UCM_results_analysis_R.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "library(\"ggplot2\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<caption>A data.frame: 23 × 11</caption>\n",
       "<thead>\n",
       "\t<tr><th scope=col>ID</th><th scope=col>dataset</th><th scope=col>attributes</th><th scope=col>instances</th><th scope=col>classes</th><th scope=col>mv_acc</th><th scope=col>ucm_acc</th><th scope=col>wmv_acc</th><th scope=col>mv_f1</th><th scope=col>ucm_f1</th><th scope=col>wmv_f1</th></tr>\n",
       "\t<tr><th scope=col>&lt;int&gt;</th><th scope=col>&lt;fct&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;int&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><td> 1</td><td>abalone        </td><td>  8</td><td> 4177</td><td>29</td><td>26.56</td><td>26.41</td><td>26.03</td><td>23.66</td><td>23.41</td><td>23.48</td></tr>\n",
       "\t<tr><td> 2</td><td>anneal         </td><td> 38</td><td>  798</td><td> 6</td><td>92.60</td><td>91.35</td><td>92.22</td><td>92.64</td><td>91.11</td><td>92.27</td></tr>\n",
       "\t<tr><td> 3</td><td>arrhythmia     </td><td>279</td><td>  452</td><td>16</td><td>66.74</td><td>65.41</td><td>66.08</td><td>58.04</td><td>55.62</td><td>57.03</td></tr>\n",
       "\t<tr><td> 4</td><td>audiology      </td><td> 69</td><td>  226</td><td>24</td><td>82.23</td><td>78.72</td><td>82.25</td><td>78.47</td><td>73.88</td><td>78.70</td></tr>\n",
       "\t<tr><td> 5</td><td>breast-cancer  </td><td>  9</td><td>  286</td><td> 2</td><td>71.55</td><td>70.48</td><td>70.18</td><td>67.95</td><td>67.31</td><td>67.96</td></tr>\n",
       "\t<tr><td> 6</td><td>breast-cancer-w</td><td>  9</td><td>  699</td><td> 2</td><td>96.85</td><td>96.85</td><td>97.00</td><td>96.87</td><td>96.87</td><td>97.01</td></tr>\n",
       "\t<tr><td> 7</td><td>car            </td><td>  6</td><td> 1728</td><td> 4</td><td>88.72</td><td>89.35</td><td>87.50</td><td>89.18</td><td>89.59</td><td>87.67</td></tr>\n",
       "\t<tr><td> 8</td><td>crx            </td><td> 15</td><td>  690</td><td> 2</td><td>84.18</td><td>84.18</td><td>83.46</td><td>82.74</td><td>83.05</td><td>82.76</td></tr>\n",
       "\t<tr><td> 9</td><td>dermatology    </td><td> 34</td><td>  366</td><td> 6</td><td>97.24</td><td>96.97</td><td>97.24</td><td>97.23</td><td>96.94</td><td>97.21</td></tr>\n",
       "\t<tr><td>10</td><td>ecoli          </td><td>  7</td><td>  336</td><td> 4</td><td>86.54</td><td>86.25</td><td>86.55</td><td>85.62</td><td>85.28</td><td>85.66</td></tr>\n",
       "\t<tr><td>11</td><td>glass          </td><td> 10</td><td>  214</td><td> 7</td><td>66.28</td><td>65.76</td><td>70.48</td><td>61.50</td><td>60.56</td><td>66.45</td></tr>\n",
       "\t<tr><td>12</td><td>ionosphere     </td><td> 34</td><td>  351</td><td> 2</td><td>92.86</td><td>91.14</td><td>90.29</td><td>92.66</td><td>90.80</td><td>89.88</td></tr>\n",
