6. Positive index
Index of abbreviations. After the above steps (extracting, phraseing, noise loss, weighting), the search engine eventually gets the only content that reflects the main contents of the page, in word form。
Next, the indexing program for the search engine extracts keywords and converts the page to a collection of keywords according to words divided by the semiwords program. Frequency, frequency, format (e. G. Title label, bold, h tag, anchor text, etc.) are also required. ) and the location of each keyword on the page (e. G. First paragraph of the page). . The indexing program of the search engine will store pages and the vocabulary structure of keywords into the index database。
7. Backward index
The forward index cannot be used directly for ranking. Assumes that the user search key. If there is only a positive index, the ranking program needs to scan all the files in the index database, find the files containing keywords and calculate the relevance。
This calculation does not meet the requirement to return the ranking results in real time. The search engine categorizes all keywords in advance, reconstructs the positive index database into a reverse index and converts the image of the file to the key word to the document. In the inverted index, the keyword is the primary key and each keyword corresponds to a series of files. For example, all documents shown on the right side of the first line below are those with keyword 1. In this way, when the user searches for the keyword, the sorting program locates the keyword in the inverted index and can immediately find the files of all the keywords。
Iv. Ranking of search results
The search engine is ready to process the user search at any time after the front spider grabs the page and the data preprocessing and indexing program is indexed backwards. After the search box enters what you want to search for, the ranking program calls the data from the index library and calculates the ranking and displays the content on the search result page。
1. Search word processing
Once the search engine has received the search word entered by the user, some processing of the search word is required before it enters the ranking process. The search process includes: chinese dictions, stop words, command processing。
When the above steps are completed, the default treatment of the remaining elements in the search engine is the use of "and" logic between keywords。
For example, in the search box, the user enters the term "decoration method" and after the word is split and stopped, the remaining keywords are "decoration" and "methodology" and the search engine is sorted with the default that the user wants to ask for both "decoration" and "methodology"。
2. Document matching
After processing the search words above, the search engine obtained a collection of keywords in the form of words. The next stage to go: the document matching phase, is the identification of documents containing all keywords. The inverted index mentioned in the index section allows for quick completion of the file matching, assuming that the user searches for "keyword 1" and "keyword 2" in the index, the ranking program can find all the page files that contain each of these words。
3. Selection of initial subsets
When matching documents containing all the keywords are found, the relevance of these documents cannot be calculated, because, in practice, dozens, millions, if not millions, of documents are often found. It will take a long time to calculate the relevance of so many documents in real time. The 100-degree search engine, with a maximum return of 760 results, will only need to calculate the relevance of the first 760 results to meet the requirements。
Since all matching files already have the most basic relevance (these documents contain all query keywords), the search engine selects 1,000 files with higher page weights, initializes a subset of the selection weights, and then calculates the relevance of this sub-totaled page。
4. Relevance calculations
When the initial subset is selected by weight, it is the step of calculating the correlation of keywords to the sub-concentration page. The calculation of relevance is the most important step in the ranking process and the main factors affecting relevance include the following:
1 usage of keywords
After multiple keywords, the value of the entire search string is not the same. The more common words contribute less to the search, the less common words contribute more to the search. So the search engine does not treat the keywords in the search string equally, but rather weights them according to their common usage. High weighting coefficients for unused words, low weighting coefficients for commonly used words, and more attention to unused words in ranking algorithms。
2 phrase frequency and density
It is generally accepted that in the absence of keywords, the more frequent and dense the search words appear on the page, the more relevant the page is. This is, of course, a general pattern, which is not necessarily the case in practice, so there are other factors in the calculation of relevance. Frequency and density are only part of the factor and are becoming less important。
3. Location and form of keyword
As mentioned in the index section, the presentation and location of page keywords are recorded in the index library. Keywords appear in more important places, such as title labels, boldfaces, h1, etc. The more the page is relevant to keywords, this part of the page is addressed by seo。
4 keyword distance
The presence of a complete matching of key words after the cut indicates that the search words are most relevant. For example, when searching for diet methods, the word "fatage method" is the most relevant word on the page. If the words "weight loss" and "methods" do not match consecutively, they appear closer and are considered by the search engine to be slightly more relevant。
5 link analysis and page weights
In addition to the page itself, links and weight relationships between pages affect the relevance of keywords, the most important of which is anchor text. The more the page has an import link to the search word anchor text, the more relevant it is. The link analysis also includes the content theme of the link page itself, text around the anchor text, etc。
Summary: it is important to understand this knowledge for us to do 100-degree web entries, for example, that the title contains the desired word that the user may search for, and that the text properly reflects the key word or split word that helps to judge the relevance of the content to the user search word。
V. Seo search engine marketing promotion
1. Targeting website outreach
A website has different objectives in the development process and may be client-seeking, increased traffic, etc., so identifying suitable outreach targets can help to select a good keyword。
2. Collecting information on markets
Market information is constantly evolving and it is essential to keep abreast of the market at all times, and to capture the dynamics of the information for the purpose of selecting keywords。
The first is to increase the number of websites by competitive means, most users do not read the content of the three pages behind the search engine, and only information that ranks ahead will receive attention from users. Obtaining names through competitive bidding is a common method for many small and medium-sized websites, which can quickly raise the profile of the sites and bring about human activity and traffic. The disadvantage is to spend money, and it is feasible to choose such an approach if needed。
Second, it optimizes internality and identifies the rule of law suitable for search engines. The search engine has a basic set of rules, and if your website follows the search engine's code, it can be significantly improved, and instead it is not ideal。
3. The selection of relatively popular search engines, such as 100 degrees, dogs, 360 searches, etc。
The most appropriate keywords are to be selected, as searchers can be easily located only if the relevant keywords are selected。
In order to ensure that the ranking is high, information searchers, when searching for keywords on search engines, find numerous registered enterprise websites, however, often focus on the top 10 or 20。
How does it fit the law of search engines
1. Reducing the number of pictures and flash files in web design, the excessive number of pictures and flash in web pages will affect the speed within the site, and the search engine will not be fully identifiable when it identifies some of the pictures and flash, and the search engine will be considered to be useless, so that the pr portion of the site will be reduced。
2. One-page keywords can be used to increase the number of names, which account for a significant proportion of the search engine, and to optimize the website。
3-friendly links are selected and used. The use of friendly links can bring a great deal of traffic to the website, which is what the site director needs to do。
A summary of the search engine extension methods:
