diff --git "a/stats.ipynb" "b/stats.ipynb" new file mode 100644--- /dev/null +++ "b/stats.ipynb" @@ -0,0 +1,775 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Statistics on CIENSFO" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "#### Importating packages\n", + "import pandas as pd\n", + "import matplotlib.pyplot as plt" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Defining dataframes" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " sent_id sent_type noun adj verb negated adv_clause prep \\\n", + "0 1 2 dispute sur \n", + "1 2 2 conscience de \n", + "2 3 2 différence de \n", + "3 4 2 dépendre 0 de \n", + "4 5 2 discuter 0 de \n", + "\n", + " graft_prep non_std_type wh wh2 marker marker2 class note \n", + "0 qui Q=S V \n", + "1 qecq qecq QsekVinf \n", + "2 comment QSV \n", + "3 pourquoi QSV \n", + "4 comment QSV \n" + ] + } + ], + "source": [ + "#### Importing annotations\n", + "df = pd.read_csv( # dataframe of annotations\n", + " \"annotations.csv\",\n", + " dtype={\n", + " 'sent_id':str, 'sent_type':int,\n", + " 'noun':str, 'adj':str, 'verb':str,\n", + " 'negated':str,\n", + " 'adv_clause':str,\n", + " 'prep':str,\n", + " 'graft_prep':str,\n", + " 'non_std_type':str,\n", + " 'wh':str, 'wh2':str,\n", + " 'marker':str, 'marker2':str,\n", + " 'class':str,\n", + " 'note':str\n", + " }\n", + ").fillna(value = '')\n", + "\n", + "print(df.iloc[0:5])" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " sent_id sent_type noun adj verb negated adv_clause prep \\\n", + "0 1 2 dispute sur \n", + "1 2 2 conscience de \n", + "2 3 2 différence de \n", + "3 4 2 dépendre 0 de \n", + "4 5 2 discuter 0 de \n", + "\n", + " graft_prep non_std_type wh wh2 marker marker2 class note dep \n", + "0 qui Q=S V noun \n", + "1 qecq qecq QsekVinf noun \n", + "2 comment QSV noun \n", + "3 pourquoi QSV verb \n", + "4 comment QSV verb \n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_8496/3777369308.py:23: SettingWithCopyWarning: \n", + "A value is trying to be set on a copy of a slice from a DataFrame.\n", + "Try using .loc[row_indexer,col_indexer] = value instead\n", + "\n", + "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", + " core_df['dep'] = core_df.apply(dep_type, axis=1)\n" + ] + } + ], + "source": [ + "#### Defining subtables\n", + "\n", + "# core_df contains the reliable occurrences:\n", + "# type 2 or 4, introduced by a preposition and no 'warning' note\n", + "# also removing conjuncted and grafts\n", + "core_df = df.loc[(df['sent_type'] != 1) & (df['sent_type'] != 3) &\n", + " (df['prep'] != \"/\") & (df['note'] == \"\") &\n", + " (df['noun'] != \"CONJ\") & (df['adj'] != \"CONJ\") &\n", + " (df['verb'] != \"CONJ\") & (df['adv_clause'] != \"CONJ\") &\n", + " (df['graft_prep'] == \"\") ]\n", + "\n", + "def dep_type(df):\n", + " if df['noun'] != \"\":\n", + " return \"noun\"\n", + " if df['adj'] != \"\":\n", + " return \"adj\"\n", + " if df['verb'] != \"\":\n", + " return \"verb\"\n", + " if df['adv_clause'] != \"\":\n", + " return \"adv_clause\"\n", + " \n", + "# Creating additional column of dependence type\n", + "core_df['dep'] = core_df.apply(dep_type, axis=1)\n", + "\n", + "print(core_df.iloc[0:5])" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " sent_id sent_type noun adj verb negated adv_clause prep \\\n", + "8 9 2 en fonction de de \n", + "25 25 2 par rapport à à \n", + "31 31 2 dans dans \n", + "50 50 2 suivant suivant \n", + "52 52 2 en fonction de de \n", + "\n", + " graft_prep non_std_type wh wh2 marker marker2 class note dep \n", + "8 si si SV