载 或盗版手段做的比较精细, 一定程度上干扰了搜索引擎 20的算法系统, 但是这样的情况是 允许发生的。
本发明实施例的搜索系统, 搜索引擎向客户端提供搜索结果时, 将搜索结果和其原版 网页关联, 对搜索结果的原版网页也进行展现, 一方面如果搜索结果的排名靠前, 在其具 有原版网页时原版网页也可展现在排名靠前的位置, 提供了一种更为便捷获得原版网页 的方式, 使得客户端可以轻松访问原版网页, 增加了原版网页被点击的概率, 提高了 客户端用户对于原版网页的重视程度, 在不修改整体搜索结果排序的情况下, 即方便了 用户又强调了搜索引擎对于原版网页的尊重和行动上的支持, 另一方面用户阅读原版网页 时往往可发现更多相关的内容和知识, 例如, 对于博客, 原作者的博客站点中会包含更多 领域相关的知识和文章, 提升用户体验。 为了更好地展现原版网页, 在客户端 10可具体地根据客户端 10配置设计不同的展现 效果, 下面简单介绍一种展现方式。具体地,在本发明的一个实施例中, 客户端 10还用于: 接收多个搜索结果并展示给用户, 并分别在至少部分搜索结果处添加提示标识, 以及当提 示标识被触发时, 将被触发的提示标识对应的搜索结果的原版网页展示给用户。
更具体地, 客户端 10获取搜索引擎 20发送的搜索结果之后, 可以解析出哪些搜索结 果具有原版网页, 哪些不具有原版网页, 客户端 10在具有原版网页的搜索结果(即至少部 分搜索结果) 的适当位置添加提示标志, 例如, 可以在其左侧、 右侧、 上方或下方。 举例 来说, 图 2 (b) 是根据本发明实施例的客户端展现搜索结果的一种示例性示意图, 如图 2 (b) 所示, 对于具有原版网页的每个搜索结果, 在其右侧通过一个图标 100 (还可以是其 他的方式, 例如, 气泡等) 提示用户具有原版网页。
同样地, 提示标志的触发方式多式多样, 例如, 使用鼠标点击或者在屏幕上进行手势 点击, 或者鼠标划过该提示标志或者手势划过该提示标志等。 举例来说, 如图 2 (b), 用 户将鼠标移动到图标上方即出现右侧的弹出式框 200, 将原版网页的信息展现给用户, 引 导用户进行进一步的点击,用户可以点击进行浏览,用户将鼠标移开即可关闭弹出式框 200。 另外, 用户还可以参与到保护原版网页中, 可以在弹出式框 200的原版网页中显示 "举报 非原版" 的操作入口, 通过用户的参与可以给搜索引擎提供更多的辅助信息。
应当理解, 展现给用户的原版网页可能包含多个, 用户一般情况下其实是可以分辨出 哪个才是真正的原版网页。
本发明实施例的搜索系统, 将原版网页通过提示标识进行提示, 方便用户阅读, 进一 步提升用户体验。
为了达到上述实施例所述的目的, 本发明的最重要的部分在于获取网页与原版网页的 对应关系。 在本发明的一个实施例中, 搜索引擎 20还用于: 抓取多个网页, 并根据分类算 法分别提取多个网页的特征值, 以及根据多个网页的特征值对多个网页进行分类以建立多 个分类集合, 并分别对每个分类集合中的网页进行分析, 以获取每个分类集合中的原版网 页。
更具体地, 通过分类算法对所有网页分类后, 产生分类集合的数量越大越好, 同时每 个分类集合中的网页数量越少越好。 理想情况是, 仅把网页内容非常相近的网页归为一个 分类集合。 其中, 分类算法可运用目前较成熟的机器学习算法, 机器学习算法是一种通用 算法, 可有较多不同的变形算法, 例如, KNN, 该算法的基本思路是: 在给定新文本后, 考虑在训练文本集中与该新文本距离最近(最相似) 的 K篇文本, 根据这 K篇文本所属的 类别判定新文本所属的类别; 又如, 支持向量机和神经网络算法在文本分类系统中应用得 也较为广泛, 此类算法一个基本的特点就是把对网页内容的计算结果向量化, 然后从数学 上比较两个向量的差异。
