Optimizing Mobile Content

Watson™ Explorer Engine provides a proxy that enables you to specify an XPath expression which identifies the section of web pages from selected sites that contains their primary content. This enables you to automatically eliminate content such as navigation elements when viewing search results from those sites.

You can customize the list of web sites that are handled by the proxy, adding other sites that are used by your search application. The list of sites for which the proxy performs special processing is stored in the Watson Explorer Engine XML function mobile-proxy-parser. To edit this function, select XML from the left-hand menu in the Watson Explorer Engine administration tool, and click All from the sub-menu that displays. Next, enter mobile-proxy-parser in the Lookup textarea, and click Select to display this function.

Once the mobile-proxy-parser displays in the Watson Explorer Engine administration tool, click edit to enable you to edit the function, and scroll down until you see the section that sets the body-xpath variable. This contains a number of site-specific valuesm, each of which looks like the following:

<xsl:when test="viv:test($page-provideurl, 'imdb\.com', 'regex')">
  <xsl:text>//div[@id='tn15content']</xsl:text>
  </xsl:when>

To quickly add a section for an additional web site, copy the conditional for an existing web site, and change the basename of the domain in the first line to match the new domain that you want to Proxy. Next, modify the XPath expression inside the <xsl:text> element in the second line, replacing it with an XPath expression that identifies the primary content area for pages on the new site.

To save your changes, click OK. Search results found on the site for which you added information will now only display within the context of the specified content area, without having to retrieve and display higher-level content such as ads, navigation, and so on.