<br><div class="gmail_quote">2011/3/3 Lime, Steve D (DNR) <span dir="ltr"><<a href="mailto:Steve.Lime@state.mn.us">Steve.Lime@state.mn.us</a>></span><br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div link="blue" vlink="purple" lang="EN-US"><div><p class="MsoNormal"><span style="font-size: 11pt; color: rgb(31, 73, 125);">Pretty friggin’ cool. Takes a minute to get ones head around what’s going on. Is this interpretation correct?</span></p>
<p class="MsoNormal"><span style="font-size: 11pt; color: rgb(31, 73, 125);"> </span></p><p><span style="font-size: 11pt; color: rgb(31, 73, 125);"><span>-<span style="font: 7pt "Times New Roman";"> </span></span></span><span style="font-size: 11pt; color: rgb(31, 73, 125);">Grey box indicates a cluster representing multiple types</span></p>
<p><span style="font-size: 11pt; color: rgb(31, 73, 125);"><span>-<span style="font: 7pt "Times New Roman";"> </span></span></span><span style="font-size: 11pt; color: rgb(31, 73, 125);">Icon with a number indicates a cluster representing a single type</span></p>
<p><span style="font-size: 11pt; color: rgb(31, 73, 125);"><span>-<span style="font: 7pt "Times New Roman";"> </span></span></span><span style="font-size: 11pt; color: rgb(31, 73, 125);">Blue dots are the points being clustered</span></p>
<p class="MsoNormal"><span style="font-size: 11pt; color: rgb(31, 73, 125);"> </span></p></div></div></blockquote><div><br>Hi Steve<br><br>Yes, that was my intention what you mentioned. The displayed image is controlled by a single (aggregated) attribute 'zoomcode' provided by the data source (see the corresponding <a href="http://vbkto.dyndns.org/cgi-bin/cluster.map">mapfile</a> for the details). But this configuration is highly customizable. For example you can define a group expression (which evaluates to a string value) to define which shapes should be considered to belong to the same group at all. This group value is also exposed as an attribute (Cluster:Group) which can also be used for classification/labeling.<br>
<br>Another powerful option is to define a filter expression which evaluates to a boolean value and defines whether the current shape should be added to a group or not. By using this, you may exclude certain shapes to take part in the clustering, or you may define a filter expression like: "([Cluster:FeatureCount]<20)" to limit the number of the features for each cluster.<br>
<br>The greatest challenge with the clustering is to select the most feasible rank expression by using a reasonable combination of the position offset, variance and the feature count. This rank expression determines which tentative clusters are selected and added to the list of the finalized clusters. We may probably expose some weighting coefficients for the user to provide better control on this.<br>
<br>I agree that a good documentation will also be required to enumerate all aspects.<br><br><br>Best regards,<br><br>Tamas<br></div></div><br><div style="visibility: hidden; left: -5000px; position: absolute; z-index: 9999; padding: 0px; margin-left: 0px; margin-top: 0px; overflow: hidden; word-wrap: break-word; color: black; font-size: 10px; text-align: left; line-height: 130%;" id="avg_ls_inline_popup">
</div>