应当理解, 对于网页内容来说, 其往往包含丰富的各种信息, 可能包括图片、 视频等, 因此比较网页之间内容相似度时需要综合考虑各种因素。
在本发明的一个实施例中, 搜索引擎 20还用于: 分别获取每个分类集合中网页的搜索 引擎 20抓取时间, 并根据搜索引擎 20抓取时间获取每个分类集合中的原版网页。 例如, 通常将抓取时间最早的网页作为原版网页。
为了增加原版网页的可靠性, 还需要增加可靠性信息。 在本发明的一个实施例中, 搜 索引擎 20还用于: 获取每个分类集合中的网页的可靠性信息, 以根据搜索引擎 20抓取时 间和可靠性信息获取原版网页。 在本发明的另一个实施例中, 可靠性信息包括网页的来源 信息、 网页的质量信息和反作弊判断信息中的一种或多种。 具体地, 获取原版网页时, 还 需要考虑网页的来源信息、 网页的质量信息和反作弊判断信息等多种反映网页可靠性的信 息, 例如, 通常正规网站的网页可靠性更高, 质量高的网页可靠性更高, 判断作弊的网页 的可靠性低等, 由此, 综合考虑可靠性信息和抓取时间可增加原版网页的准确性。
在本发明的一个实施例中, 搜索引擎 20还用于: 在获取多个分类集合中的原版网页之 后, 建立每个分类集合中的网页与该分类集合中的原版网页之间的关联关系, 以根据至少 部分搜索结果和关联关系获取至少部分搜索结果对应的原版网页。 由此, 根据建立的关联 关系获取搜索结果对应的原版网页, 快捷、 方便。
为了实现上述实施例, 本发明的实施例还提出一种搜索引擎。
一种搜索引擎, 包括: 搜索词获取模块, 用于获取客户端发送的搜索词; 搜索结果获 取模块, 用于根据搜索词获取多个搜索结果, 并分别获取多个搜索结果中至少部分搜索结 果对应的原版网页; 搜索结果发送模块, 用于将多个搜索结果和至少部分搜索结果对应的 原版网页提供至客户端, 并通过客户端展现。
图 8是根据本发明一个实施例的搜索引擎的结构示意图。
如图 8所示, 搜索引擎包括: 搜索词获取模块 210、搜索结果获取模块 220和搜索结果 发送模块 230。
具体地, 搜索词获取模块 210用于获取客户端发送的搜索词。
搜索结果获取模块 220用于根据搜索词获取多个搜索结果, 并分别获取多个搜索结果 中至少部分搜索结果对应的原版网页。 更具体地, 搜索词获取模块 210获取搜索词之后, 搜索结果获取模块 220根据搜索词获取对应的多个搜索结果, 再分别根据每个搜索结果获 取对应的原版网页, 如果搜索结果具有原版网页, 则获取对应的原版网页, 如果搜索结果 没有原版网页, 则有可能该搜索结果即为原版网页, 或者搜索引擎无法获取到对应的原版 网页。 理想情况下, 如果搜索结果具有原版网页, 则对应一个原版网页, 但是在实际的计算 过程中, 可能存在一个以上的原版网页, 这种情况出现的可能原因是, 在计算网页内容相 关性和原版网页匹配程度时出现多个符合算法定义的阀值范围的网页, 因此匹配了多个原 版网页。 现实情况中, 原版网页理论上只会有一个, 出现这个情况说明转载或盗版手段做 的比较精细, 一定程度上干扰了搜索引擎的算法系统, 但是这样的情况是允许发生的。
搜索结果发送模块 230用于将多个搜索结果和至少部分搜索结果对应的原版网页提供 至客户端, 以供客户端进行展现。
