cltk-latin-pipeline-output.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Building a Text Analysis Pipeline for Ancient Languages with CLTK"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This workshop will introduce the Classical Language Toolkit, an open-source Python framework dedicated to text analysis and natural language processing for historical languages. Participants will be taken through the basic stages of a text analysis pipeline, namely corpus loading, preprocessing, sentence and word tokenization, lemmatization, part-of-speech and morphological tagging, prosody identification, and more. Examples will given primarily using Latin texts (and English translations), though some attention will be given to other languages supported by the project, including Ancient Greek and Akkadian. Participants are encouraged to bring their laptops; the demonstration can be followed along with interactively in a web browser without any installation or setup."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pprint import pprint #RUN CELLS BY PRESSING SHIFT & RETURN TOGETHER"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Working with CLTK Corpora"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Installing Corpora"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Set up corpora\n",
    "\n",
    "## You will need the models/datasets that the new lemmatizer uses\n",
    "## Note that this will generate an error if this and older \n",
    "## version of this corpus is already installed. If that happens,\n",
    "## backup the old version, move or delete it, and reimport the corpus.\n",
    "\n",
    "#from cltk.corpus.utils.importer import CorpusImporter\n",
    "\n",
    "#corpus_importer = CorpusImporter('latin')\n",
    "#corpus_importer.list_corpora\n",
    "\n",
    "#corpus_importer.import_corpus('latin_models_cltk')\n",
    "\n",
    "## We will be using the Latin Library corpus for today's workshop,\n",
    "## so we will also need to import that as well.\n",
    "\n",
    "#corpus_importer.import_corpus('latin_text_latin_library')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading a CLTK corpus"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['12tables.txt', '1644.txt', 'abbofloracensis.txt', 'abelard/dialogus.txt', 'abelard/epistola.txt', 'abelard/historia.txt', 'addison/barometri.txt', 'addison/burnett.txt', 'addison/hannes.txt', 'addison/machinae.txt', 'addison/pax.txt', 'addison/praelium.txt', 'addison/preface.txt', 'addison/resurr.txt', 'addison/sphaer.txt', 'adso.txt', 'aelredus.txt', 'agnes.txt', 'alanus/alanus1.txt', 'alanus/alanus2.txt', 'albertanus/albertanus.arsloquendi.txt', 'albertanus/albertanus.liberconsol.txt', 'albertanus/albertanus.sermo.txt', 'albertanus/albertanus.sermo1.txt', 'albertanus/albertanus.sermo2.txt', 'albertanus/albertanus.sermo3.txt', 'albertanus/albertanus.sermo4.txt', 'albertanus/albertanus1.txt', 'albertanus/albertanus2.txt', 'albertanus/albertanus3.txt', 'albertanus/albertanus4.txt', 'albertofaix/hist1.txt', 'albertofaix/hist10.txt', 'albertofaix/hist11.txt', 'albertofaix/hist12.txt', 'albertofaix/hist2.txt', 'albertofaix/hist3.txt', 'albertofaix/hist4.txt', 'albertofaix/hist5.txt', 'albertofaix/hist6.txt', 'albertofaix/hist7.txt', 'albertofaix/hist8.txt', 'albertofaix/hist9.txt', 'alcuin/cella.txt', 'alcuin/conflictus.txt', 'alcuin/epitaphium.txt', 'alcuin/luscinia.txt', 'alcuin/propos.txt', 'alcuin/rec.txt', 'alcuin/rhetorica.txt']\n"
     ]
    }
   ],
   "source": [
    "# Get the Latin Library corpus\n",
    "\n",
    "from cltk.corpus.readers import get_corpus_reader\n",
    "ll = get_corpus_reader(language='latin', \n",
    "                       corpus_name='latin_text_latin_library')\n",
    "\n",
    "# The CLTK Latin Library corpus is a web-scraped collection of plaintext files\n",
    "# from thelatinlibrary.com.\n",
    "\n",
    "# We can access the individual files as follows...\n",
    "\n",
    "files = ll.fileids()\n",
    "print(files[:50]) # The first 50 files in the corpus\n",
    "\n",
    "# Note the [:50] slice to limit our list to the first 50 items."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 2141 files in this corpus.\n"
     ]
    }
   ],
   "source": [
    "# Stats\n",
    "\n",
    "file_count = len(files)\n",