adv_clause \n", + "25 comment QSV adv_clause \n", + "31 qui Q=S V adv_clause \n", + "50 comment QSV adv_clause \n", + "52 quel QSV adv_clause \n" + ] + } + ], + "source": [ + "# pp_core contains the reliable occurrences complement of a noun,\n", + "# dependent on a noun, adjective or verb\n", + "pp_core = core_df.loc[core_df['dep'] != \"adv_clause\"]\n", + "noun_core = core_df.loc[core_df['dep'] == \"noun\"]\n", + "adj_core = core_df.loc[core_df['dep'] == \"adj\"]\n", + "verb_core = core_df.loc[core_df['dep'] == \"verb\"]\n", + "\n", + "# advcl_core contains the reliable occurrences of adverbial clauses\n", + "advcl_core = core_df.loc[core_df['dep'] == \"adv_clause\"]\n", + "print(advcl_core.iloc[0:5])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Statistics" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 10, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Histogram of preposition used\n", + "pp_core.hist(\"prep\", by=\"dep\")\n", + "advcl_core.hist(\"adv_clause\", by=\"dep\")" + ] + }, + { + "cell_type": "code", + "execution_count": 40, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 40, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Statistics on the interrogative word\n", + "# (only the first one is taken if the interrogative has multiple qu-words)\n", + "plt1 = pp_core.hist(\"wh\", by=\"dep\", sharex=True)\n", + "advcl_core.hist(\"wh\", by=\"dep\")" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 21, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Statistics on the interrogative class\n", + "plt1 = pp_core.hist(\"class\", by=\"dep\", sharex=True)\n", + "advcl_core.hist(\"class\", by=\"dep\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Tables for the CMLF paper" + ] + }, + { + "cell_type": "code", + "execution_count": 35, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
prepCIENSFOCEFC
0en fonction de132
1par rapport à10
2dans10
3suivant25
4selon76
5au niveau de01
\n", + "
" + ], + "text/plain": [ + " prep CIENSFO CEFC\n", + "0 en fonction de 13 2\n", + "1 par rapport à 1 0\n", + "2 dans 1 0\n", + "3 suivant 2 5\n", + "4 selon 7 6\n", + "5 au niveau de 0 1" + ] + }, + "execution_count": 35, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "### Table of prepositions of advcl\n", + "\n", + "# Creating table\n", + "prep_table = pd.DataFrame()\n", + "prep_table['prep'] = advcl_core['adv_clause'].unique()\n", + "\n", + "\n", + "# Function to count values of unique prepositions\n", + "def value_count(df:pd.DataFrame, sent_type:int) -> None:\n", + " prep = df['prep']\n", + " prep_col = advcl_core[advcl_core['sent_type'] == sent_type]['adv_clause']\n", + " counts = prep_col.value_counts()\n", + " if prep in counts.index:\n", + " return counts[prep]\n", + " return 0\n", + "\n", + "\n", + "# Applying the function to create two columns\n", + "prep_table['CIENSFO'] = prep_table.apply(lambda df: value_count(df, 2), axis=1) \n", + "prep_table['CEFC'] = prep_table.apply(lambda df: value_count(df, 4), axis=1) \n", + "\n", + "prep_table" + ] + }, + { + "cell_type": "code", + "execution_count": 46, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_13023/2771199511.py:4: SettingWithCopyWarning: \n", + "A value is trying to be set on a copy of a slice from a DataFrame.\n", + "Try using .loc[row_indexer,col_indexer] = value instead\n", + "\n", + "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n", + " advcl_core['wh_mark'] = advcl_core[['wh','marker']].apply(lambda x: ''.join(x), axis=1)\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