本发明实施例的搜索引擎, 在向客户端提供搜索结果时, 将搜索结果和其原版网页关 联, 使得客户端对搜索结果的原版网页也进行展现, 一方面如果搜索结果的排名靠前, 在 其具有原版网页时原版网页也可展现在排名靠前的位置,提供了一种更为便捷获得原版网 页的方式, 使得客户端可以轻松访问原版网页, 增加了原版网页被点击的概率, 提高 了客户端用户对于原版网页的重视程度, 在不修改整体搜索结果排序的情况下, 即方便 了用户又强调了搜索引擎对于原版网页的尊重和行动上的支持, 另一方面用户阅读原版网 页时往往可发现更多相关的内容和知识, 例如, 对于博客, 原作者的博客站点中会包含更 多领域相关的知识和文章, 提升用户体验。
图 9是根据本发明另一个实施例的搜索引擎的结构示意图。
如图 9所示, 搜索引擎包括: 搜索词获取模块 210、 搜索结果获取模块 220、 搜索结果 发送模块 230、 网页抓取模块 240、 网页分类模块 250和原版网页分析模块 260。
具体地, 网页抓取模块 240用于抓取多个网页, 例如, 可通过网络爬虫获取互联网上 的多个网页。
网页分类模块 250用于根据分类算法分别提取多个网页的特征值, 并根据多个网页的 特征值对多个网页进行分类以建立多个分类集合。 更具体地, 通过分类算法对所有网页分 类后, 产生分类集合的数量越大越好, 同时每个分类集合中的网页数量越少越好。 理想情 况是, 仅把网页内容非常相近的网页归为一个分类集合。 其中, 分类算法可运用目前较成 熟的机器学习算法,机器学习算法是一种通用算法,可有较多不同的变形算法,例如, KNN, 该算法的基本思路是: 在给定新文本后, 考虑在训练文本集中与该新文本距离最近 (最相 似) 的 K篇文本, 根据这 K篇文本所属的类别判定新文本所属的类别; 又如, 支持向量机 和神经网络算法在文本分类系统中应用得也较为广泛, 此类算法一个基本的特点就是把对 网页内容的计算结果向量化, 然后从数学上比较两个向量的差异。 应当理解, 对于网页内 容来说, 其往往包含丰富的各种信息, 可能包括图片、 视频等, 因此比较网页之间内容相 似度时需要综合考虑各种因素。
原版网页分析模块 260用于分别对每个分类集合中的网页进行分析, 以获取每个分类 集合中的原版网页。
在本发明的一个实施例中, 原版网页分析 260模块还用于: 分别获取每个分类集合中 的网页的搜索引擎抓取时间, 并根据搜索引擎抓取时间获取每个分类集合中的原版网页。 例如, 通常将抓取时间最早的网页作为原版网页。
在本发明的另一个实施例中, 原版网页分析模块 260还用于: 分别获取每个分类集合 中的网页的可靠性信息, 以根据搜索引擎抓取时间和可靠性信息获取每个分类集合中的原 版网页。 在本发明的一个实施例中, 可靠性信息包括网页的来源信息、 网页的质量信息和 反作弊判断信息中的一种或多种。 具体地, 获取原版网页时, 还需要考虑网页的来源信息、 网页的质量信息和反作弊判断信息等多种反映网页可靠性的信息, 例如, 通常正规网站的 网页可靠性更高, 质量高的网页可靠性更高, 判断作弊的网页的可靠性低等, 由此, 综合 考虑可靠性信息和抓取时间可增加原版网页的准确性。
图 10是根据本发明又一个实施例的搜索引擎的结构示意图。
如图 10所示, 搜索引擎包括: 搜索词获取模块 210、 搜索结果获取模块 220、 搜索结 果发送模块 230、 网页抓取模块 240、 网页分类模块 250、 原版网页分析模块 260和关联关 系建立模块 270。