       "\t<tr><td>13</td><td>iris           </td><td>  5</td><td>  150</td><td> 3</td><td>95.33</td><td>95.33</td><td>95.33</td><td>95.29</td><td>95.29</td><td>95.29</td></tr>\n",
       "\t<tr><td>14</td><td>kr-vs-kp       </td><td> 36</td><td> 3196</td><td> 2</td><td>96.87</td><td>96.40</td><td>95.93</td><td>96.86</td><td>96.37</td><td>95.90</td></tr>\n",
       "\t<tr><td>15</td><td>labor-neg      </td><td> 16</td><td>   57</td><td> 2</td><td>94.33</td><td>94.33</td><td>96.33</td><td>94.19</td><td>94.19</td><td>96.19</td></tr>\n",
       "\t<tr><td>16</td><td>letter         </td><td> 16</td><td>20000</td><td>26</td><td>92.36</td><td>91.39</td><td>94.76</td><td>92.41</td><td>91.43</td><td>94.78</td></tr>\n",
       "\t<tr><td>17</td><td>liver disorders</td><td>  6</td><td>  345</td><td> 2</td><td>72.94</td><td>72.06</td><td>73.24</td><td>72.81</td><td>71.23</td><td>72.69</td></tr>\n",
       "\t<tr><td>18</td><td>lymphography   </td><td> 20</td><td>  148</td><td> 4</td><td>80.95</td><td>80.29</td><td>81.67</td><td>79.71</td><td>78.79</td><td>80.60</td></tr>\n",
       "\t<tr><td>19</td><td>nursery        </td><td>  8</td><td>12960</td><td> 5</td><td>90.57</td><td>90.90</td><td>89.29</td><td>89.76</td><td>90.10</td><td>88.64</td></tr>\n",
       "\t<tr><td>20</td><td>page-blocks    </td><td> 10</td><td> 5473</td><td> 5</td><td>96.05</td><td>96.18</td><td>96.03</td><td>95.57</td><td>95.76</td><td>95.71</td></tr>\n",
       "\t<tr><td>21</td><td>segment        </td><td> 21</td><td> 2310</td><td> 7</td><td>95.93</td><td>96.32</td><td>96.19</td><td>95.87</td><td>96.28</td><td>96.14</td></tr>\n",
       "\t<tr><td>22</td><td>sonar          </td><td>208</td><td>   60</td><td> 2</td><td>66.71</td><td>66.74</td><td>67.64</td><td>65.39</td><td>65.87</td><td>66.96</td></tr>\n",
       "\t<tr><td>23</td><td>spambase       </td><td> 57</td><td> 4601</td><td> 2</td><td>93.91</td><td>93.98</td><td>93.83</td><td>93.88</td><td>93.96</td><td>93.81</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "A data.frame: 23 × 11\n",
       "\\begin{tabular}{lllllllllll}\n",
       " ID & dataset & attributes & instances & classes & mv\\_acc & ucm\\_acc & wmv\\_acc & mv\\_f1 & ucm\\_f1 & wmv\\_f1\\\\\n",
       " <int> & <fct> & <int> & <int> & <int> & <dbl> & <dbl> & <dbl> & <dbl> & <dbl> & <dbl>\\\\\n",
       "\\hline\n",
       "\t  1 & abalone         &   8 &  4177 & 29 & 26.56 & 26.41 & 26.03 & 23.66 & 23.41 & 23.48\\\\\n",
       "\t  2 & anneal          &  38 &   798 &  6 & 92.60 & 91.35 & 92.22 & 92.64 & 91.11 & 92.27\\\\\n",
       "\t  3 & arrhythmia      & 279 &   452 & 16 & 66.74 & 65.41 & 66.08 & 58.04 & 55.62 & 57.03\\\\\n",
       "\t  4 & audiology       &  69 &   226 & 24 & 82.23 & 78.72 & 82.25 & 78.47 & 73.88 & 78.70\\\\\n",
       "\t  5 & breast-cancer   &   9 &   286 &  2 & 71.55 & 70.48 & 70.18 & 67.95 & 67.31 & 67.96\\\\\n",
       "\t  6 & breast-cancer-w &   9 &   699 &  2 & 96.85 & 96.85 & 97.00 & 96.87 & 96.87 & 97.01\\\\\n",
