Add files via upload

lectures/Week7.ipynb

@@ -317,12 +317,12 @@
    "metadata": {},
    "source": [
     "*Exercise 7*: Community detection.\n",
-    "> * Consider the BotW network. \n",
+    "> * Consider the BotW network (GCC as previously)\n",
     "> * Use [the Python Louvain-algorithm implementation](http://perso.crans.org/aynaud/communities/) to find communities in the BotW network. Report the value of modularity found by the algorithm. Is it higher or lower than what you found above for races as communities? What does this comparison reveal about them?\n",
     "    >   * **Note**: This implementation is also available as Anaconda package. Install with `conda` as expained [here](https://anaconda.org/auto/python-louvain). \n",
     "    >   * You can also try the *Infomap* algorithm instead if you're curious. Go to [this page](http://www.mapequation.org/code.html) and search for 'python'. It's harder to install, but a better community detection algorithm.\n",
     ">\n",
-    "> * Visualize the network, using the Force Atlas algorithm (see Lecture 5, exercise 2). This time assign each node a different color based on their _community_. Describe the structure you observe.\n",
+    "> * Visualize the network, using the Force Atlas algorithm (see Lecture 5, exercise 3). This time assign each node a different color based on their _community_. Describe the structure you observe.\n",
     "> * Compare the communities found by your algorithm with the races:\n",
     ">     * create a matrix $\\mathbf{D}$ with dimension ($R \\times C$), where $R$ is the number of races and $C$ is the number of communities. \n",
     ">     * set the entry $\\mathbf{D}_{i,j}$ to be the number of nodes that race $i$ has in common with community $j$. \n",