具体地, 关联关系建立模块 270用于在原版网页分析模块 260获取多个分类集合中的 原版网页之后, 建立每个分类集合中网页与该分类集合中的原版网页之间的关联关系, 以 根据至少部分搜索结果和关联关系获取至少部分搜索结果对应的原版网页。 之后, 并该关 联关系进行永久保存, 例如, 可存储在数据库中或文件系统中, 在此不对存储装置做限定, 存储的方式也可多样化, 例如, 典型的存储格式为 [网页 A, 原版网页 1, 原版网页 2, …: |, [网页 B, 原版网页 3, 原版网页 4, …]等, 在此不对存储格式做限定。 存储的好处是方便 信息的快速、 直接获取, 搜索引擎获取多个搜索结果之后, 可以直接读取对应的原版网页。
为了实现上述实施例, 本发明的实施例还提出一种客户端。
一种客户端, 包括: 搜索词发送模块, 用于向搜索引擎发送搜索词; 搜索结果获取模 块, 用于获取搜索引擎根据搜索词发送的多个搜索结果和多个搜索结果中至少部分搜索结 果对应的原版网页; 以及搜索结果展现模块, 用于展现多个搜索结果和至少部分搜索结果 对应的原版网页。
图 11是根据本发明一个实施例的客户端的结构示意图。
如图 11所示, 客户端包括: 搜索词发送模块 110、 搜索结果获取模块 120和搜索结果 展现模块 130。
在本发明的一个实施例中, 客户端为 PC或移动终端, 其中移动终端可是智能手机、平 板电脑、 个人数字助理、 电子书等具有各种操作系统的硬件设备。 具体地, 搜索词发送模块 110用于向搜索引擎发送搜索词, 更具体地, 搜索词发送模 块 110可通过浏览器类应用程序访问搜索引擎网站并向搜索引擎发送搜索词, 也可以通过 其他快捷途径 (例如搜索栏等) 向搜索引擎发送搜索词, 在此不做限定。
搜索结果获取模块 120用于获取搜索引擎根据搜索词发送的多个搜索结果和多个搜索 结果中至少部分搜索结果对应的原版网页。
搜索结果展现模块 130用于展现多个搜索结果和至少部分搜索结果对应的原版网页。 本发明实施例的客户端, 可获取搜索引擎发送的搜索结果和其原版网页, 对搜索结果 的原版网页进行展现, 一方面如果搜索结果的排名靠前, 在其具有原版网页时原版网页也 可展现在排名靠前的位置, 提供了一种更为便捷获得原版网页的方式, 使得客户端可以轻 松访问原版网页, 增加了原版网页被点击的概率, 提高了客户端用户对于原版网页的重视 程度, 另一方面用户阅读原版网页时往往可发现更多相关的内容和知识, 例如, 对于博客, 原作者的博客站点中会包含更多领域相关的知识和文章, 提升用户体验。
为了更好地展现原版网页, 在客户端可具体地根据客户端配置设计不同的展现效果, 在本发明的一个实施例中, 搜索结果展现模块 130还用于: 接收多个搜索结果并展示给用 户, 并分别在至少部分搜索结果处添加提示标识, 以及当提示标识被触发时, 将被触发的 提示标识对应的搜索结果的原版网页展示给用户。
更具体地, 搜索结果展现模块 130可以解析出哪些搜索结果具有原版网页, 哪些不具 有原版网页, 在具有原版网页的搜索结果 (即至少部分搜索结果) 的适当位置添加提示标 志, 例如, 可以在其左侧、 右侧、 上方或下方。 举例来说, 如图 2 (b ) 所示, 对于具有原 版网页的每个搜索结果, 在其右侧通过一个图标 100提示用户具有原版网页。 此外, 提示 标志的触发方式多式多样, 例如, 使用鼠标点击或者在屏幕上进行手势点击, 或者鼠标划 过该提示标志或者手势划过该提示标志等。 举例来说, 如图 2 (b), 用户将鼠标移动到图 标上方即出现右侧的弹出式框 200, 将原版网页的信息展现给用户, 引导用户进行进一步 的点击, 用户可以点击进行浏览, 用户将鼠标移开即可关闭弹出式框 200。 另外, 用户还 可以参与到保护原版网页中, 可以在弹出式框 200的原版网页中显示 "举报非原版" 的操 作入口, 通过用户的参与可以给搜索引擎提供更多的辅助信息, 帮助将伪原版的网页找出 来, 搜索引擎还需根据策略来有效识别用户的举报是否真实可靠, 挖掘出有效的信息, 去 除恶意攻击的信息等。
