Pre-Visualization.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pre-Visualization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Read Pre-Processed data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data Shape:  (49074, 300)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>...</th>\n",
       "      <th>291</th>\n",
       "      <th>292</th>\n",
       "      <th>293</th>\n",
       "      <th>294</th>\n",
       "      <th>295</th>\n",
       "      <th>296</th>\n",
       "      <th>297</th>\n",
       "      <th>298</th>\n",
       "      <th>299</th>\n",
       "      <th>300</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9254293.0</th>\n",
       "      <td>-0.069380</td>\n",
       "      <td>0.043574</td>\n",
       "      <td>-0.000726</td>\n",
       "      <td>0.048484</td>\n",
       "      <td>-0.025510</td>\n",
       "      <td>0.006262</td>\n",
       "      <td>-0.036216</td>\n",
       "      <td>-0.022857</td>\n",
       "      <td>-0.020438</td>\n",
       "      <td>0.010014</td>\n",
       "      <td>...</td>\n",
       "      <td>0.029088</td>\n",
       "      <td>0.003316</td>\n",
       "      <td>-0.036513</td>\n",
       "      <td>0.029949</td>\n",
       "      <td>-0.013912</td>\n",
       "      <td>-0.062233</td>\n",
       "      <td>-0.043268</td>\n",
       "      <td>-0.008448</td>\n",
       "      <td>-0.042122</td>\n",
       "      <td>0.056087</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9254046.0</th>\n",
       "      <td>-0.068804</td>\n",
       "      <td>0.047289</td>\n",
       "      <td>-0.004199</td>\n",
       "      <td>0.047923</td>\n",
       "      <td>-0.023679</td>\n",
       "      <td>0.008813</td>\n",
       "      <td>-0.034846</td>\n",
       "      <td>-0.021046</td>\n",
       "      <td>-0.021554</td>\n",
       "      <td>0.008313</td>\n",
       "      <td>...</td>\n",
       "      <td>0.028636</td>\n",
       "      <td>0.000343</td>\n",
       "      <td>-0.039810</td>\n",
       "      <td>0.032698</td>\n",
       "      <td>-0.010946</td>\n",
       "      <td>-0.064177</td>\n",
       "      <td>-0.043474</td>\n",
       "      <td>-0.011102</td>\n",
       "      <td>-0.042165</td>\n",
       "      <td>0.058833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9254708.0</th>\n",
       "      <td>-0.064731</td>\n",
       "      <td>0.041257</td>\n",
       "      <td>-0.002079</td>\n",
       "      <td>0.050523</td>\n",
       "      <td>-0.021690</td>\n",
       "      <td>0.010098</td>\n",
       "      <td>-0.031933</td>\n",
       "      <td>-0.018337</td>\n",
       "      <td>-0.019479</td>\n",
       "      <td>0.010856</td>\n",
       "      <td>...</td>\n",
       "      <td>0.027773</td>\n",
       "      <td>0.001253</td>\n",
       "      <td>-0.037309</td>\n",
       "      <td>0.031383</td>\n",
       "      <td>-0.012615</td>\n",
       "      <td>-0.061032</td>\n",
       "      <td>-0.041807</td>\n",
       "      <td>-0.008018</td>\n",
       "      <td>-0.043039</td>\n",
       "      <td>0.055232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9254085.0</th>\n",
       "      <td>-0.065514</td>\n",
       "      <td>0.047664</td>\n",
       "      <td>-0.003868</td>\n",
       "      <td>0.048916</td>\n",
       "      <td>-0.019847</td>\n",
       "      <td>0.009603</td>\n",
       "      <td>-0.032904</td>\n",
       "      <td>-0.015890</td>\n",
       "      <td>-0.018976</td>\n",
       "      <td>0.007979</td>\n",
       "      <td>...</td>\n",
       "      <td>0.026406</td>\n",
       "      <td>-0.002981</td>\n",
       "      <td>-0.041753</td>\n",
       "      <td>0.033225</td>\n",
       "      <td>-0.010989</td>\n",
       "      <td>-0.062158</td>\n",
       "      <td>-0.039481</td>\n",
       "      <td>-0.008778</td>\n",
       "      <td>-0.039722</td>\n",
       "      <td>0.055824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9254330.0</th>\n",
       "      <td>-0.068589</td>\n",
       "      <td>0.048235</td>\n",
       "      <td>-0.003841</td>\n",
       "      <td>0.049804</td>\n",
       "      <td>-0.022346</td>\n",
       "      <td>0.009306</td>\n",
       "      <td>-0.033885</td>\n",
       "      <td>-0.017681</td>\n",
       "      <td>-0.017290</td>\n",
       "      <td>0.007400</td>\n",
       "      <td>...</td>\n",
       "      <td>0.026265</td>\n",
       "      <td>-0.001228</td>\n",
