topsis simple example

notebooks/topsis_example.ipynb

@@ -0,0 +1,180 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Simple TOPSIS example\n",
+    "\n",
+    "Let make a simple topsis example"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "from skcriteria import topsis"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "- Every candidate will be a posible alternative inside the method. I call every alternative as A1, A2...\n",
+    "- Every alternative has the same set of criteria C1, C2...\n",
+    "- Every criteria represent a numerical value of some measuremente of some characteristic of the alternative\n",
+    "  for example Age, If they has talks or not, how far away of the venue they live, etc.\n",
+    "- Every criteria can be maximized or minimiszed but not the two of them.\n",
+    "\n",
+    "For example:\n",
+    "\n",
+    "- **Ivan:** Speaker, Lives in Floripa, Give a talk last year\n",
+    "- **Raniere:** Speaker, Lives in Campinhas, Give a talk last year\n",
+    "- **Juan:** Not Speaker, Lives in Cordoba, Not Give a talk last year\n",
+    "\n",
+    "The first criteria is a float equals to how many hours they gona talk in the conference (tlaks increment 1 and tutotiral increments 3)\n",
+    "Now if we think the distance between where they live and where is the venue we can have the second criteria.\n",
+    "The last one is a boollean (1: they give a talk in scipy2015, 0: Nope)\n",
+    "\n",
+    "So the matrix is like"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "mtx = [\n",
+    "    [1., 0., 1], # ivan\n",
+    "    [4, 770., 1], # raniere\n",
+    "    [0, 1885., 0], # juan\n",
+    "]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now we need the criteria:\n",
+    "    - Is better to to give more hours of talk (maximization criteria)\n",
+    "    - Is better to be away from the vennue (maximization)\n",
+    "    - Is better to give a talk the last year (max)\n",
+    "    \n",
+    "So the criteria array must be like"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "criteria = [1, 1, 1] # minimization is '-1'"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The last thing is give a series of weight of every criteria (this is optional) for example we can think that the first criteria have the 45% of importance, the second criteria the 40% and the last one only the 15%"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "weights = [45, 40, 15]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now lets apply [TOPSIS](https://en.wikipedia.org/wiki/TOPSIS)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "rank, points = topsis.topsis(mtx, criteria, weights)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The `rank` variable contains the ranking of every order alternative and the `points` is the distance of every alternative to the ideal and anti-ideal points (read this for understand topsis: https://en.wikipedia.org/wiki/TOPSIS)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1 raniere\n",
+      "2 juan\n",
+      "3 ivan\n"
+     ]
+    }
+   ],
+   "source": [
+    "for r, name in sorted(zip(rank, [\"ivan\", \"raniere\", \"juan\"])):\n",
+    "    print r, name"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 2",
+   "language": "python",
+   "name": "python2"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.5+"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
+}