应当理解, 展现给用户的原版网页可能包含多个, 用户一般情况下其实是可以分辨出 哪个才是真正的原版网页。
本发明实施例的客户端, 将原版网页通过提示标识进行提示, 方便用户阅读, 进一步 提升用户体验。 为了实现上述实施例, 本发明还提出一种应用程序。 该应用程序用于在运行时执行如 本发明实施例所述的搜索方法。
根据本发明实施例的应用程序, 在搜索引擎向客户端提供搜索结果时, 将搜索结果和 其原版网页关联, 对搜索结果的原版网页也进行展现, 一方面如果搜索结果的排名靠前, 在其具有原版网页时原版网页也可展现在排名靠前的位置,提供了一种更为便捷获得原版 网页的方式, 使得客户端可以轻松访问原版网页, 增加了原版网页被点击的概率, 提 高了客户端用户对于原版网页的重视程度, 在不修改整体搜索结果排序的情况下, 即方 便了用户又强调了搜索引擎对于原版网页的尊重和行动上的支持, 另一方面用户阅读原版 网页时往往可发现更多相关的内容和知识, 例如, 对于博客, 原作者的博客站点中会包含 更多领域相关的知识和文章, 提升用户体验。
为了实现上述实施例, 本发明还提出一种存储介质, 用于存储应用程序, 该应用程序 用于在运行时执行如本发明实施例所述的搜索方法。
根据本发明实施例的存储介质, 在搜索引擎向客户端提供搜索结果时, 将搜索结果和 其原版网页关联, 对搜索结果的原版网页也进行展现, 一方面如果搜索结果的排名靠前, 在其具有原版网页时原版网页也可展现在排名靠前的位置,提供了一种更为便捷获得原版 网页的方式, 使得客户端可以轻松访问原版网页, 增加了原版网页被点击的概率, 提 高了客户端用户对于原版网页的重视程度, 在不修改整体搜索结果排序的情况下, 即方 便了用户又强调了搜索引擎对于原版网页的尊重和行动上的支持, 另一方面用户阅读原版 网页时往往可发现更多相关的内容和知识, 例如, 对于博客, 原作者的博客站点中会包含 更多领域相关的知识和文章, 提升用户体验。
应理解, 本发明的各部分可用硬件、 软件、 固件或它们的组合来实现。 在上述实施方 式中, 多个步骤或方法可用存储在存储器中且由合适的指令执行系统执行的软件或固件来 实现。 如, 如果用硬件来实现, 和在另一实施方式中一样, 可用本领域公知的下列技术中 的任一项或他们的组合来实现: 具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻 辑电路, 具有合适的组合逻辑门电路的专用集成电路, 可编程门阵列 (PGA), 现场可编程 门阵列 (FPGA) 等。
在本说明书的描述中, 参考术语"一个实施例"、 "一些实施例"、 "示例"、 "具体示例"、 或"一些示例"等的描述意指结合该实施例或示例描述的具体特征、 结构、 材料或者特点包 含于本发明的至少一个实施例或示例中。 在本说明书中, 对上述术语的示意性表述不一定 指的是相同的实施例或示例。 而且, 描述的具体特征、 结构、 材料或者特点可以在任何的 一个或多个实施例或示例中以合适的方式结合。
尽管已经示出和描述了本发明的实施例, 本领域的普通技术人员可以理解: 在不脱离 本发明的原理和宗旨的情况下可以对这些实施例进行多种变化、 修改、 替换和变型, 本发 明的范围由权利要求及其等同物限定。

