Building a web‐snippet clustering system based on a mixed clustering method

Based on The Document-Link and Time-Clue Relationships Between Blog Posts to Improve the Performance of Google Blog Search

Both the blog search engine and the general search engine automatically crawl the pages from the web and produce relevant search results based on the user's query. The first difference between the two types is that the blog search engine focuses on dealing with blog posts and filters out other types of pages. This difference allows bloggers only to care about posts rather than all pages that are indexed by general search engines. The second difference is the post, considering more time-related issues compared to the page. The semantic analysis model is widely used to analyze the various semantic relationships that may arise in the document. In this article, the authors propose a new semantic analysis model to find possible time relationships between posts. The main contribution of this paper has two points: first is that this paper builds a high-performance search system that considers the discussion topic and updated time between posts; second, is that the authors consider the time relationships between posts that can rank the relevant blog topics based on the popularity of the posts.