    "print(f'There are {file_count} files in this corpus.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['vergil/aen1.txt', 'vergil/aen10.txt', 'vergil/aen11.txt', 'vergil/aen12.txt', 'vergil/aen2.txt', 'vergil/aen3.txt', 'vergil/aen4.txt', 'vergil/aen5.txt', 'vergil/aen6.txt', 'vergil/aen7.txt', 'vergil/aen8.txt', 'vergil/aen9.txt', 'vergil/ec1.txt', 'vergil/ec10.txt', 'vergil/ec2.txt', 'vergil/ec3.txt', 'vergil/ec4.txt', 'vergil/ec5.txt', 'vergil/ec6.txt', 'vergil/ec7.txt', 'vergil/ec8.txt', 'vergil/ec9.txt', 'vergil/geo1.txt', 'vergil/geo2.txt', 'vergil/geo3.txt', 'vergil/geo4.txt']\n"
     ]
    }
   ],
   "source": [
    "virgil_files = [file for file in files if \"vergil\" in file]\n",
    "print(virgil_files)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Arma virumque cano, Troiae qui primus ab oris \n",
      "Italiam, fato profugus, Laviniaque venit \n",
      "litora, multum ille et terris iactatus et alto \n",
      "vi superum saevae memorem Iunonis ob iram; \n",
      "multa quoque et bello passus, dum conderet urbem,    5 \n",
      "inferretque deos Latio, genus unde Latinum, \n",
      "Albanique patres, atque altae moenia Romae.\n",
      " \n",
      "\n",
      " \n",
      "Musa, mihi causas memora, quo numine laeso, \n",
      "quidve dolens, regina deum tot volvere casus \n",
      "insignem pietate virum, tot adire labores    10 \n",
      "impulerit. Tantaene animis caelestibus irae?\n"
     ]
    }
   ],
   "source": [
    "print(ll.raw(virgil_files[0])[101:616])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Vergil: Aeneid I\n",
      "\t\t \n",
      "\n",
      "\t\t \n",
      "\t\t\n",
      "\t\t \n",
      "\t\t\n",
      "\t\t \n",
      "\t\t \n",
      "\t \n",
      "\t\n",
      " \n",
      "\n",
      "\n",
      " P. VERGILI MARONIS AENEIDOS LIBER PRIMVS \n",
      "\n",
      "\n",
      " \n",
      "\n",
      "Arma virumque cano, Troiae qui primus ab oris \n",
      "Italiam, fato profugus, Laviniaque venit \n",
      "litora, mu\n"
     ]
    }
   ],
   "source": [
    "print(ll.raw(virgil_files[0])[:200])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dias,' inquit, 'Danaum, casusque tuorum, \n",
      "erroresque tuos; nam te iam septima portat    755 \n",
      "omnibus errantem terris et fluctibus aestas.'\n",
      " \n",
      "\n",
      " \n",
      "\t \n",
      "\t\t Vergil \n",
      "\t\t The Latin Library \n",
      "\t\t The Classics Page\n"
     ]
    }
   ],
   "source": [
    "print(ll.raw(virgil_files[0])[-200:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6350\n"
     ]
    }
   ],
   "source": [
    "word_count = len(list(ll.words(virgil_files[0])))\n",
    "print(word_count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "244\n"
     ]
    }
   ],
   "source": [
    "sent_count = len(list(ll.sents(virgil_files[0])))\n",
    "print(sent_count)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Imports for preprocessing\n",
    "\n",
    "import re # Regex module, useful for pattern matching\n",
    "import html # Useful for handling entities\n",
    "\n",
    "# Import/load a CLTK tool for normalizing i/j and u/v in Latin texts\n",
    "from cltk.stem.latin.j_v import JVReplacer\n",
    "replacer = JVReplacer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def preprocess(text):\n",
    "\n",
    "    # Remove Latin Library-specific paratexts with regex\n",
    "    \n",
    "    remove_list = [\n",
    "            r'\\bP. VERGILI MARONIS AENEIDOS LIBER .+\\b',\n",
    "            r'Vergil: Aeneid .+',\n",
    "            r'\\bThe Latin Library\\b',\n",
    "            r'\\bThe Classics Page\\b',\n",
    "            r'\\bVergil\\b',\n",
    "        ]\n",
    "    \n",
    "    for pattern in remove_list:\n",
    "        text = re.sub(pattern, '', text)\n",
    "\n",
    "    # Remove html entities and related html artifacts\n",
    "    \n",
    "    text = html.unescape(text) # Handle html entities\n",
    "    text = re.sub(r'&nbsp;?', ' ',text) #&nbsp; stripped incorrectly in corpus?\n",
    "    text = re.sub(r'\\x00',' ',text) #Another space problem?\n",
    "    text = re.sub(r' \\xa0 ', '    ', text)\n",
    "    \n",
    "    # Lowercase text\n",
    "    text = text.lower()\n",
    "\n",
    "    # Normalize text\n",
    "    text = replacer.replace(text) #Normalize u/v & i/j\n",
    "    \n",
    "    # Remove punctuation with translate\n",
    "    punctuation =\"\\\"#$%&\\'()+,-/:;<=>@[\\]^_`{|}~.?!«»—\"\n",