       "\t  7 & car             &   6 &  1728 &  4 & 88.72 & 89.35 & 87.50 & 89.18 & 89.59 & 87.67\\\\\n",
       "\t  8 & crx             &  15 &   690 &  2 & 84.18 & 84.18 & 83.46 & 82.74 & 83.05 & 82.76\\\\\n",
       "\t  9 & dermatology     &  34 &   366 &  6 & 97.24 & 96.97 & 97.24 & 97.23 & 96.94 & 97.21\\\\\n",
       "\t 10 & ecoli           &   7 &   336 &  4 & 86.54 & 86.25 & 86.55 & 85.62 & 85.28 & 85.66\\\\\n",
       "\t 11 & glass           &  10 &   214 &  7 & 66.28 & 65.76 & 70.48 & 61.50 & 60.56 & 66.45\\\\\n",
       "\t 12 & ionosphere      &  34 &   351 &  2 & 92.86 & 91.14 & 90.29 & 92.66 & 90.80 & 89.88\\\\\n",
       "\t 13 & iris            &   5 &   150 &  3 & 95.33 & 95.33 & 95.33 & 95.29 & 95.29 & 95.29\\\\\n",
       "\t 14 & kr-vs-kp        &  36 &  3196 &  2 & 96.87 & 96.40 & 95.93 & 96.86 & 96.37 & 95.90\\\\\n",
       "\t 15 & labor-neg       &  16 &    57 &  2 & 94.33 & 94.33 & 96.33 & 94.19 & 94.19 & 96.19\\\\\n",
       "\t 16 & letter          &  16 & 20000 & 26 & 92.36 & 91.39 & 94.76 & 92.41 & 91.43 & 94.78\\\\\n",
       "\t 17 & liver disorders &   6 &   345 &  2 & 72.94 & 72.06 & 73.24 & 72.81 & 71.23 & 72.69\\\\\n",
       "\t 18 & lymphography    &  20 &   148 &  4 & 80.95 & 80.29 & 81.67 & 79.71 & 78.79 & 80.60\\\\\n",
       "\t 19 & nursery         &   8 & 12960 &  5 & 90.57 & 90.90 & 89.29 & 89.76 & 90.10 & 88.64\\\\\n",
       "\t 20 & page-blocks     &  10 &  5473 &  5 & 96.05 & 96.18 & 96.03 & 95.57 & 95.76 & 95.71\\\\\n",
       "\t 21 & segment         &  21 &  2310 &  7 & 95.93 & 96.32 & 96.19 & 95.87 & 96.28 & 96.14\\\\\n",
       "\t 22 & sonar           & 208 &    60 &  2 & 66.71 & 66.74 & 67.64 & 65.39 & 65.87 & 66.96\\\\\n",
       "\t 23 & spambase        &  57 &  4601 &  2 & 93.91 & 93.98 & 93.83 & 93.88 & 93.96 & 93.81\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "A data.frame: 23 × 11\n",
       "\n",
       "| ID &lt;int&gt; | dataset &lt;fct&gt; | attributes &lt;int&gt; | instances &lt;int&gt; | classes &lt;int&gt; | mv_acc &lt;dbl&gt; | ucm_acc &lt;dbl&gt; | wmv_acc &lt;dbl&gt; | mv_f1 &lt;dbl&gt; | ucm_f1 &lt;dbl&gt; | wmv_f1 &lt;dbl&gt; |\n",
       "|---|---|---|---|---|---|---|---|---|---|---|\n",
       "|  1 | abalone         |   8 |  4177 | 29 | 26.56 | 26.41 | 26.03 | 23.66 | 23.41 | 23.48 |\n",
       "|  2 | anneal          |  38 |   798 |  6 | 92.60 | 91.35 | 92.22 | 92.64 | 91.11 | 92.27 |\n",
       "|  3 | arrhythmia      | 279 |   452 | 16 | 66.74 | 65.41 | 66.08 | 58.04 | 55.62 | 57.03 |\n",
       "|  4 | audiology       |  69 |   226 | 24 | 82.23 | 78.72 | 82.25 | 78.47 | 73.88 | 78.70 |\n",
       "|  5 | breast-cancer   |   9 |   286 |  2 | 71.55 | 70.48 | 70.18 | 67.95 | 67.31 | 67.96 |\n",
       "|  6 | breast-cancer-w |   9 |   699 |  2 | 96.85 | 96.85 | 97.00 | 96.87 | 96.87 | 97.01 |\n",
