<div class="gmail_quote">2010/7/30 Bill Allombert <span dir="ltr"><<a href="mailto:Bill.Allombert@math.u-bordeaux1.fr">Bill.Allombert@math.u-bordeaux1.fr</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="im">
<br>
</div>Generally, the answer is no, because data mining would violate the privacy expectation of the<br>
user. For example, you could use data-mining to desanonymize partially the data.<br>
<br></blockquote><div><br>By no means I intend to violate users privacy.<br><br>What I am thinking about is having some sort of package recommender system that would base its recommendation on the real use of packages. Thus I would use popcon data base to deduce what packages usually come along with other packages, things like that.<br>
<br>I'm interested in how packages relate to each other in adition to package dependencies, recommends and conflicts, which we already know since they are defined by the mantainer. <br><br>My focus would be the "use driven" package relationship.<br>
<br>[ ]'s<br><br>Tassia.<br></div></div>