diff --git a/2017/09-clustering/cl_francisco_carlos.ipynb b/2017/09-clustering/cl_francisco_carlos.ipynb
new file mode 100644
index 0000000..0afb269
--- /dev/null
+++ b/2017/09-clustering/cl_francisco_carlos.ipynb
@@ -0,0 +1,461 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": true
+ },
+ "source": [
+ "## Exercícios\n",
+ "\n",
+ "1 - Aplique os algoritmos K-means [1] e AgglomerativeClustering [2] em qualquer dataset que você desejar (recomendação: iris). Compare os resultados utilizando métricas de avaliação de clusteres (completeness e homogeneity, por exemplo) [3].\n",
+ "\n",
+ "* [1] http://scikit-learn.org/stable/modules/clustering.html#k-means\n",
+ "\n",
+ "* [2] http://scikit-learn.org/0.17/modules/clustering.html#hierarchical-clustering\n",
+ "\n",
+ "* [3] http://scikit-learn.org/stable/modules/clustering.html#clustering-evaluation\n",
+ "\n",
+ "2 - Qual o valor de K (número de clusteres) você escolheu para a questão anterior? Desenvolva o Método do Cotovelo (não utilizar lib!) e descubra o K mais adequado. Após descobrir, aplique novamente o K-means com o K adequado. \n",
+ "\n",
+ "* Ajuda: atributos do [k-means](http://scikit-learn.org/0.17/modules/generated/sklearn.cluster.KMeans.html#sklearn.cluster.KMeans)\n",
+ "\n",
+ "3 - Após a questão 2, você aplicou o algoritmo com K apropriado. Refaça o cálculo das métricas de acordo com os resultados de clusters obtidos com a questão anterior e verifique se o resultado melhorou.\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "collapsed": true
+ },
+ "source": [
+ "## Parte 1"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 166,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "#Bibliotecas\n",
+ "import pandas as pd\n",
+ "import numpy as np\n",
+ "import matplotlib.pyplot as plt\n",
+ "from sklearn import datasets\n",
+ "from sklearn.decomposition import PCA\n",
+ "from sklearn.preprocessing import normalize\n",
+ "from sklearn.cluster import KMeans\n",
+ "from sklearn import metrics"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 167,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " 0 | \n",
+ " 1 | \n",
+ " 2 | \n",
+ " 3 | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " | 0 | \n",
+ " 5.1 | \n",
+ " 3.5 | \n",
+ " 1.4 | \n",
+ " 0.2 | \n",
+ "
\n",
+ " \n",
+ " | 1 | \n",
+ " 4.9 | \n",
+ " 3.0 | \n",
+ " 1.4 | \n",
+ " 0.2 | \n",
+ "
\n",
+ " \n",
+ " | 2 | \n",
+ " 4.7 | \n",
+ " 3.2 | \n",
+ " 1.3 | \n",
+ " 0.2 | \n",
+ "
\n",
+ " \n",
+ " | 3 | \n",
+ " 4.6 | \n",
+ " 3.1 | \n",
+ " 1.5 | \n",
+ " 0.2 | \n",
+ "
\n",
+ " \n",
+ " | 4 | \n",
+ " 5.0 | \n",
+ " 3.6 | \n",
+ " 1.4 | \n",
+ " 0.2 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "
\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " | \n",
+ " 0 | \n",
+ " 1 | \n",
+ " 2 | \n",
+ " 3 | \n",
+ "
\n",
+ " \n",
+ " \n",
+ " \n",
+ " | 0 | \n",
+ " 0.401886 | \n",
+ " 0.275804 | \n",
+ " 0.110322 | \n",
+ " 0.015760 | \n",
+ "
\n",
+ " \n",
+ " | 1 | \n",
+ " 0.414066 | \n",
+ " 0.253510 | \n",
+ " 0.118305 | \n",
+ " 0.016901 | \n",
+ "
\n",
+ " \n",
+ " | 2 | \n",
+ " 0.402667 | \n",
+ " 0.274156 | \n",
+ " 0.111376 | \n",
+ " 0.017135 | \n",
+ "
\n",
+ " \n",
+ " | 3 | \n",
+ " 0.400015 | \n",
+ " 0.269575 | \n",
+ " 0.130440 | \n",
+ " 0.017392 | \n",
+ "
\n",
+ " \n",
+ " | 4 | \n",
+ " 0.395482 | \n",
+ " 0.284747 | \n",
+ " 0.110735 | \n",
+ " 0.015819 | \n",
+ "
\n",
+ " \n",
+ "
\n",
+ "