qu_markCIENSFOCEFC
0si63
1comment73
2qui13
3quel40
4pourquoi10
5combien10
6commentecq10
7où14
8qecq00
\n", + "
" + ], + "text/plain": [ + " qu_mark CIENSFO CEFC\n", + "0 si 6 3\n", + "1 comment 7 3\n", + "2 qui 1 3\n", + "3 quel 4 0\n", + "4 pourquoi 1 0\n", + "5 combien 1 0\n", + "6 commentecq 1 0\n", + "7 où 1 4\n", + "8 qecq 0 0" + ] + }, + "execution_count": 46, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "### Table of interrogative word of conditional advcl\n", + "\n", + "# Adding table of wh or marking:\n", + "advcl_core['wh_mark'] = advcl_core[['wh','marker']].apply(lambda x: ''.join(x), axis=1)\n", + "\n", + "# Creating table\n", + "qu_table = pd.DataFrame()\n", + "qu_table['qu_mark'] = advcl_core['wh_mark'].unique()\n", + "\n", + "\n", + "# Function to count values of unique prepositions\n", + "def value_count(df:pd.DataFrame, sent_type:int) -> None:\n", + " prep = df['qu_mark']\n", + " # excluding non conditional advcl\n", + " prep_col = advcl_core[(advcl_core['sent_type'] == sent_type)\n", + " & (advcl_core['adv_clause'] != \"par rapport à\")\n", + " & (advcl_core['adv_clause'] != \"dans\")\n", + " & (advcl_core['adv_clause'] != \"au niveau de\")]['wh_mark']\n", + " counts = prep_col.value_counts()\n", + " if prep in counts.index:\n", + " return counts[prep]\n", + " return 0\n", + "\n", + "\n", + "# Applying the function to create two columns\n", + "qu_table['CIENSFO'] = qu_table.apply(lambda df: value_count(df, 2), axis=1) \n", + "qu_table['CEFC'] = qu_table.apply(lambda df: value_count(df, 4), axis=1) \n", + "\n", + "qu_table" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "35\n", + "0.3142857142857143\n" + ] + } + ], + "source": [ + "print(6+7+1+4+1+1+1+1+3+3+3+4)\n", + "print(11/35)" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
classCIENSFOCEFC
0si SV63
1QSV138
2Q=S V11
3QESV10
4Q10
5Qsek GN00
6Q=S sekV01
\n", + "
" + ], + "text/plain": [ + " class CIENSFO CEFC\n", + "0 si SV 6 3\n", + "1 QSV 13 8\n", + "2 Q=S V 1 1\n", + "3 QESV 1 0\n", + "4 Q 1 0\n", + "5 Qsek GN 0 0\n", + "6 Q=S sekV 0 1" + ] + }, + "execution_count": 7, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "### Table of interrogative structures of conditional advcl\n", + "\n", + "# Creating table\n", + "class_table = pd.DataFrame()\n", + "class_table['class'] = advcl_core['class'].unique()\n", + "\n", + "\n", + "# Function to count values of unique prepositions\n", + "def value_count(df:pd.DataFrame, sent_type:int) -> None:\n", + " prep = df['class']\n", + " # excluding non conditional advcl\n", + " prep_col = advcl_core[(advcl_core['sent_type'] == sent_type)\n", + " & (advcl_core['adv_clause'] != \"par rapport à\")\n", + " & (advcl_core['adv_clause'] != \"dans\")\n", + " & (advcl_core['adv_clause'] != \"au niveau de\")]['class']\n", + " counts = prep_col.value_counts()\n", + " if prep in counts.index:\n", + " return counts[prep]\n", + " return 0\n", + "\n", + "\n", + "# Applying the function to create two columns\n", + "class_table['CIENSFO'] = class_table.apply(lambda df: value_count(df, 2), axis=1) \n", + "class_table['CEFC'] = class_table.apply(lambda df: value_count(df, 4), axis=1) \n", + "\n", + "class_table" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "35\n", + "0.6\n", + "0.2571428571428571\n" + ] + } + ], + "source": [ + "print(6+13+1+1+1+3+8+1+1)\n", + "print(21/35)\n", + "print(9/35)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "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": "3.8.10" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +}