python_03-notebook-55c9ffda55.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "3d56ef8f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-12-31T21:38:08.407774Z",
     "iopub.status.busy": "2024-12-31T21:38:08.407365Z",
     "iopub.status.idle": "2024-12-31T21:41:01.213569Z",
     "shell.execute_reply": "2024-12-31T21:41:01.212139Z"
    },
    "papermill": {
     "duration": 172.816183,
     "end_time": "2024-12-31T21:41:01.219110",
     "exception": false,
     "start_time": "2024-12-31T21:38:08.402927",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "checking the current working directory address ...\n",
      "/kaggle/working\n",
      "listing all files in the current working directory ...\n",
      "['__notebook__.ipynb']\n",
      "moving on to the practice code ...\n",
      "\n",
      "> TDM info summary ...\n",
      "\n",
      "<<TermDocumentMatrix (terms: 5, documents: 7162)>>\n",
      "Non-/sparse entries: 55017/-19207\n",
      "Sparsity           : -54%\n",
      "Maximal term length: 21\n",
      "Weighting          : term frequency (tf)\n",
      "Sample             :\n",
      "               Docs.\n",
      "Terms     1 _ one.txt  2 _ two.txt 3 _ three.txt 4 _ four.txt 5 _ five.txt\n",
      "\n",
      "... (skipping this matrix due to it being too big) ...\n",
      "\n",
      "\n",
      "> plotting TDM dendrogram ...\n",
      "\n",
      "TDM dendrogram graph saved as 'py-tdm_dendrogram.png'\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x700 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Frequent terms (appearing 35 or more times):\n",
      "'10'100694'11'121'15'17'2011'2012'2013'2014'2015'2016'2017'2018'2022'2023'access'accid'activ'addit'aerial'aerodynam'aerospac'agricultur'air'aircraft'airship'airspac'al'algorithm'also'among'analysi'anoth'appli'applic'approach'architectur'area'around'attack'authent'author'autonom'avail'avoid'balloon'base'benefit'blockchain'call'camera'capabl'capac'case'categori'caus'challeng'chang'china'civilian'coaxial'collect'collis'combin'commerci'common'communic'compani'compar'complex'comput'concept'concern'confer'configur'consid'control'cost'creat'critic'cruis'current'darvishpoor'data'deliveri'deploy'design'detect'develop'devic'differ'direct'discuss'drone'dual'duct'due'dynam'effect'effici'emerg'encrypt'endur'energi'engin'environ'et'even'exampl'exist'explor'fan'fig'first'flap'fli'flight'follow'forc'framework'futur'gcs'general'generat'global'govern'gps'ground'group'hardwar'hassanalian'helicopt'help'high'hover'howev'http'https'hybrid'id'ieee'imag'implement'import'improv'includ'increas'industri'inform'inspect'integr'intellig'intern'internet'intrus'investig'israel'issu'journal'june'key'laghari'land'law'learn'less'li'lift'lightweight'like'limit'low'made'magnus'main'make'manag'mani'manufactur'map'market'mass'may'mechan'method'micro'militari'mission'model'monitor'moreov'motion'motor'move'multirotor'need'network'new'number'numer'one'onlin'oper'part'particular'payload'perform'phase'physic'popular'potenti'power'pp'present'privaci'privat'problem'process'progress'propel'propos'protect'protocol'provid'public'quadrotor'race'rang'receiv'recent'reduc'regul'relat'remot'requir'rescu'research'result'retriev'review'risk'robot'rotat'rotor'rule'safeti'scheme'sci'scienc'search'section'secur'see'sensor'servic'sever'show'shown'signal'similar'sinc'small'softwar'solut'space'specif'speed'stabil'state'still'structur'studi'surfac'surveil'survey'swarm'system'tabl'tail'take'takeoff'techniqu'technolog'test'therefor'threat'three'thrust'tilt'time'torqu'two'type'ua'uass'uav'unit'univers'unman'use'user'util'various'vehicl'vertic'view'vtol'vulner'wang'way'well'wing'wireless'without'work'would'\n",
      "\n",
      "> DTM info summary ...\n",
      "\n",
      "<<DocumentTermMatrix (terms: 7162, documents: 5)>>\n",
      "Non-/sparse entries: 55017/-19207\n",
      "Sparsity           : -54%\n",
      "Maximal term length: 21\n",
      "Weighting          : term frequency (tf)\n",
      "Sample             :\n",
      "               Docs.\n",
      "Terms     1 _ one.txt  2 _ two.txt 3 _ three.txt 4 _ four.txt 5 _ five.txt\n",
      "develop             30           29           46           24          169\n",
      "fig                  2           26            0           10          300\n",
      "al                   7          139            4           15          203\n",
      "system              15          131           22           99          118\n",
      "control             17           43            7           60          270\n",
      "secur               25          335            4           34            5\n",
      "use                 27          145           25           75          172\n",
      "et                   7          231            0           17          210\n",
      "uav                 46          595           16          243          100\n",
      "drone              178           32          386           40          498\n",
      "\n",
      "> plotting DTM dendrogram ...\n",
      "\n",
      "DTM dendrogram graph saved as 'py-dtm_dendrogram.png'\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x700 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# an approximate \"python_03\" translation of the \"r_studio\" script made previously for research paper text analysis ;\n",
    "# this research paper text analysis work was originally done as a part of a project for an abstract, titled \"UNREGULATED DRONES AND THEIR IMPACT\", submitted to a research conference, named \"STUDENTS ON THEIR WAY TO SCIENCE\", in april of 2024 CE ;\n",
    "# the final version of the abstract can be found online in the conference proceedings publicaly available on the official website (direct link = \"https://www.sws.lbtu.lv/sites/sws/files/pages/attachments/SWS%20PROCEEDINGS_2024_FINAL.pdf\") ;\n",
    "# this practice code for \"python_03\" was originally prepared by \"swarn s. w. (a. y.)\" with the \"cc-by-nc-sa\" license ;\n",
    "import os\n",
    "import nltk\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.tokenize import word_tokenize\n",
    "from nltk.stem import SnowballStemmer\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from scipy.cluster.hierarchy import dendrogram, linkage\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.sparse import csr_matrix\n",
    "\n",
    "# the required NLTK data sets ('stopwords' & 'punkt') are already available in kaggle at a certain location ;\n",
    "nltk.data.path.append(\"/usr/share/nltk_data\")\n",
    "\n",
    "# checking the current working directory address ;\n",
    "print(\"checking the current working directory address ...\")\n",
    "print(os.getcwd())\n",
    "\n",
    "# listing all files in the current working directory ;\n",
    "print(\"listing all files in the current working directory ...\")\n",
    "print(os.listdir())\n",
    "\n",
    "# moving on to the practice code ;\n",
    "print(\"moving on to the practice code ...\")\n",
    "\n",
    "# defining a function to obtain the text corpus ;\n",
    "def read_corpus(directory):\n",
    "    \"\"\"Read all text files from a directory\"\"\"\n",
    "    texts = []\n",
    "    filenames = []\n",
    "    for filename in sorted(os.listdir(directory)):\n",
    "        if filename.endswith('.txt'):\n",
    "            with open(os.path.join(directory, filename), 'r', encoding='utf-8') as file:\n",
    "                texts.append(file.read())\n",
    "                filenames.append(filename)\n",
    "    return texts, filenames\n",