       "      <td>-0.040819</td>\n",
       "      <td>0.032922</td>\n",
       "      <td>-0.012522</td>\n",
       "      <td>-0.064146</td>\n",
       "      <td>-0.039414</td>\n",
       "      <td>-0.011086</td>\n",
       "      <td>-0.039173</td>\n",
       "      <td>0.058802</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 300 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                1         2         3         4         5         6    \\\n",
       "0                                                                       \n",
       "9254293.0 -0.069380  0.043574 -0.000726  0.048484 -0.025510  0.006262   \n",
       "9254046.0 -0.068804  0.047289 -0.004199  0.047923 -0.023679  0.008813   \n",
       "9254708.0 -0.064731  0.041257 -0.002079  0.050523 -0.021690  0.010098   \n",
       "9254085.0 -0.065514  0.047664 -0.003868  0.048916 -0.019847  0.009603   \n",
       "9254330.0 -0.068589  0.048235 -0.003841  0.049804 -0.022346  0.009306   \n",
       "\n",
       "                7         8         9         10   ...       291       292  \\\n",
       "0                                                  ...                       \n",
       "9254293.0 -0.036216 -0.022857 -0.020438  0.010014  ...  0.029088  0.003316   \n",
       "9254046.0 -0.034846 -0.021046 -0.021554  0.008313  ...  0.028636  0.000343   \n",
       "9254708.0 -0.031933 -0.018337 -0.019479  0.010856  ...  0.027773  0.001253   \n",
       "9254085.0 -0.032904 -0.015890 -0.018976  0.007979  ...  0.026406 -0.002981   \n",
       "9254330.0 -0.033885 -0.017681 -0.017290  0.007400  ...  0.026265 -0.001228   \n",
       "\n",
       "                293       294       295       296       297       298  \\\n",
       "0                                                                       \n",
       "9254293.0 -0.036513  0.029949 -0.013912 -0.062233 -0.043268 -0.008448   \n",
       "9254046.0 -0.039810  0.032698 -0.010946 -0.064177 -0.043474 -0.011102   \n",
       "9254708.0 -0.037309  0.031383 -0.012615 -0.061032 -0.041807 -0.008018   \n",
       "9254085.0 -0.041753  0.033225 -0.010989 -0.062158 -0.039481 -0.008778   \n",
       "9254330.0 -0.040819  0.032922 -0.012522 -0.064146 -0.039414 -0.011086   \n",
       "\n",
       "                299       300  \n",
       "0                              \n",
       "9254293.0 -0.042122  0.056087  \n",
       "9254046.0 -0.042165  0.058833  \n",
       "9254708.0 -0.043039  0.055232  \n",
       "9254085.0 -0.039722  0.055824  \n",
       "9254330.0 -0.039173  0.058802  \n",
       "\n",
       "[5 rows x 300 columns]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "#  Index 0 represents the award number which is unique to each document \n",
    "docs2vec = pd.read_csv('docs_vector.csv', index_col = 0, header=None)\n",
    "#docs2vec.set_index(docs2vec.columns[0])\n",
    "print('Data Shape: ', docs2vec.shape)\n",
    "docs2vec.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Normalization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn import preprocessing\n",
    "# df = preprocessing.normalize(docs2vec, norm='l2', axis=1, copy=False, return_norm=False)\n",
    "# df = pd.DataFrame(df)\n",
    "# print('Data Shape: ', df.shape)\n",
    "# df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PCA Decomposition for 2D pre-visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Variation Ratio:  [0.22646793 0.13580476]\n",
      "Data shape:  (49074, 2)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pc1</th>\n",
       "      <th>pc2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9254293.0</th>\n",
       "      <td>0.014243</td>\n",
       "      <td>-0.006449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9254046.0</th>\n",
       "      <td>-0.009191</td>\n",
       "      <td>-0.002264</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9254708.0</th>\n",
       "      <td>0.013431</td>\n",
       "      <td>-0.004799</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9254085.0</th>\n",
       "      <td>-0.010991</td>\n",
       "      <td>0.018573</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9254330.0</th>\n",
       "      <td>-0.009560</td>\n",
       "      <td>0.021950</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                pc1       pc2\n",
       "0                            \n",
       "9254293.0  0.014243 -0.006449\n",
       "9254046.0 -0.009191 -0.002264\n",
       "9254708.0  0.013431 -0.004799\n",
