Intermediary transaction http://www.aliyun.com/zixun/aggregation/6858.html ">seo diagnose Taobao guest cloud host technology Hall
The search engine is also a model of WEB2.0, because the user input data needs to be processed so that it can interact well with the user. But for search engines, and not the user input of all the words are directly used in ranking, but first will be processed by their own, the processing of the words after the direct user ranking. This article according to own summary and everybody share the search engine is how to handle our input search term. If the wrong place, also welcome everyone to correct me.
First, the first participle: participle is a major feature, whether we enter a long sentence or a phrase, search engines will be the word segmentation, the search string cut into a word based on the combination of keywords. As long as this, in order to better through their own preprocessed database for query and comparison, all of these words are not listed to return to the user. The principle of this participle and page participle is somewhat similar.
Second, to stop the word: We have a lot of subjectivity in the search, many times the colloquial words are used in the search term, but this word is often to search for content without substantial help, but we have developed the habit of inputting these words, so search engines must be able to stop the word, To the maximum extent to ensure the accuracy and relevance of the search.
Third, the instruction processing: usually we will enter many words in the search, in fact, the default case, the search engine put multiple default words in accordance with the word "and". For example, when you search the "Weight loss" "Site", the search engine will default we entered the "Weight loss site" to deal with, the individual contains "weight loss" or "website" content in theory will not return, but the actual situation will often appear to contain a single word site. In addition, there are a variety of advanced search instructions to deal with the words, such as the plus and minus signs.
Four, error correction: Many times we input will have the wrong problem, pinyin is more homophonic, the wrong is normal, but if the search engine does not do error correction, it will return a lot we do not want to see the results. So, sometimes we even enter the whole spell, search engine will be based on this to prompt the correct Chinese.
The five, the classification treats: the different words are generally different treats, this is a search engine's characteristic function. For example, we search for general keywords, return to the general Web page majority, and when we search for a star's name, there will be a lot of news, video, pictures and other comprehensive search results. Therefore, the search engine should be based on the needs of users to judge. This judgment is in the preprocessing phase.
Well, search engine for the processing of search terms is basically from these five aspects, here to make a summary: participle, to stop the word, the processing of instructions, error correction, classification treatment. Although this does not have the direct relation to our website, but understands some basic knowledge to believe will have the indirect function. website, user, search engine originally three is one, these three compose the whole internet. A lot of truth is also figured out, know these simple knowledge, there may be some other aspects of our thinking and inspiration.
This article is here, will continue to share other aspects of knowledge, welcome and 11544.html "> I contact, this article from: Niu Niu Games, Web site: http://www.niuniuxiaoyouxi.com/, also welcome reprint, reproduced please retain the copyright."