    "\n",
    "# reading the text corpus from the kaggle input directory ;\n",
    "corpus, filenames = read_corpus('/kaggle/input/my-corpus/my_corpus')\n",
    "\n",
    "# initializing stemmer ;\n",
    "stemmer = SnowballStemmer('english')\n",
    "\n",
    "# defining custom stop words ;\n",
    "custom_stops = [\"View\", \"PDFView\", \"articleView\", \"ScopusGoogle\", \"Scholar\",\n",
    "               \"the\", \"The\", \"articlea0\", \"2019\", \"2020\", \"2021\", \"can\", \n",
    "               \"use\", \"question\"]\n",
    "\n",
    "# defining preprocessing function ;\n",
    "def preprocess_text(text):\n",
    "    # tokenizing text ;\n",
    "    tokens = word_tokenize(text)\n",
    "    \n",
    "    # removing punctuation & converting to lower case ;\n",
    "    tokens = [word.lower() for word in tokens if word.isalnum()]\n",
    "    \n",
    "    # removing standard & custom stop words ;\n",
    "    stop_words = set(stopwords.words('english') + custom_stops)\n",
    "    tokens = [word for word in tokens if word not in stop_words]\n",
    "    \n",
    "    # stemming words ;\n",
    "    tokens = [stemmer.stem(word) for word in tokens]\n",
    "    \n",
    "    return ' '.join(tokens)\n",
    "\n",
    "# preprocessing all documents ;\n",
    "processed_corpus = [preprocess_text(doc) for doc in corpus]\n",
    "\n",
    "# creating a term-document matrix ;\n",
    "vectorizer = CountVectorizer()\n",
    "tdm = vectorizer.fit_transform(processed_corpus)\n",
    "\n",
    "# defining a function to display matrix info similar to r_studio's output ;\n",
    "def display_matrix_info(matrix, terms, docs, matrix_type=\"TermDocumentMatrix\"):\n",
    "    n_terms, n_docs = matrix.shape\n",
    "    non_zero = matrix.sum()\n",
    "    sparsity = (1.0 - (non_zero / (n_terms * n_docs))) * 100\n",
    "    max_term_length = max(len(term) for term in terms)\n",
    "    \n",
    "    print(f\"<<{matrix_type} (terms: {n_terms}, documents: {n_docs})>>\")\n",
    "    print(f\"Non-/sparse entries: {int(non_zero)}/{(n_terms * n_docs) - int(non_zero)}\")\n",
    "    print(f\"Sparsity           : {sparsity:.0f}%\")\n",
    "    print(f\"Maximal term length: {max_term_length}\")\n",
    "    print(f\"Weighting          : term frequency (tf)\")\n",
    "    print(\"Sample             :\")\n",
    "    \n",
    "    # displaying sample terms & their frequencies ;\n",
    "    print(\"               Docs.\")\n",
    "    print(\"Terms    \" + \" \".join(f\"{doc:>12}\" for doc in filenames))\n",
    "    \n",
    "    # selecting some interesting terms to show (the ones with highest total frequency) ;\n",
    "    if matrix_type == \"TermDocumentMatrix\":\n",
    "        # selecting terms that appear in multiple documents and have significant frequencies ;\n",
    "        term_sums = matrix.sum(axis=1).A1\n",
    "        # counting docs where term appears ;\n",
    "        term_docs = (matrix > 0).sum(axis=1).A1  \n",
    "        # filtering terms that appear in at least two documents and have meaningful total frequency ;\n",
    "        meaningful_terms = [(i, s) for i, (s, d) in enumerate(zip(term_sums, term_docs)) \n",
    "                           # adjusting these thresholds as needed ;\n",
    "                           if d >= 2 and s >= 10]  \n",
    "        # sorting by total frequency and take top ten ;\n",
    "        top_terms_idx = sorted(meaningful_terms, key=lambda x: x[1], reverse=True)[:10]\n",
    "        top_terms_idx = [idx for idx, _ in top_terms_idx]\n",
    "    else:\n",
    "        # keeping the original logic for the DTM info summary ;\n",
    "        term_sums = matrix.sum(axis=1).A1\n",
    "        top_terms_idx = np.argsort(term_sums)[-10:]\n",
    "    \n",
