I am working on a project using Python and Jupyter. This is what I have to work with: # $-*-$ coding: utf$-8-*-$

$"""$

Created on Fri Sep $202024$

@author: CS$634$ Group $1,$ Fall $2024$

$"""$

import pandas as pd

import json

import numpy as np

# Assign Dataset JSON File Paths

EN$\_$FILE$\_$PATH $=$ "DataSets$\backslash$Eng$\_$sentences.json"

FRA$\_$FILE$\_$PATH $=$ "DataSets$\backslash$Fra$\_$sentences.json"

MULTI$\_$LANG$\_$SENTENCE$\_$FILE$\_$PATH $=$ "DataSets$\backslash$CC$0\_$sentences.json"

LINKS$\_$FILE$\_$PATH $=$ "DataSets$\backslash$Links$.$json" # Contains Translations to match the sentences from CC$0\_$sentences.json

# Read the JSON file into a DataFrame

en$\_$data $=$ pd$.$read$\_$json$($EN$\_$FILE$\_$PATH, lines$=$True$)$ # English

fra$\_$data $=$ pd$.$read$\_$json$($FRA$\_$FILE$\_$PATH, lines$=$True$)$ # French

sentences$\_$data $=$ pd$.$read$\_$json$($MULTI$\_$LANG$\_$SENTENCE$\_$FILE$\_$PATH, lines$=$True$)$ # Multi Lang Sentences

with open$($LINKS$\_$FILE$\_$PATH, $\text{'}$r$\text{'})$ as link$\_$file: # Link$/$Translation

link$\_$data $=[]$

for line in link$\_$file:

link$\_$data.append$($json$.$loads$($line$))$

# Create a DataFrame

en$\_$data.set$\_$index$(\text{'}$Sentence id$\text{'},$ inplace$=$True$)$ # English

fra$\_$data.set$\_$index$(\text{'}$Sentence id$\text{'},$ inplace$=$True$)$ # French

sentences$\_$data.set$\_$index$(\text{'}$Sentence id$\text{'},$ inplace$=$True$)$ # Multi Lang Sentences

link$\_$df $=$ pd$.$DataFrame$($link$\_$data$)$ # Link$/$Translation

# Reset index to include 'Sentence id$\text{'}$ as a column

en$\_$data.reset$\_$index$($inplace$=$True$)$ # English

fra$\_$data.reset$\_$index$($inplace$=$True$)$ # French

sentences$\_$data.reset$\_$index$($inplace$=$True$)$ # Multi Lang Sentences

link$\_$df$.$reset$\_$index$($drop$=$True, inplace$=$True$)$ # Link$/$Translation

# Display the first few rows of the DataFrame

print$($en$\_$data.head$())$

print$($fra$\_$data.head$())$

print$($sentences$\_$data.head$())$

print$($link$\_$df$.$head$())$

# $=============================================================================$

# Owen's part

# $=============================================================================$

# Sentences$\_$data contains all the sentences available on the website

# Count number of sentences in each available languages

lang$\_$count $=$ sentences$\_$data.groupby$([\text{'}$Lang$\text{'}])[\text{'}$Sentence id$\text{'}].$agg$(\text{'}$count$\text{'}).$reset$\_$index$()$

lang$\_$count $=$ lang$\_$count.rename$($columns$=\{\text{'}$Sentence id$\text{'}$:'lang$\_$count'$\})$

idx $=$ lang$\_$count.lang$\_$count.idxmax$()$

print$(\text{'}$

$\{\}$ language has $\{\}$ available sentences which is the most in dataset'.format$($lang$\_$count.iloc$[$idx$][0],$ lang$\_$count.iloc$[$idx$][1]))$

# link$\_$df contains the links between the sentences. $177$ means that sentence #$77$ is the translation of sentence #$1.$

# The reciprocal link is also present, so the file will also contain a line that says $771.$

# Count number of sentences that has been used

link$\_$translate $=$ link$\_$df$.$groupby$([\text{'}$Sentence id$\text{'}]).$agg$(\text{'}$count$\text{'}).$reset$\_$index$()$

link$\_$translate $=$ link$\_$translate.rename$($columns$=\{\text{'}$Translation id$\text{'}$:'translation$\_$count'$\})$

# Get the translations

link$\_$translate $=$ link$\_$translate.merge$($sentences$\_$data.iloc$[$:$,$:$3],$ how $=$ 'left', on $=$ 'Sentence id$\text{'})$

# Keep only the valid translation

valid$\_$translation $=$ link$\_$translate$[$~link$\_$translate.Lang.isna$()]$

valid$\_$translation$\_$count $=$ valid$\_$translation.groupby$(\text{'}$Lang$\text{'})[\text{'}$Sentence id$\text{'}].$agg$(\text{'}$count$\text{'}).$reset$\_$index$()$

# Count number of each language valid translations

valid$\_$translation$\_$count $=$ valid$\_$translation$\_$count.rename$($columns$=\{\text{'}$Sentence id$\text{'}$:'valid$\_$count'$\})$

idx $=$ valid$\_$translation$\_$count.valid$\_$count.idxmax$()$

print$(\text{'}$

$\{\}$ language has $\{\}$ valid sentences which is the most in dataset'.format$($valid$\_$translation$\_$count.iloc$[$idx$][0],$ valid$\_$translation$\_$count.iloc$[$idx$][1]))$

# Most popular translated sentence

idx $=$ valid$\_$translation.translation$\_$count.idxmax$()$

print$(\text{'}$

$\{\}$ is the most translated sentences'.format$($valid$\_$translation.loc$[$idx$][-1]))$

# Translation that has the most different meaning in English

en$\_$data $=$ en$\_$data.merge$($link$\_$df$,$ how $=$ 'left', right$\_$on $=$ 'Sentence id$\text{'},$ left$\_$on $=$ 'Sentence id$\text{'})$

en$\_$translate$\_$count $=$ en$\_$data$[\text{'}$Translation id$\text{'}].$value$\_$counts$()$

idx $=$ en$\_$translate$\_$count.index$[0]$

print$(\text{'}$

Translation that has the most different meaning in English:

$\text{'})$

en$\_$data$[$en$\_$data$[\text{'}$Translation id$\text{'}]==$ idx$][[\text{'}$Translation id$\text{'},$'Text'$]]$

# Translation that has the most different meaning in French

fra$\_$data $=$ fra$\_$data.merge$($link$\_$df$,$ how $=$ 'left', right$\_$on $=$ 'Sentence id$\text{'},$ left$\_$on $=$ 'Sentence id$\text{'})$

fra$\_$translate$\_$count $=$ fra$\_$data$[\text{'}$Translation id$\text{'}].$value$\_$counts$()$

idx $=$ fra$\_$translate$\_$count.index$[0]$

print$(\text{'}$

Translation that has the most different meaning in French:

$\text{'})$

fra$\_$data$[$fra$\_$data$[\text{'}$Translation id$\text{'}]==$ idx$][[\text{'}$Translation id$\text{'},$'Text'$]]$