Unveiling Chinese search engine technology: Sorting Technology (5)

Source: e800.com.cn

Development Trend of Sorting Technology The technical improvements and optimizations of various search engines are directly reflected in the sorting of search results. Many search engines are further studying new sorting methods to improve customer satisfaction. Professionals believe that the current search engine sortingAlgorithmThere are still two problems. 1. relevance is not solved. Relevance refers to the degree of relevance between the search term and the page. The search term andArticleNot to mention the fact that many of these features do not exist at the same time. This is also the reason why many methods can do harm to search engines. In addition, some articles do not contain search words, but are about content that is very relevant to search words, such as searching for "terrorists". However, some webpages refer to some destructive actions of bin Laden, no sub-eyes of terrorists appear in the text, and the search engine cannot find the webpage. Surface features can only be controlled. The root cause should be to increase semantic understanding, such as the extraction of keywords and keywords, and semantic analysis to determine the degree of relevance between search words and webpages. the more accurate the analysis, the better the effect. 2. Simplification of search results. In the search engine, anyone who searches for the same word returns the same result. In this way, visitors are obviously not satisfied. Scientists may search for "planet" to learn about the planet, but ordinary people may want to find a "star wars" movie, but the search engine returns the same result. How to satisfy these different types of visitors requires personalization of search results. Foreign Vivisimo Company (http://www.vivisimo.com/) is to solve this problem, they use the method of automatic clustering of search results to meet the needs of different types of customers. Vivisimo has taken a step towards sorting search results from simplification to personalization, but the most ideal result should be for each visitor. The sorting results are directly related to their search habits and willingness. When searching for "Sports", people who like football should put the results of football at the top, and those who like basketball should put the results at the top. The Sorting Technology of search engines should also develop towards solving these two problems: Semantic Relevance and sorting personalization. The former requires a comprehensive natural language processing technology, and the latter needs to record huge Visitor Information and complex computing. It is not easy to meet any of the requirements. How can we solve these problems, the task falls on the shoulders of scientists and engineers. Which search engine solves these problems may be called the overlord of the next search world. More references Note: Since the following references are not published in some journals in the form of papers, there is no apparent source, you can get the download link for the relevant article by searching the article title on Google or Baidu search engine. [1] Chinese search engine technology unveiling: Chinese word segmentation. [2] Chinese search engine technology: web spider. [3] unveiling the Chinese search engine technology: system architecture. [4] robots & spiders & crawlers: How web and intranet search engines follow links to build indexes. Author: Avi rapports.2001. [5] guidelines for Robot writers. Author: Martijn Koster, 1993.