       "9254085.0 -0.010991  0.018573\n",
       "9254330.0 -0.009560  0.021950"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "pca = PCA(n_components=2)\n",
    "principalComponents = pca.fit_transform(docs2vec)\n",
    "print('Variation Ratio: ', pca.explained_variance_ratio_)\n",
    "principalDf = pd.DataFrame(data = principalComponents, columns = ['pc1', 'pc2'])\n",
    "principalDf.set_index(docs2vec.index,inplace=True)\n",
    "print('Data shape: ', principalDf.shape)\n",
    "principalDf.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Pre-Visualization: All Points"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "All data visulization after 2-component principal component decomposition."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "fig = plt.figure(figsize = (8,8))\n",
    "ax = fig.add_subplot(1,1,1) \n",
    "ax.set_xlabel('Principal Component 1', fontsize = 15)\n",
    "ax.set_ylabel('Principal Component 2', fontsize = 15)\n",
    "ax.set_title('2 component PCA', fontsize = 20)\n",
    "ax.scatter(principalDf['pc1'], principalDf['pc2'])\n",
    "ax.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These outlier points corresponds to not ture abstract. <br>\n",
    "Example Abstracts: <br>\n",
    "    9401662:  Bhowmik    This is an Individual Travel award.  ***' <br>\n",
    "    9254516: NNNNNNNNNNNNNNNNNNN//"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                pc1       pc2\n",
      "0                            \n",
      "9254516.0  0.601818 -0.035106\n",
      "9252605.0  0.021598  0.096890\n",
      "9417988.0  0.324703  0.012575\n",
      "9400192.0  0.342002 -0.036687\n",
      "9401662.0  0.188964 -0.095735\n",
      "9355771.0  0.299530 -0.006020\n",
      "9313269.0  0.349852  0.017075\n",
      "9313839.0  0.259327 -0.017730\n",
      "9315155.0  0.342002 -0.036687\n",
      "9315089.0  0.342002 -0.036687\n",
      "9312030.0  0.329727 -0.053656\n",
      "9316234.0  0.342002 -0.036687\n",
      "9318322.0  0.232119 -0.097058\n",
      "9318321.0  0.253980 -0.033251\n",
      "9153716.0  0.066292  0.098928\n",
      "9047965.0  0.001473 -0.077489\n"
     ]
    }
   ],
   "source": [
    "principalDf_Outlier = principalDf[(principalDf['pc1'] > 0.25) | (principalDf['pc2'] < -0.075) | (principalDf['pc2'] > 0.075)]\n",
    "print(principalDf_Outlier)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Remove outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Float64Index([9254516.0, 9252605.0, 9417988.0, 9400192.0, 9401662.0, 9355771.0,\n",
      "              9313269.0, 9313839.0, 9315155.0, 9315089.0, 9312030.0, 9316234.0,\n",
      "              9318322.0, 9318321.0, 9153716.0, 9047965.0],\n",
      "             dtype='float64', name=0)\n"
     ]
    }
   ],
   "source": [
    "print(principalDf_Outlier.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "principalDf_final = principalDf.drop(index=principalDf_Outlier.index)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Pre-Visualization: Sample Points"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "fig = plt.figure(figsize = (8,8))\n",
    "ax = fig.add_subplot(1,1,1) \n",
    "ax.set_xlabel('Principal Component 1', fontsize = 15)\n",
    "ax.set_ylabel('Principal Component 2', fontsize = 15)\n",
    "ax.set_title('2 component PCA', fontsize = 20)\n",
    "ax.scatter(principalDf_final['pc1'], principalDf_final['pc2'])\n",
    "ax.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sample data n = 1000 visulization after 2-component principal component decomposition."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "principalDf_sample = principalDf_final.sample(n =1000)\n",
    "fig = plt.figure(figsize = (8,8))\n",
    "ax = fig.add_subplot(1,1,1) \n",
    "ax.set_xlabel('Principal Component 1', fontsize = 15)\n",
    "ax.set_ylabel('Principal Component 2', fontsize = 15)\n",
    "ax.set_title('2 component PCA', fontsize = 20)\n",
    "#plt.xlim([-0.05, 0.1])\n",
    "#plt.ylim([-0.05, 0.1])\n",
    "ax.scatter(principalDf_sample['pc1'], principalDf_sample['pc2'])\n",
    "ax.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Remove outlier and write the final version of the file\n",
    "'docs_vector_final.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = docs2vec.drop(index=principalDf_Outlier.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv('docs_vector_final.csv',header=None)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}