       "|  7 | car             |   6 |  1728 |  4 | 88.72 | 89.35 | 87.50 | 89.18 | 89.59 | 87.67 |\n",
       "|  8 | crx             |  15 |   690 |  2 | 84.18 | 84.18 | 83.46 | 82.74 | 83.05 | 82.76 |\n",
       "|  9 | dermatology     |  34 |   366 |  6 | 97.24 | 96.97 | 97.24 | 97.23 | 96.94 | 97.21 |\n",
       "| 10 | ecoli           |   7 |   336 |  4 | 86.54 | 86.25 | 86.55 | 85.62 | 85.28 | 85.66 |\n",
       "| 11 | glass           |  10 |   214 |  7 | 66.28 | 65.76 | 70.48 | 61.50 | 60.56 | 66.45 |\n",
       "| 12 | ionosphere      |  34 |   351 |  2 | 92.86 | 91.14 | 90.29 | 92.66 | 90.80 | 89.88 |\n",
       "| 13 | iris            |   5 |   150 |  3 | 95.33 | 95.33 | 95.33 | 95.29 | 95.29 | 95.29 |\n",
       "| 14 | kr-vs-kp        |  36 |  3196 |  2 | 96.87 | 96.40 | 95.93 | 96.86 | 96.37 | 95.90 |\n",
       "| 15 | labor-neg       |  16 |    57 |  2 | 94.33 | 94.33 | 96.33 | 94.19 | 94.19 | 96.19 |\n",
       "| 16 | letter          |  16 | 20000 | 26 | 92.36 | 91.39 | 94.76 | 92.41 | 91.43 | 94.78 |\n",
       "| 17 | liver disorders |   6 |   345 |  2 | 72.94 | 72.06 | 73.24 | 72.81 | 71.23 | 72.69 |\n",
       "| 18 | lymphography    |  20 |   148 |  4 | 80.95 | 80.29 | 81.67 | 79.71 | 78.79 | 80.60 |\n",
       "| 19 | nursery         |   8 | 12960 |  5 | 90.57 | 90.90 | 89.29 | 89.76 | 90.10 | 88.64 |\n",
       "| 20 | page-blocks     |  10 |  5473 |  5 | 96.05 | 96.18 | 96.03 | 95.57 | 95.76 | 95.71 |\n",
       "| 21 | segment         |  21 |  2310 |  7 | 95.93 | 96.32 | 96.19 | 95.87 | 96.28 | 96.14 |\n",
       "| 22 | sonar           | 208 |    60 |  2 | 66.71 | 66.74 | 67.64 | 65.39 | 65.87 | 66.96 |\n",
       "| 23 | spambase        |  57 |  4601 |  2 | 93.91 | 93.98 | 93.83 | 93.88 | 93.96 | 93.81 |\n",
       "\n"
      ],
      "text/plain": [
       "   ID dataset         attributes instances classes mv_acc ucm_acc wmv_acc mv_f1\n",
       "1   1 abalone           8         4177     29      26.56  26.41   26.03   23.66\n",
       "2   2 anneal           38          798      6      92.60  91.35   92.22   92.64\n",
       "3   3 arrhythmia      279          452     16      66.74  65.41   66.08   58.04\n",
       "4   4 audiology        69          226     24      82.23  78.72   82.25   78.47\n",
       "5   5 breast-cancer     9          286      2      71.55  70.48   70.18   67.95\n",
       "6   6 breast-cancer-w   9          699      2      96.85  96.85   97.00   96.87\n",
       "7   7 car               6         1728      4      88.72  89.35   87.50   89.18\n",
       "8   8 crx              15          690      2      84.18  84.18   83.46   82.74\n",
       "9   9 dermatology      34          366      6      97.24  96.97   97.24   97.23\n",
       "10 10 ecoli             7          336      4      86.54  86.25   86.55   85.62\n",
       "11 11 glass            10          214      7      66.28  65.76   70.48   61.50\n",