Claims

权利要求书
1、 一种搜索方法, 其特征在于, 包括以下步骤:
搜索引擎获取客户端发送的搜索词;
所述搜索引擎根据所述搜索词获取多个搜索结果, 并分别获取所述多个搜索结果 中至少部分搜索结果对应的原版网页;
所述搜索引擎将所述多个搜索结果和所述至少部分搜索结果对应的原版网页提供 至所述客户端, 并通过所述客户端展现。
2、 如权利要求 1所述的方法, 其特征在于, 还包括:
所述搜索引擎抓取多个网页;
所述搜索引擎根据分类算法分别提取所述多个网页的特征值;
所述搜索引擎根据所述多个网页的特征值对所述多个网页进行分类以建立多个分 类集合; 以及
所述搜索引擎分别对每个分类集合中的网页进行分析, 以获取每个分类集合中的 原版网页。
3、 如权利要求 2所述的方法, 其特征在于, 所述搜索引擎分别对每个分类集合中 的网页进行分析, 以获取每个分类集合中的原版网页, 具体包括:
所述搜索引擎获取每个分类集合中的网页的搜索引擎抓取时间; 以及
所述搜索引擎根据所述搜索引擎抓取时间获取每个分类集合中的原版网页。
4、 如权利要求 3所述的方法, 其特征在于, 所述搜索引擎分别对每个分类集合中 的网页进行分析, 以获取每个分类集合中的原版网页, 还包括:
所述搜索引擎获取每个分类集合中的网页的可靠性信息, 以根据所述搜索引擎抓 取时间和可靠性信息获取每个分类集合中的原版网页。
5、如权利要求 4所述的方法, 其特征在于, 所述可靠性信息包括网页的来源信息、 网页的质量信息和反作弊判断信息中的一种或多种。
6、 一种搜索系统, 其特征在于, 包括: 客户端和搜索引擎, 其中,
所述客户端, 用于向所述搜索引擎发送搜索词, 并对所述搜索引擎返回的内容进 行展现;
所述搜索引擎, 用于获取所述搜索词, 并根据所述搜索词获取多个搜索结果, 并 分别获取所述多个搜索结果中至少部分搜索结果对应的原版网页, 以及将所述多个搜 索结果和所述至少部分搜索结果对应的原版网页提供至所述客户端。
7、 如权利要求 6所述的系统, 其特征在于, 所述搜索引擎还用于: 抓取多个网页, 并根据分类算法分别提取所述多个网页的特征值, 以及根据所述 多个网页的特征值对所述多个网页进行分类以建立多个分类集合, 并分别对每个分类 集合中的网页进行分析, 以获取每个分类集合中的原版网页。
8、 如权利要求 7所述的系统, 其特征在于, 所述搜索引擎还用于:
获取每个分类集合中的网页的搜索引擎抓取时间, 并根据所述搜索引擎抓取时间 获取每个分类集合中的原版网页。
9、 如权利要求 8所述的系统, 其特征在于, 所述搜索引擎还用于:
获取每个分类集合中的网页的可靠性信息, 并根据所述搜索引擎抓取时间和可靠 性信息获取每个分类集合中的原版网页。
10、 如权利要求 9所述的系统, 其特征在于, 所述可靠性信息包括网页的来源信 息、 网页的质量信息和反作弊判断信息中的一种或多种。
1 1、 如权利要求 6- 10任一项所述的系统, 其特征在于, 所述客户端还用于: 接收所述多个搜索结果并展示给用户, 并分别在所述至少部分搜索结果处添加提 示标识, 以及当所述提示标识被触发时, 将被触发的提示标识对应的搜索结果的原版 网页展示给所述用户。
12、 一种搜索引擎, 其特征在于, 包括:
搜索词获取模块, 用于获取客户端发送的搜索词;
搜索结果获取模块, 用于根据所述搜索词获取多个搜索结果, 并分别获取所述多 个搜索结果中至少部分搜索结果对应的原版网页;
搜索结果发送模块, 用于将所述多个搜索结果和所述至少部分搜索结果对应的原 版网页提供至所述客户端, 以供所述客户端进行展现。
13、 如权利要求 12所述的搜索引擎, 其特征在于, 还包括:
网页抓取模块, 用于抓取多个网页;
网页分类模块, 用于根据分类算法分别提取所述多个网页的特征值, 并根据所述 多个网页的特征值对所述多个网页进行分类以建立多个分类集合; 以及
原版网页分析模块, 用于对每个分类集合中的网页进行分析, 以获取每个分类集 合中的原版网页。
14、如权利要求 13所述的搜索引擎, 其特征在于, 所述原版网页分析模块还用于: 获取每个分类集合中的网页的搜索引擎抓取时间, 并根据所述搜索引擎抓取时间获取 每个分类集合中的原版网页。
15、如权利要求 14所述的搜索引擎, 其特征在于, 所述原版网页分析模块还用于: 获取每个分类集合中的网页的可靠性信息, 以根据所述搜索引擎抓取时间和可靠性信 息获取每个分类集合中的原版网页。
16、 如权利要求 15所述的搜索引擎, 其特征在于, 所述可靠性信息包括网页的来 源信息、 网页的质量信息和反作弊判断信息中的一种或多种。
17、 一种客户端, 其特征在于, 还包括:
搜索词发送模块, 用于向搜索引擎发送搜索词;
搜索结果获取模块, 用于获取所述搜索引擎根据所述搜索词发送的多个搜索结果 和所述多个搜索结果中至少部分搜索结果对应的原版网页; 以及
搜索结果展现模块, 用于展现所述多个搜索结果和所述至少部分搜索结果对应的 原版网页。
18、 如权利要求 17所述的客户端, 其特征在于, 所述搜索结果展现模块还用于: 接收所述多个搜索结果并展示给用户, 并分别在所述至少部分搜索结果处添加提示标 识, 以及当所述提示标识被触发时, 将被触发的提示标识对应的搜索结果的原版网页 展示给所述用户。
19、 一种应用程序, 其特征在于, 用于在运行时执行如权利要求 1-5任一项所述的搜 索方法。
20、 一种存储介质, 其特征在于, 用于存储应用程序, 所述应用程序用于在运行时执 行如权利要求 1-5任一项所述的搜索方法。
PCT/CN2014/084940 2013-08-21 2014-08-21 搜索方法、系统、搜索引擎和客户端 WO2015024522A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP14837976.1A EP3037985A4 (en) 2013-08-21 2014-08-21 Search method and system, search engine and client
US14/913,368 US20160203222A1 (en) 2013-08-21 2014-08-21 Search method, search system, and search engine

