Webpage labeling method and device
A technology for marking and labeling webpages, applied in the Internet field, can solve problems such as low recall rate, small coverage of artificially designed label system, and failure to meet the real needs of users, and achieve high recall rate and wide label coverage
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
example 1
[0054] Example 1: The web page provides the download service of the TV series "The Legend of Zhen Huan". When the user clicks on the web page, a corresponding query statement will be generated in the query log. If the NER tool is used to analyze the query statement, the obtained named entity word is "Zhen Huan Biography" and the demand word is "Download", then the combination mode of the named entity word and the demand word is "Zhen Huan Biography+Download".
example 2
[0055] Example 2: The web page provides ticket price information for Shanghai Disneyland. When the user clicks on the web page, a corresponding query statement will be generated in the query log. If the NER tool is used to analyze the query sentence, the named entity words obtained are "Shanghai" and "Disneyland", and the demand word obtained is "ticket price", then the combination mode of the named entity word and the demand word is "Shanghai + Disneyland+ ticket prices".
[0056] S22: Obtain the page views corresponding to the query statement.
[0057] In this embodiment, after obtaining the query statement conforming to the combination pattern of the named entity word and the demand word, the number of page views corresponding to the part of the query statement is further obtained.
[0058] It should be understood that the number of page views is the total number of visits to the webpage by the user.
[0059] S23: Sort the query statements according to the number of page...
example 3
[0122] Example 3: Use the combination of the required label and the feature to classify the corresponding label for the webpage to be labeled.
[0123] The combination of requirement labels and features is used to mark the corresponding labels for the webpages to be marked by calculating the similarity between the requirement labels and the characteristics. When using this labeling method to label tags, a similarity threshold needs to be set in advance to judge whether the calculated similarity reaches the similarity threshold. If the similarity calculated according to the characteristics of the requirement tag and the webpage to be marked reaches the similarity threshold, the requirement tag is marked on the webpage to be marked.
[0124] Optionally, when various classifiers are used to label corresponding labels for webpages to be labeled, appropriate labels can also be selected manually in combination with prior rules, so as to make the labeled labels more accurate. For ex...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


