Search engine principle and optimization thought analysis

Intermediary transaction http://www.aliyun.com/zixun/aggregation/6858.html ">seo diagnose Taobao guest cloud host technology Hall

The basic composition and function of search engine

• A search engine program mainly by the searcher, indexers, and other four parts of the user interface, the main storage device by the page memory and storage bucket two components.

• Searcher: Crawler Crawl Compact Repository

• Indexer: The repository extracts web page information, analyzes and decomposes, establishes the keyword index, the preliminary sort processing, the storage bucket, namely the hardware storage unit.

• Users submit queries through the user interface, the search is based on input keywords, in the indexer and bucket to find, and using algorithms to the final ranking of results

Page priority algorithm based on web information

• Web content based algorithm: keywords in a special position in the situation: for example, Titile,meta,des.

• Keywords in the text of the page appears: The total number of keywords appear, the average interval of words, the frequency of the keywords appear.

• Web page link based algorithm: for example, PageRank algorithm hits algorithm for hits insufficient to add some of the column algorithm.

The algorithm of user behavior related page

• The user's opinion about the relevance of the search results cannot be ignored. Adjust page priority by analyzing web logs

• The dir ect Hit algorithm based on CTR: The popularity of the page is judged by the rate of clicks returned by the search results and the length of time associated with the page.

• Other user behavior: for example, by filtering the user behavior two times, the gap between search results and user expectations is gradually narrowed. Cookie records, hot keywords, etc...

Make the site included

How to make the site included

If not included, whether to the search engine blocked?

• All site data is trending down, even zero, and multiple search engines are showing this

• Through the site log analysis spiders visit the site: no links, invalid links, no work and return.

We want to attract links.

How to make more pages included

• Eliminate spider traps: robots.txt settings do not use spiders not access to the technology to display content, such as pop-up windows, frames, Flash,img,js use JS to write the Drop-down menu. Such an unrecognized content, two can not follow the link crawling. Dynamic URL address too long, too many dynamic parameters,? & = et cetera, avoid entering the black hole. Make 404 pages to ensure server response. Open the Web site in at least 10 seconds.

• Reduce neglected content: The Web page, the spider crawled over a certain size of the page will stop crawling, add too much content of the Web page, you can use the unnecessary content of JS to write. Flash inside to ensure that you do not want to be included in the content, avoid the use of frames.

• Build Spider program channel: Design site map.

Optimize content

Search ranking elements: two main categories

• Page elements: Link popularity, user behavior, url length and depth, freshness: content, site structure, do not cheat

• Search request elements: keyword prominence, density, frequency, content, TF, search term proximity

Attract links to your site

• The most important ranking factor at present is determined by the link.

• Content for the king on the previous internet is standing statistically, but it is not the content that causes the Internet to change, and is the link. The internet is easy to move from one part of the content to another. The 1998 Google emerged, breaking the traditional ranking algorithm based on keyword search, but based on link analysis, using links to judge the quality of the page level. PR

• Popularity of Links: number of links, link quality, anchor text,

• Link dependencies: Simple anchor text to determine the correlation is not enough, search engine will look at the anchor text around the word, view the entire page or even the entire link source site words.

Weight value of links

• Internal links < Within the same family < bidirectional links < congested one-way links < sparse one-way links

• What is the same family link: IP WHOIS repeat similar anchor text to these weights is not high.

This article from www.chenhuayi.com Original, reprint please indicate the source.