    "    # displaying the selected terms ;\n",
    "    if matrix_type == \"TermDocumentMatrix\":\n",
    "        print(\"\\n... (skipping this matrix due to it being too big) ...\\n\")\n",
    "    else:\n",
    "        for idx in top_terms_idx:\n",
    "            term = terms[idx]\n",
    "            freqs = matrix[idx].toarray()[0]\n",
    "            print(f\"{term:<9} \" + \" \".join(f\"{freq:>12}\" for freq in freqs))\n",
    "\n",
    "# converting to a data frame & displaying the TDM info summary ;\n",
    "terms = vectorizer.get_feature_names_out()\n",
    "df = pd.DataFrame(tdm.toarray(), columns=terms)\n",
    "print(\"\\n> TDM info summary ...\\n\")\n",
    "display_matrix_info(tdm, terms, filenames)\n",
    "\n",
    "# scaling the data set ;\n",
    "scaler = StandardScaler()\n",
    "df_scaled = scaler.fit_transform(df)\n",
    "\n",
    "# calculating distance matrix & perform hierarchical clustering ;\n",
    "d = linkage(df_scaled, method='ward')\n",
    "\n",
    "# plotting a dendrogram for the TDM ;\n",
    "print(\"\\n> plotting TDM dendrogram ...\\n\")\n",
    "plt.figure(figsize=(10, 7))\n",
    "dendrogram(d)\n",
    "plt.title('Hierarchical Clustering Dendrogram (Term-Document Matrix)')\n",
    "plt.xlabel('Sample Index')\n",
    "plt.ylabel('Distance')\n",
    "\n",
    "# saving the dendrogram plot in a file for the TDM ;\n",
    "plt.savefig('py-tdm_dendrogram.png')  \n",
    "print(\"TDM dendrogram graph saved as 'py-tdm_dendrogram.png'\")\n",
    "\n",
    "plt.show()\n",
    "\n",
    "# finding frequent terms & displaying them ;\n",
    "def find_frequent_terms(df, min_freq):\n",
    "    term_freq = df.sum()\n",
    "    frequent_terms = term_freq[term_freq >= min_freq].index.tolist()\n",
    "    # sorting alphabetically like r_studio's output\n",
    "    return sorted(frequent_terms)  \n",
    "\n",
    "print(\"\\nFrequent terms (appearing 35 or more times):\")\n",
    "freq_terms = find_frequent_terms(df, 35)\n",
    "print(\"'\" + \"'\".join(freq_terms) + \"'\")\n",
    "\n",
    "# creating the document-term matrix info summary & displaying it ;\n",
    "dtm = csr_matrix(tdm.T)\n",
    "print(\"\\n> DTM info summary ...\\n\")\n",
    "display_matrix_info(dtm, terms, filenames, \"DocumentTermMatrix\")\n",
    "\n",
    "# scaling the DTM ;\n",
    "dtm_scaled = scaler.fit_transform(dtm.toarray())\n",
    "\n",
    "# calculating distance matrix & perform hierarchical clustering for the DTM ;\n",
    "d1 = linkage(dtm_scaled, method='ward')\n",
    "\n",
    "# plotting a dendrogram for the DTM ;\n",
    "print(\"\\n> plotting DTM dendrogram ...\\n\")\n",
    "plt.figure(figsize=(10, 7))\n",
    "dendrogram(d1)\n",
    "plt.title('Hierarchical Clustering Dendrogram (Document-Term Matrix)')\n",
    "plt.xlabel('Sample Index')\n",
    "plt.ylabel('Distance')\n",
    "\n",
    "# saving the dendrogram plot in a file for the DTM ;\n",
    "plt.savefig('py-dtm_dendrogram.png')  \n",
    "print(\"DTM dendrogram graph saved as 'py-dtm_dendrogram.png'\")\n",
    "\n",
    "plt.show()\n",
    "\n",
    "# the end ;"
   ]
  }
 ],
 "metadata": {
  "kaggle": {
   "accelerator": "none",
   "dataSources": [
    {
     "datasetId": 6345257,
     "sourceId": 10257289,
     "sourceType": "datasetVersion"
    }
   ],
   "dockerImageVersionId": 30822,
   "isGpuEnabled": false,
   "isInternetEnabled": false,
   "language": "python",
   "sourceType": "notebook"
  },
  "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": "3.10.12"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 176.911153,
   "end_time": "2024-12-31T21:41:03.051694",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2024-12-31T21:38:06.140541",
   "version": "2.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}