    "    translator = str.maketrans({key: \" \" for key in punctuation})\n",
    "    text = text.translate(translator)\n",
    "    \n",
    "    # Remove numbers\n",
    "    translator = str.maketrans({key: \" \" for key in '0123456789'})\n",
    "    text = text.translate(translator)\n",
    "    \n",
    "    # Handle spacing\n",
    "    text = re.sub('[ ]+',' ', text) # Remove double spaces\n",
    "    text = re.sub('\\t',' ', text) # Remove tabs\n",
    "    text = re.sub('^\\s+','', text)\n",
    "    text = re.sub(' \\n', '\\n', text)\n",
    "    text = re.sub('\\n\\n', '~', text)\n",
    "    text = re.sub('~+', '\\n\\n', text)\n",
    "    \n",
    "    return text.strip()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Vergil: Aeneid I\n",
      "\t\t \n",
      "\n",
      "\t\t \n",
      "\t\t\n",
      "\t\t \n",
      "\t\t\n",
      "\t\t \n",
      "\t\t \n",
      "\t \n",
      "\t\n",
      " \n",
      "\n",
      "\n",
      " P. VERGILI MARONIS AENEIDOS LIBER PRIMVS \n",
      "\n",
      "\n",
      " \n",
      "\n",
      "Arma virumque cano, Troiae qui primus ab oris \n",
      "Italiam, fato profugus, Laviniaque venit \n",
      "litora, multum ille et terris iactatus et alto \n",
      "vi superum saevae memorem Iunonis ob iram; \n",
      "multa quoque et bello passus, dum conderet urbem,    5 \n",
      "inferretque deos Latio, genus unde Latinum, \n",
      "Albanique patres, atque altae moenia Romae.\n",
      " \n",
      "\n",
      " \n",
      "Musa, mihi causas memora, quo numine laeso, \n",
      "quidve dolens, regina d\n"
     ]
    }
   ],
   "source": [
    "virgil_raw = ll.raw(virgil_files[0])\n",
    "print(virgil_raw[:500])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "arma uirumque cano troiae qui primus ab oris\n",
      "italiam fato profugus lauiniaque uenit\n",
      "litora multum ille et terris iactatus et alto\n",
      "ui superum saeuae memorem iunonis ob iram\n",
      "multa quoque et bello passus dum conderet urbem\n",
      "inferretque deos latio genus unde latinum\n",
      "albanique patres atque altae moenia romae\n",
      "\n",
      "musa mihi causas memora quo numine laeso\n",
      "quidue dolens regina deum tot uoluere casus\n",
      "insignem pietate uirum tot adire labores\n",
      "impulerit tantaene animis caelestibus irae\n",
      "\n",
      "urbs antiqua fuit tyrii t\n"
     ]
    }
   ],
   "source": [
    "virgil_pp = preprocess(ll.raw(virgil_files[0]))\n",
    "print(virgil_pp[:500])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tokenization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set up CLTK Latin word tokenizer\n",
    "\n",
    "from cltk.tokenize.word import WordTokenizer\n",
    "word_tokenizer = WordTokenizer('latin')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['arma', 'uirum', '-que', 'cano', 'troiae', 'qui', 'primus', 'ab', 'oris', 'italiam', 'fato', 'profugus', 'lauinia', '-que', 'uenit', 'litora', 'multum', 'ille', 'et', 'terris', 'iactatus', 'et', 'alto', 'ui', 'superum', 'saeuae', 'memorem', 'iunonis', 'ob', 'iram', 'multa', 'quoque', 'et', 'bello', 'passus', 'dum', 'conderet', 'urbem', 'inferret', '-que', 'deos', 'latio', 'genus', 'unde', 'latinum', 'albani', '-que', 'patres', 'atque', 'altae', 'moenia', 'romae', 'musa', 'mihi', 'causas', 'memora', 'quo', 'numine', 'laeso', 'quid', '-ue', 'dolens', 'regina', 'deum', 'tot', 'uoluere', 'casus', 'insignem', 'pietate', 'uirum', 'tot', 'adire', 'labores', 'impulerit', 'tantae', '-ne', 'animis', 'caelestibus', 'irae', 'urbs', 'antiqua', 'fuit', 'tyrii', 'tenuere', 'coloni', 'karthago', 'italiam', 'contra', 'tiberina', '-que', 'longe', 'ostia', 'diues', 'opum', 'studiis', '-que', 'asperrima', 'belli', 'quam', 'iuno', 'fertur', 'terris', 'magis', 'omnibus', 'unam', 'posthabita', 'coluisse', 'samo', 'hic', 'illius', 'arma', 'hic', 'currus', 'fuit', 'hoc', 'regnum', 'dea', 'gentibus', 'esse', 'si', 'qua', 'fata', 'sinant', 'iam', 'tum']\n",
      "\n",
      "\n",
      "There are 5167 tokens in Virgil 1.\n"
     ]
    }
   ],
   "source": [
    "# Here is what the same poem looks like as a list of tokens\n",
    "\n",
    "tokens = word_tokenizer.tokenize(virgil_pp)\n",
    "\n",
    "print(tokens[:125])\n",
    "print('\\n')\n",
    "print(f'There are {len(tokens)} tokens in Virgil 1.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Lemmatization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # Set up CLTK Latin backoff lemmatizer\n",
    "\n",
    "from cltk.lemmatize.latin.backoff import BackoffLatinLemmatizer\n",
    "lemmatizer = BackoffLatinLemmatizer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "lemmas = lemmatizer.lemmatize(tokens)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('arma', 'arma'), ('uirum', 'uir'), ('-que', '-que'), ('cano', 'cano'), ('troiae', 'troia'), ('qui', 'qui'), ('primus', 'primus'), ('ab', 'ab'), ('oris', 'ora'), ('italiam', 'italia'), ('fato', 'fatum'), ('profugus', 'profugus'), ('lauinia', 'lauinia'), ('-que', '-que'), ('uenit', 'uenio'), ('litora', 'litus'), ('multum', 'multus'), ('ille', 'ille'), ('et', 'et'), ('terris', 'terra'), ('iactatus', 'jacto'), ('et', 'et'), ('alto', 'altus'), ('ui', 'uis'), ('superum', 'superi'), ('saeuae', 'saevus'), ('memorem', 'memoro'), ('iunonis', 'iuno'), ('ob', 'ob'), ('iram', 'ira'), ('multa', 'multus'), ('quoque', 'quoque'), ('et', 'et'), ('bello', 'bellum'), ('passus', 'patior'), ('dum', 'dum'), ('conderet', 'condo'), ('urbem', 'urbs'), ('inferret', 'infero'), ('-que', '-que'), ('deos', 'deus'), ('latio', 'latium'), ('genus', 'genus'), ('unde', 'unde'), ('latinum', 'latinum'), ('albani', 'albani'), ('-que', '-que'), ('patres', 'pater'), ('atque', 'atque'), ('altae', 'alo'), ('moenia', 'moenia'), ('romae', 'roma'), ('musa', 'musa'), ('mihi', 'ego'), ('causas', 'causa'), ('memora', 'memoro'), ('quo', 'quo'), ('numine', 'numen'), ('laeso', 'laedo'), ('quid', 'quis'), ('-ue', '-ue'), ('dolens', 'doleo'), ('regina', 'regina'), ('deum', 'deus'), ('tot', 'tot'), ('uoluere', 'volvo'), ('casus', 'casus'), ('insignem', 'insignis'), ('pietate', 'pietas'), ('uirum', 'uir'), ('tot', 'tot'), ('adire', 'adeo'), ('labores', 'labor'), ('impulerit', 'impello'), ('tantae', 'tantus'), ('-ne', '-ne'), ('animis', 'animus'), ('caelestibus', 'caelestis'), ('irae', 'ira'), ('urbs', 'urbs'), ('antiqua', 'antiquus'), ('fuit', 'sum'), ('tyrii', 'tyrii'), ('tenuere', 'teneo'), ('coloni', 'colonus'), ('karthago', 'karthago'), ('italiam', 'italia'), ('contra', 'contra'), ('tiberina', 'tiberinus'), ('-que', '-que'), ('longe', 'longus'), ('ostia', 'ostium'), ('diues', 'diues'), ('opum', 'opos'), ('studiis', 'studium'), ('-que', '-que'), ('asperrima', 'asper1'), ('belli', 'bellum'), ('quam', 'quam'), ('iuno', 'iuno')]\n"
     ]
    }
   ],
   "source": [
    "print(lemmas[:100])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('arma', 'arma'),\n",
      " ('uirum', 'uir'),\n",
      " ('-que', '-que'),\n",
      " ('cano', 'cano'),\n",
      " ('troiae', 'troia'),\n",
      " ('qui', 'qui'),\n",
      " ('primus', 'primus'),\n",
      " ('ab', 'ab'),\n",
      " ('oris', 'ora'),\n",
      " ('italiam', 'italia'),\n",
      " ('fato', 'fatum'),\n",
      " ('profugus', 'profugus'),\n",
      " ('lauinia', 'lauinia'),\n",
      " ('-que', '-que'),\n",
      " ('uenit', 'uenio'),\n",
      " ('litora', 'litus'),\n",
      " ('multum', 'multus'),\n",
      " ('ille', 'ille'),\n",
      " ('et', 'et'),\n",
      " ('terris', 'terra'),\n",
      " ('iactatus', 'jacto'),\n",
      " ('et', 'et'),\n",
      " ('alto', 'altus'),\n",
      " ('ui', 'uis'),\n",
      " ('superum', 'superi')]\n"
     ]
    }
   ],
   "source": [
    "from pprint import pprint\n",
    "\n",
    "pprint(lemmas[:25])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## POS & Morphological Tagging"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from cltk.tag.pos import POSTag\n",
    "\n",
    "tagger = POSTag('latin')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('arma', 'N-P---NA-'),\n",
      " ('uirum', None),\n",
      " ('-', 'U--------'),\n",
      " ('que', 'C--------'),\n",
      " ('cano', None),\n",
      " ('troiae', 'N-S---FG-'),\n",
      " ('qui', 'P-P---MN-'),\n",
      " ('primus', 'A-S---MN-'),\n",
      " ('ab', 'R--------'),\n",
      " ('oris', 'N-P---NB-')]\n"
     ]
    }
   ],
   "source": [
    "pos_tags = tagger.tag_ngram_123_backoff(' '.join(tokens))\n",
    "pprint(pos_tags[:10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# pos_tags_2 = tagger.tag_tnt(' '.join(tokens))\n",
    "# pprint(pos_tags_2[:10])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prosody Tagging"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "import Levenshtein\n",