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310367193.X 2013-08-21
CN201310367193XA CN103399957A (zh) 2013-08-21 2013-08-21 搜索方法、系统、搜索引擎和客户端

Publications (1)

Publication Number Publication Date
WO2015024522A1 true WO2015024522A1 (zh) 2015-02-26

Family

ID=49563585

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/084940 WO2015024522A1 (zh) 2013-08-21 2014-08-21 搜索方法、系统、搜索引擎和客户端

Country Status (4)

Country Link
US (1) US20160203222A1 (zh)
EP (1) EP3037985A4 (zh)
CN (1) CN103399957A (zh)
WO (1) WO2015024522A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103399957A (zh) * 2013-08-21 2013-11-20 百度在线网络技术(北京)有限公司 搜索方法、系统、搜索引擎和客户端
CN106649302A (zh) * 2015-10-28 2017-05-10 腾讯科技(深圳)有限公司 一种搜索排序方法以及装置
CN105718533A (zh) * 2016-01-15 2016-06-29 百度在线网络技术(北京)有限公司 信息推送方法和装置
CN111552778B (zh) * 2020-04-26 2024-05-14 北京达佳互联信息技术有限公司 音频资源管理方法、装置、计算机可读存储介质及设备
CN113034995B (zh) * 2021-04-26 2023-04-11 读书郎教育科技有限公司 一种学生平板生成听写内容的方法及系统
CN117715049B (zh) * 2024-02-05 2024-04-12 成都一心航科技有限公司 一种针对手机浏览器的反作弊系统以及反作弊方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101025737A (zh) * 2006-02-22 2007-08-29 王东 基于关注度的同源信息搜索引擎聚合显示方法及相关系统
CN101114284A (zh) * 2006-07-27 2008-01-30 阿里巴巴公司 一种显示网页内容相关信息的方法及系统
US20080092051A1 (en) * 2006-10-11 2008-04-17 Laurent Frederick Sidon Method of dynamically creating real time presentations responsive to search expression
CN101499098A (zh) * 2009-03-04 2009-08-05 阿里巴巴集团控股有限公司 一种网页评估值的确定及运用的方法、系统
CN103399957A (zh) * 2013-08-21 2013-11-20 百度在线网络技术(北京)有限公司 搜索方法、系统、搜索引擎和客户端

