bizearch网站信息源质量到底如何，外贸客户开发渠道，以测试产品为例，每页30家公司，1980页为59400家公司信息，此站到底可不可用，一探究竟。外贸实战笔记，转载注明出处。

http://bizearch.com如何使用

进入网站后，查看company database，非常多产品种类，点击自己产品，进入后，又可以细分。

以agriculture为例，有5万多家公司。

站点结构分析

每一家公司页面都有这些信息，来分析后抓取。product services，需要抓取，后续通过Excel筛选出与自己产品更符合的。

公司名，一般我自己不抓，这里也将Web Scraper设置出来了。网站website是最需要的。地址，需要，后续排除掉国内。再比如business type，甚至可以将manufacturer排掉。

Web Scraper

这串代码graph模仿官网思路，非常优秀，减少页面结构复杂程度，避免不必要运行。

				
					{"_id":"from1980","startUrl":["http://bizearch.com/company/Chemicals.5-[2-1980].htm"],"selectors":[{"id":"1allcompany","parentSelectors":["_root"],"type":"SelectorLink","selector":".ls a","multiple":true},{"id":"companyname","parentSelectors":["_root","1allcompany"],"type":"SelectorText","selector":"strong","multiple":false,"regex":""},{"id":"contactperson","parentSelectors":["_root","1allcompany"],"type":"SelectorText","selector":"tr:contains('Contact Person') td","multiple":false,"regex":""},{"id":"address","parentSelectors":["_root","1allcompany"],"type":"SelectorText","selector":"tr:contains('Company Address') td","multiple":false,"regex":""},{"id":"web","parentSelectors":["_root","1allcompany"],"type":"SelectorText","selector":"tr:contains('Website') td","multiple":false,"regex":""},{"id":"businesstype","parentSelectors":["_root","1allcompany"],"type":"SelectorText","selector":"tr:contains('Business Type') td","multiple":true,"regex":""},{"id":"industryfocus","parentSelectors":["_root","1allcompany"],"type":"SelectorText","selector":"tr:contains('Industry Focus') td","multiple":false,"regex":""}]}

上述Web Scraper代码可爬取公司名，联系人名称，网站，公司类型，主营业务。你也可以将此站代码复制进自己的web scraper进行修改。

仔细研究下页面selector的结构顺序。