    "\n",
    "from cltk.prosody.latin.hexameter_scanner import HexameterScanner\n",
    "scanner = HexameterScanner()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['ārm', 'u', 'i', 'rūm', 'que', 'ca', 'nō', 'tro', 'jae', 'quī', 'pri', 'mus', 'ab', 'o', 'ris']\n",
      "-    UU -   U  U -   -  -   -   - U  U  - U \n"
     ]
    }
   ],
   "source": [
    "scansion = scanner.scan(virgil_pp[:44])\n",
    "print(scansion.syllables)\n",
    "print(scansion.scansion)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from cltk.tokenize.line import LineTokenizer\n",
    "line_tokenizer = LineTokenizer('latin')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "lines = line_tokenizer.tokenize(virgil_pp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-UU-UU-----UU-U\n",
      "-UU---UU--UUUU-U\n",
      "-UU-------UU-U\n",
      "-UU-UU-UU-UU-UU-U\n",
      "-UU-------UU-U\n",
      "---UU-UUUUU-UU-U\n",
      "-UU-------UU--\n",
      "-UUU--UUUUUUU-U\n",
      "-UUU-UUUUUUUUUUU-U\n",
      "---UUUUU-UUUUU-U\n"
     ]
    }
   ],
   "source": [
    "for line in lines[:10]:\n",
    "    scansion = scanner.scan(line)\n",
    "    print(scansion.scansion.replace(' ',''))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Named Entity Recognition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('ut',),\n",
      " ('Venus',),\n",
      " (',',),\n",
      " ('ut',),\n",
      " ('Sirius', 'Entity'),\n",
      " (',',),\n",
      " ('ut',),\n",
      " ('Spica', 'Entity'),\n",
      " (',',),\n",
      " ('ut',),\n",
      " ('aliae',),\n",
      " ('quae',),\n",
      " ('primae',),\n",
      " ('dicuntur',),\n",
      " ('esse',),\n",
      " ('mangitudinis',),\n",
      " ('.',)]\n"
     ]
    }
   ],
   "source": [
    "from cltk.tag import ner\n",
    "\n",
    "text_str = \"\"\"ut Venus, ut Sirius, ut Spica, ut aliae quae primae dicuntur esse mangitudinis.\"\"\"\n",
    "\n",
    "pprint(ner.tag_ner('latin', input_text=text_str, output_type=list))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "def preprocess_2(text):\n",
    "\n",
    "    # Remove Latin Library-specific paratexts with regex\n",
    "    \n",
    "    remove_list = [\n",
    "            r'\\bP. VERGILI MARONIS AENEIDOS LIBER .+\\b',\n",
    "            r'Vergil: Aeneid .+',\n",
    "            r'\\bThe Latin Library\\b',\n",
    "            r'\\bThe Classics Page\\b',\n",
    "            r'\\bVergil\\b',\n",
    "        ]\n",
    "    \n",
    "    for pattern in remove_list:\n",
    "        text = re.sub(pattern, '', text)\n",
    "\n",
    "    # Remove html entities and related html artifacts\n",
    "    \n",
    "    text = html.unescape(text) # Handle html entities\n",
    "    text = re.sub(r'&nbsp;?', ' ',text) #&nbsp; stripped incorrectly in corpus?\n",
    "    text = re.sub(r'\\x00',' ',text) #Another space problem?\n",
    "    text = re.sub(r' \\xa0 ', '    ', text)\n",
    "    \n",
    "    # Lowercase text\n",
    "#     text = text.lower()\n",
    "\n",
    "    # Normalize text\n",
    "    text = replacer.replace(text) #Normalize u/v & i/j\n",
    "    \n",
    "    # Remove punctuation with translate\n",
    "    punctuation =\"\\\"#$%&\\'()+,-/:;<=>@[\\]^_`{|}~.?!«»—\"\n",
    "    translator = str.maketrans({key: \" \" for key in punctuation})\n",
    "    text = text.translate(translator)\n",
    "    \n",
    "    # Remove numbers\n",
    "    translator = str.maketrans({key: \" \" for key in '0123456789'})\n",
    "    text = text.translate(translator)\n",
    "    \n",
    "    # Handle spacing\n",
    "    text = re.sub('[ ]+',' ', text) # Remove double spaces\n",
    "    text = re.sub('\\t',' ', text) # Remove tabs\n",
    "    text = re.sub('^\\s+','', text)\n",
    "    text = re.sub(' \\n', '\\n', text)\n",
    "    text = re.sub('\\n\\n', '~', text)\n",
    "    text = re.sub('~+', '\\n\\n', text)\n",
    "    \n",
    "    return text.strip()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Arma uirumque cano Troiae qui primus ab oris\n",
      "Italiam fato profugus Lauiniaque uenit\n",
      "litora multum il\n"
     ]
    }
   ],
   "source": [
    "virgil_pp_2 = preprocess_2(ll.raw(virgil_files[0]))\n",
    "print(virgil_pp_2[:100])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('Arma',),\n",
      " ('uirumque',),\n",
      " ('cano',),\n",