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101086530B1 (ko) * 2008-10-02 2011-11-23 엔에이치엔(주) 웹 문서 원본 판별 방법 및 시스템, 이를 위한 웹 문서 이력 정보 제공 방법 및 시스템
CN103235821B (zh) * 2013-04-27 2015-06-24 百度在线网络技术(北京)有限公司 原创内容的搜索方法和搜索服务器

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101025737A (zh) * 2006-02-22 2007-08-29 王东 基于关注度的同源信息搜索引擎聚合显示方法及相关系统
CN101114284A (zh) * 2006-07-27 2008-01-30 阿里巴巴公司 一种显示网页内容相关信息的方法及系统
US20080092051A1 (en) * 2006-10-11 2008-04-17 Laurent Frederick Sidon Method of dynamically creating real time presentations responsive to search expression
CN101499098A (zh) * 2009-03-04 2009-08-05 阿里巴巴集团控股有限公司 一种网页评估值的确定及运用的方法、系统
CN103399957A (zh) * 2013-08-21 2013-11-20 百度在线网络技术(北京)有限公司 搜索方法、系统、搜索引擎和客户端

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3037985A4 *

Also Published As

Publication number Publication date
EP3037985A1 (en) 2016-06-29
CN103399957A (zh) 2013-11-20
US20160203222A1 (en) 2016-07-14
EP3037985A4 (en) 2017-05-03

Similar Documents

Publication Publication Date Title
JP6487201B2 (ja) 推奨ページを生成するための方法及び装置
WO2019218514A1 (zh) 网页目标信息的提取方法、装置及存储介质
US7917514B2 (en) Visual and multi-dimensional search
CN110537180B (zh) 用于直接浏览器内标记因特网内容中的元素的系统和方法
US10380197B2 (en) Network searching method and network searching system
US7739221B2 (en) Visual and multi-dimensional search
CN107657048B (zh) 用户识别方法及装置
US20150169710A1 (en) Method and apparatus for providing search results
WO2019041521A1 (zh) 用户关键词提取装置、方法及计算机可读存储介质
WO2015024522A1 (zh) 搜索方法、系统、搜索引擎和客户端
US10311120B2 (en) Method and apparatus for identifying webpage type
US10229190B2 (en) Latent semantic indexing in application classification
US9639767B2 (en) Context-aware handwriting recognition for application input fields
JP6500086B2 (ja) 二次元コードの解析方法および装置、コンピュータ読み取り可能な記憶媒体、コンピュータプログラムおよび端末機器
WO2021068681A1 (zh) 标签分析方法、装置及计算机可读存储介质
EP3175375A1 (en) Image based search to identify objects in documents
JP2015225669A (ja) 注釈表示支援装置及び注釈表示支援方法
Zhou et al. Visual similarity based anti-phishing with the combination of local and global features
US20180039646A1 (en) Method and apparatus for storing access log based on keyword
CN104750692B (zh) 一种信息处理方法、信息检索方法及其对应的装置
JP2020091607A (ja) 検索システム、及び検索方法
US10534810B1 (en) Computerized systems and methods for enriching a knowledge base for search queries
Xu et al. Estimating similarity of rich internet pages using visual information
US20210117438A1 (en) System and method for generating queryeable structured document from an unstructured document using machine learning
JP2018072873A (ja) 情報処理装置、情報処理方法、およびプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14837976

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 14913368

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2014837976

Country of ref document: EP