Three requirements for information update by vertical search

Intermediary transaction SEO diagnosis Taobao guest Cloud host technology Hall

Vertical search has special requirements for updating information, which can be considered from the following considerations:

1. The stability of the information source (cannot let the information source website feel the spider pressure)

2. The cost of grasping

3. Improve the user experience.

According to the above points to develop a better strategy, to do just right.

The strategy can evaluate the website/web page update coefficient, the site/page key factor, the user clicks the factor (or the exposure coefficient), the website stability coefficient ..., based on these coefficients to determine the frequency of these sites/Web pages update. Again because of the new information and updated information on the list page front or first, so a good rating of the Web page can be very low-cost to solve the problem of updates, the coefficient of the lower page January update once, a slightly higher weekly update once, medium days to once a day, High in hours to minutes. Similar search engine's big storehouse, Zhou Cu, Day storehouse, hour storehouse ...

Based on the visual Block analysis technology, the simulation of IE browser display mode, the Web page analysis.

According to the principle of human vision, the results of the analysis of the Web pages are divided into blocks, and then according to the needs of these blocks for processing, such as: acquisition orientation, introduction extraction and some of the necessary content extraction text extraction ...

Structured information extraction technology, the unstructured data in Web pages are extracted into structured data according to certain requirements.

There are two ways, the simple template approach, in addition to the Web page does not rely on web-based structured information extraction methods, these two ways to each other to take advantage of the simplest and most effective way to meet the needs. The biggest difference between vertical search engine and general search engine is the deep processing of the structured data after the structure extraction of the webpage information, and provides the professional search service. Therefore, the technology of web structured information extraction is an important technical index to determine the quality of vertical search engine. In fact, the web structure of information extraction in Baidu, Google has been widely used, such as: MP3, image search, Google's local search is extracted from the Web Library enterprise information, add to its map search, Google through this technology is subversive to do the content of the way. The same technology is also used in Qihoo, Sogou shopping, shopping and other applications.

Simple grammar analysis, simple grammar analysis is very important in the search engine, can improve the quality of data by simple grammar analysis, low-cost access to some kind of information, improve sorting, look for the content of need ...

Information processing technology, information processing includes a wide range of

Mainly includes the heavy, the clustering, the analysis ..., this according to need the related technology to be very many.

Data mining, finding the relevance of your information is important and effective for vertical search, and can provide users with more detailed services in these dependencies.

Word segmentation technology, search-oriented word segmentation technology, build and your industry-related thesaurus.

Note that this is a search-oriented participle, not a recognition-oriented and accurate participle. It's not too much to have more than 10 people on the job.

Indexing technology, indexing technology for vertical search is very critical, a Web site-level search engine must support the distribution index, hierarchical database, distributed search, flexible update, flexible weight adjustment, flexible indexing and flexible upgrade extension, high reliability and stability redundancy. There is also a need to support extensions of various technologies, such as offset calculations.

Other techniques, slightly.

The technical evaluation of vertical search engine should be judged from the following points

1. Comprehensive

2. Updated

3. Accuracy

4. Functional