      " ('Troiae', 'Entity'),\n",
      " ('qui',),\n",
      " ('primus',),\n",
      " ('ab',),\n",
      " ('oris',),\n",
      " ('Italiam', 'Entity'),\n",
      " ('fato',),\n",
      " ('profugus',),\n",
      " ('Lauiniaque', 'Entity'),\n",
      " ('uenit',),\n",
      " ('litora',),\n",
      " ('multum',),\n",
      " ('ille',),\n",
      " ('et',),\n",
      " ('terris',),\n",
      " ('iactatus',),\n",
      " ('et',),\n",
      " ('alto',),\n",
      " ('ui',),\n",
      " ('superum',),\n",
      " ('saeuae',),\n",
      " ('memorem',)]\n"
     ]
    }
   ],
   "source": [
    "pprint(ner.tag_ner('latin', input_text=virgil_pp_2, output_type=list)[:25])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Sample Text Analyses & Visualizations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Counting Tokens"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('-que', 252), ('et', 166), ('in', 49), ('per', 29), ('cum', 27), ('atque', 25), ('aut', 24), ('iam', 23), ('quae', 23), ('ad', 21), ('-ne', 19), ('hic', 19), ('nec', 19), ('aeneas', 19), ('tum', 18), ('se', 18), ('te', 17), ('non', 17), ('sic', 16), ('est', 15), ('qui', 14), ('o', 14), ('nunc', 14), ('ab', 13), ('si', 13)]\n"
     ]
    }
   ],
   "source": [
    "word_count = Counter(tokens)\n",
    "print(word_count.most_common(25))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top 25 words in Virgil 1:\n",
      "\n",
      "       TOKEN       COUNT       TOKEN %     RUNNING %   \n",
      "    1. -que        252         4.88%       4.88%       \n",
      "    2. et          166         3.21%       8.09%       \n",
      "    3. in          49          0.95%       9.04%       \n",
      "    4. per         29          0.56%       9.6%        \n",
      "    5. cum         27          0.52%       10.12%      \n",
      "    6. atque       25          0.48%       10.61%      \n",
      "    7. aut         24          0.46%       11.07%      \n",
      "    8. iam         23          0.45%       11.52%      \n",
      "    9. quae        23          0.45%       11.96%      \n",
      "   10. ad          21          0.41%       12.37%      \n",
      "   11. -ne         19          0.37%       12.73%      \n",
      "   12. hic         19          0.37%       13.1%       \n",
      "   13. nec         19          0.37%       13.47%      \n",
      "   14. aeneas      19          0.37%       13.84%      \n",
      "   15. tum         18          0.35%       14.19%      \n",
      "   16. se          18          0.35%       14.53%      \n",
      "   17. te          17          0.33%       14.86%      \n",
      "   18. non         17          0.33%       15.19%      \n",
      "   19. sic         16          0.31%       15.5%       \n",
      "   20. est         15          0.29%       15.79%      \n",
      "   21. qui         14          0.27%       16.06%      \n",
      "   22. o           14          0.27%       16.33%      \n",
      "   23. nunc        14          0.27%       16.61%      \n",
      "   24. ab          13          0.25%       16.86%      \n",
      "   25. si          13          0.25%       17.11%      \n"
     ]
    }
   ],
   "source": [
    "running = 0\n",
    "\n",
    "print('Top 25 words in Virgil 1:\\n')\n",
    "print(\"{number:>5}  {word:<12}{count:<12}{percent:<12}{running:<12}\". \\\n",
    "        format(number=\"\", word=\"TOKEN\", count=\"COUNT\", percent=\"TOKEN %\", running = \"RUNNING %\"))\n",
    "for i, pair in enumerate(word_count.most_common(25)):\n",
    "    running += pair[1]\n",
    "    print(\"{number:>5}. {word:<12}{count:<12}{percent:<12}{running:<12}\". \\\n",
    "        format(number=i+1, word=pair[0], count=pair[1], \\\n",
    "        percent=str(round(pair[1] / len(tokens)*100, 2))+\"%\", running = str(round(running / len(tokens)*100, 2))+\"%\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### KWIC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk import Text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Displaying 19 of 19 matches:\n",
      "non alligat ancora morsu huc septem aeneas collectis nauibus omni ex numero su\n",
      "ere parant flammis et frangere saxo aeneas scopulum interea conscendit et omne\n",
      "iam exaudire uocatos praecipue pius aeneas nunc acris oronti nunc amyci casum \n",
      "mperiis et fulmine terres quid meus aeneas in te committere tantum quid troes \n",
