|
|
The visualization allows you to see the connections and weightings of thousands of URLs at a glance (depending on when your computer gives up). In addition, it is not possible to show in tables how closely individual URLs are interwoven with one another. This is possible with Gephi. How does Gephi work? How the software works exactly should not be part of the article. On the one hand, this goes beyond the scope and on the other hand, others have already made the effort. It's best to get information from one of these sources: For friends of the paper industry, we recommend issue 44 of the Boosting website (07-08/2017), pages 28-42.
There is a very good explanation in English at Searchengineland. The presentation by India Car Owner Phone Number List Sebastian Erlhofer from mindshape (PDF) is also very helpful from slide 15, but without written explanations. However, this is not a prerequisite for understanding this article. My goal is to recommend approaches to evaluation to you - or to get you excited about the tool in the first place. What can you do in Gephi? Basically, you have the following options to customize your structure of links: Adjust the distribution or spread of the network (“layout”) Change the size of URLs (≈nodes). Change the color of URLs Change the color of links (≈edges).
It is recommended that you think carefully about which metrics you want to use for your URLs before you start. You then upload this as a node table with columns for these metrics or attributes. Numerical values should not be imported as a “string”, but rather, for example, as a “long” in the window after the upload ( more information on possible data types ). Otherwise you won't be able to set up gradual weighting or coloring according to these values later. Let's start with the most obvious one: coloring by topic - or by page type or or or.
|
|
發表於 14:00:31