removed unnecessary picture

4. K-means clustering and unsupervised learning.ipynb

@@ -85,7 +85,7 @@
     "![alt-text](http://pubs.rsc.org/services/images/RSCpubs.ePlatform.Service.FreeContent.ImageService.svc/ImageService/Articleimage/2012/AN/c2an16122b/c2an16122b-f3.gif \"k-means clustering algorithm\")\n",
     "\n",
     "##### (a) \n",
-    "Okay, our data seems to have a distinct structure. Let’s use this information to initialize our k-means algorithm with k = 3 clusters. Right now we’re operating under the assumption that we know how many clusters we want, but we’ll go into more detail about relaxing this assumption and how to choose “the best possible k” at the end of the workshop. \n",
+    "Okay, our data seems to have some sort of underlying structure. Let’s use this information to initialize our k-means algorithm with k = 3 clusters. Right now we’re operating under the assumption that we know how many clusters we want, but we’ll go into more detail about relaxing this assumption and how to choose “the best possible k” at the end of the workshop. \n",
     "\n",
     "##### (b) \n",
     "k-Means works like this. First we “throw down” three random cluster centroids. We initialize these clusters randomly because every iteration of k-means will \"correct\" them towards the right clusters. Since we are heading to a correct answer anyway, we don't really care about where we start. \n",