      "animum mentem -que benignam at pius aeneas per noctem plurima uoluens ut primu\n",
      "cis tempestas adpulit oris sum pius aeneas raptos qui ex hoste penates classe \n",
      "spectat desuper arces miratur molem aeneas magalia quondam miratur portas stre\n",
      "fortunati quorum iam moenia surgunt aeneas ait et fastigia suspicit urbis infe\n",
      "es oblata timorem leniit hic primum aeneas sperare salutem ausus et adflictis \n",
      "ustis aut sorte trahebat cum subito aeneas concursu accedere magno anthea serg\n",
      "emores fandi atque nefandi rex erat aeneas nobis quo iustior alter nec pietate\n",
      "x ipse noto compulsus eodem adforet aeneas equidem per litora certos dimittam \n",
      "i dictis et fortis achates et pater aeneas iamdudum erumpere nubem ardebant pr\n",
      " in aethera purgat apertum restitit aeneas clara -que in luce refulsit os umer\n",
      "t coram quem quaeritis adsum troius aeneas libycis ereptus ab undis o sola inf\n",
      "immanibus applicat oris tu -ne ille aeneas quem dardanio anchisae alma uenus p\n",
      "cta uiros antiqua ab origine gentis aeneas neque enim patrius consistere mente\n",
      " supplex tua numina posco frater ut aeneas pelago tuus omnia circum litora iac\n",
      "ponda mediam -que locauit iam pater aeneas et iam troiana iuuentus conueniunt \n"
     ]
    }
   ],
   "source": [
    "virgil_Text = Text(tokens)\n",
    "virgil_Text.concordance('Aeneas')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dispersion Plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline  \n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Text: arma uirum -que cano troiae qui primus ab...>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "virgil_Text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 5))\n",
    "virgil_Text.dispersion_plot(['aeneas', 'uenus', 'dido'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['vergil/aen1.txt', 'vergil/aen10.txt', 'vergil/aen11.txt', 'vergil/aen12.txt', 'vergil/aen2.txt', 'vergil/aen3.txt', 'vergil/aen4.txt', 'vergil/aen5.txt', 'vergil/aen6.txt', 'vergil/aen7.txt', 'vergil/aen8.txt', 'vergil/aen9.txt']\n"
     ]
    }
   ],
   "source": [
    "# Show books sorting problem\n",
    "\n",
    "aeneid_files = [file for file in files if \"vergil/aen\" in file]\n",
    "print(aeneid_files)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['vergil/aen1.txt', 'vergil/aen2.txt', 'vergil/aen3.txt', 'vergil/aen4.txt', 'vergil/aen5.txt', 'vergil/aen6.txt', 'vergil/aen7.txt', 'vergil/aen8.txt', 'vergil/aen9.txt', 'vergil/aen10.txt', 'vergil/aen11.txt', 'vergil/aen12.txt']\n"
     ]
    }
   ],
   "source": [
    "# Fix books sorting problem\n",
    "\n",
    "aeneid_order = [int(\" \".join(re.findall(r'\\d+', item))) for item in aeneid_files]\n",
    "aeneid_files = [x for _, x in sorted(zip(aeneid_order, aeneid_files))]\n",
    "print(aeneid_files)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "aeneid_pp = preprocess(ll.raw(aeneid_files))\n",
    "aeneid_tokens = word_tokenizer.tokenize(aeneid_pp)\n",
    "aeneid_lemmas = [x for _, x in lemmatizer.lemmatize(aeneid_tokens)]\n",
    "\n",
    "# Error in lemmatizer; need to fix\n",
    "aeneid_lemmas = [lemma if lemma != 'dis-do' else 'dido' for lemma in aeneid_lemmas]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Dispersion plot of entire Aeneid\n",
    "\n",
    "aeneid_Text = Text(aeneid_tokens)\n",
    "plt.figure(figsize=(20, 5))\n",
    "aeneid_Text.dispersion_plot(['aeneas', 'uenus', 'dido'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Lemmatized dispersion plot\n",
    "\n",
    "aeneid_Text = Text(aeneid_lemmas)\n",
    "plt.figure(figsize=(20, 5))\n",
    "aeneid_Text.dispersion_plot(['aeneas', 'uenus', 'dido'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Graphed Frequency Distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11f5d42d0>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from nltk.probability import FreqDist\n",
    "\n",
    "fdist = FreqDist(virgil_Text)\n",
    "\n",
    "plt.figure(figsize=(20, 10))\n",
    "fdist.plot(50, cumulative=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Working with other CLTK Languages"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Working with Ancient Greek in CLTK"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "import requests\n",