       "12 12 ionosphere       34          351      2      92.86  91.14   90.29   92.66\n",
       "13 13 iris              5          150      3      95.33  95.33   95.33   95.29\n",
       "14 14 kr-vs-kp         36         3196      2      96.87  96.40   95.93   96.86\n",
       "15 15 labor-neg        16           57      2      94.33  94.33   96.33   94.19\n",
       "16 16 letter           16        20000     26      92.36  91.39   94.76   92.41\n",
       "17 17 liver disorders   6          345      2      72.94  72.06   73.24   72.81\n",
       "18 18 lymphography     20          148      4      80.95  80.29   81.67   79.71\n",
       "19 19 nursery           8        12960      5      90.57  90.90   89.29   89.76\n",
       "20 20 page-blocks      10         5473      5      96.05  96.18   96.03   95.57\n",
       "21 21 segment          21         2310      7      95.93  96.32   96.19   95.87\n",
       "22 22 sonar           208           60      2      66.71  66.74   67.64   65.39\n",
       "23 23 spambase         57         4601      2      93.91  93.98   93.83   93.88\n",
       "   ucm_f1 wmv_f1\n",
       "1  23.41  23.48 \n",
       "2  91.11  92.27 \n",
       "3  55.62  57.03 \n",
       "4  73.88  78.70 \n",
       "5  67.31  67.96 \n",
       "6  96.87  97.01 \n",
       "7  89.59  87.67 \n",
       "8  83.05  82.76 \n",
       "9  96.94  97.21 \n",
       "10 85.28  85.66 \n",
       "11 60.56  66.45 \n",
       "12 90.80  89.88 \n",
       "13 95.29  95.29 \n",
       "14 96.37  95.90 \n",
       "15 94.19  96.19 \n",
       "16 91.43  94.78 \n",
       "17 71.23  72.69 \n",
       "18 78.79  80.60 \n",
       "19 90.10  88.64 \n",
       "20 95.76  95.71 \n",
       "21 96.28  96.14 \n",
       "22 65.87  66.96 \n",
       "23 93.96  93.81 "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "performance_df <- read.csv(\"performance.csv\")\n",
    "performance_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABaAAAAHgCAIAAADc1V3gAAAACXBIWXMAABJ0AAASdAHeZh94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\nAAAAAHSPgAMAAAAAAOje/wFKbL5UX8SQDQAAAABJRU5ErkJggg==",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 240,
       "width": 720
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "performance_df$ucm_mv_acc_diff <- performance_df$ucm_acc - performance_df$mv_acc\n",
    "\n",
    "options(repr.plot.width=12, repr.plot.height=4)\n",
    "\n",
    "p <- ggplot(data=performance_df, aes(x=ID, y=ucm_mv_acc_diff)) + \n",
    "    geom_bar(stat=\"identity\", color=\"#8c9bb5\", fill=\"#8c9bb5\") +\n",
    "    xlab(\"Dataset ID\") + ylab(\"Accuracy Difference\") +\n",
    "    theme(text = element_text(size=18))\n",
    "p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "R",
   "language": "R",
   "name": "ir"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "3.3.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}