    "\n",
    "response = requests.get('https://raw.githubusercontent.com/tesserae/tesserae/master/texts/grc/homer.iliad/homer.iliad.part.1.tess')\n",
    "iliad = response.text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<hom. il. 1.1>\tμῆνιν ἄειδε θεὰ Πηληϊάδεω ̓Αχιλῆος\n",
      "<hom. il. 1.2>\tοὐλομένην, ἣ μυρί' ̓Αχα\n"
     ]
    }
   ],
   "source": [
    "print(iliad[:100])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "μῆνιν ἄειδε θεὰ Πηληϊάδεω ̓Αχιλῆος\n",
      "οὐλομένην, ἣ μυρί' ̓Αχαιοῖς ἄλγε' ἔθηκε,\n",
      "πολλὰς\n"
     ]
    }
   ],
   "source": [
    "iliad = re.sub(r'<.+?>\\t', '', iliad)\n",
    "print(iliad[:100])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "import unicodedata\n",
    "\n",
    "iliad = unicodedata.normalize('NFC', iliad)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['μῆνιν', 'ἄειδε', 'θεὰ', 'Πηληϊάδεω', '̓Αχιλῆος', 'οὐλομένην', ',', 'ἣ', 'μυρί', \"'\", '̓Αχαιοῖς', 'ἄλγε', \"'\", 'ἔθηκε', ',', 'πολλὰς', 'δ', \"'\", 'ἰφθίμους', 'ψυχὰς', '̓́Αϊδι', 'προί̈αψεν', 'ἡρώων', ',', 'αὐτοὺς', 'δὲ', 'ἑλώρια', 'τεῦχε', 'κύνεσσιν', 'οἰωνοῖσί', 'τε', 'πᾶσι', ',', 'Διὸς', 'δ', \"'\", 'ἐτελείετο', 'βουλή', ',', 'ἐξ', 'οὗ', 'δὴ', 'τὰ', 'πρῶτα', 'διαστήτην', 'ἐρίσαντε', '̓Ατρεί̈δης', 'τε', 'ἄναξ', 'ἀνδρῶν', 'καὶ', 'δῖος', '̓Αχιλλεύς', '.', 'τίς', 'τ', \"'\", 'ἄρ', 'σφωε', 'θεῶν', 'ἔριδι', 'ξυνέηκε', 'μάχεσθαι', ';', 'Λητοῦς', 'καὶ', 'Διὸς', 'υἱός', ':', 'ὃ', 'γὰρ', 'βασιλῆϊ', 'χολωθεὶς', 'νοῦσον', 'ἀνὰ', 'στρατὸν', 'ὄρσε', 'κακήν', ',', 'ὀλέκοντο', 'δὲ', 'λαοί', ',', 'οὕνεκα', 'τὸν', 'Χρύσην', 'ἠτίμασεν', 'ἀρητῆρα', '̓Ατρεί̈δης', ':', 'ὃ', 'γὰρ', 'ἦλθε', 'θοὰς', 'ἐπὶ', 'νῆας', '̓Αχαιῶν', 'λυσόμενός', 'τε', 'θύγατρα']\n"
     ]
    }
   ],
   "source": [
    "from cltk.tokenize.word import WordTokenizer\n",
    "\n",
    "word_tokenizer_greek = WordTokenizer('greek')\n",
    "tokens = word_tokenizer_greek.tokenize(iliad)\n",
    "print(tokens[:100])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "from cltk.tokenize.line import LineTokenizer\n",
    "\n",
    "line_tokenizer = LineTokenizer('greek')\n",
    "lines = line_tokenizer.tokenize(iliad)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('μῆνιν', 'N-S---FA-'),\n",
       " ('ἄειδε', 'V2SPMA---'),\n",
       " ('θεὰ', 'N-S---FV-'),\n",
       " ('Πηληϊάδεω', None),\n",
       " ('̓', None),\n",
       " ('Αχιλῆος', None)]"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from cltk.tag.pos import POSTag\n",
    "tagger = POSTag('greek')\n",
    "\n",
    "tagger.tag_ngram_123_backoff(lines[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Working with Akkadian in CLTK"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "# In[1]: import os\n",
    "\n",
    "from cltk.tokenize.word import WordTokenizer\n",
    "\n",
    "word_tokenizer_akkadian = WordTokenizer('akkadian')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('u2-wa-a-ru', 'akkadian'),\n",
      " ('at-ta', 'akkadian'),\n",
      " ('e2-kal2-la-ka', 'akkadian'),\n",
      " ('_e2_-ka', 'sumerian'),\n",
      " ('wu-e-er', 'akkadian')]\n"
     ]
    }
   ],
   "source": [
    "line = 'u2-wa-a-ru at-ta e2-kal2-la-ka _e2_-ka wu-e-er'\n",
    "tokens = word_tokenizer_akkadian.tokenize(line)\n",
    "pprint(tokens)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['e', 'piš', 'ta', 'šu']"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from cltk.stem.akkadian.syllabifier import Syllabifier\n",
    "\n",
    "word = \"epištašu\"\n",
    "syll = Syllabifier()\n",
    "syll.syllabify(word)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('ilum', {'case': 'nominative', 'number': 'singular'}),\n",
       " ('ilam', {'case': 'accusative', 'number': 'singular'}),\n",
       " ('ilim', {'case': 'genitive', 'number': 'singular'}),\n",
       " ('ilān', {'case': 'nominative', 'number': 'dual'}),\n",
       " ('ilīn', {'case': 'oblique', 'number': 'dual'}),\n",
       " ('ilū', {'case': 'nominative', 'number': 'plural'}),\n",
       " ('ilī', {'case': 'oblique', 'number': 'plural'})]"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from cltk.stem.akkadian.declension import NaiveDecliner\n",
    "\n",
    "word = 'ilum'\n",
    "decliner = NaiveDecliner()\n",
    "decliner.decline_noun(word, 'm')"
   ]
  }
 ],
 "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": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}