notebooks/1_sentiment_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Sentiment Analysis\n",
    "\n",
    "**Exploring the Relationship Between User Prompt Sentiment and Vooyai's Country Recommendations**\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, lets go back to some previous results on how vooyai destination suggestions differ from real tourism stats.\n",
    "\n",
    "\n",
    " ![Ranking Differences](../figures/ranking_diffs.png)\n",
    " \n",
    "\n",
    "We can see how we have a notable difference with countries like Turkey, United Kingdom or Austria. Let's try to find the cause for this, in case there is some bias affecting the suggestion results.\n",
    "\n",
    "In this notebook we will work under **the hypothesis that prompt sentiment affects the results**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Imports and data loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Imports\n",
    "from dataloader import VooyaiDataloader\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "\n",
    "dataset = VooyaiDataloader(\"./data/\")\n",
    "ranking_df = pd.read_csv(\"./data/ranking.csv\")\n",
    "\n",
    "# Rename 'country' column to 'Country' in ranking_df\n",
    "ranking_df = ranking_df.rename(columns={'country': 'Country'})\n",
    "\n",
    "\n",
    "es_df = dataset.destinations_df[dataset.destinations_df['language'] == 'es']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading sentiment tools and preparing data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will use a model for spanish prompts and a different model for english prompts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pysentimiento import create_analyzer\n",
    "import transformers\n",
    "\n",
    "#Initialize sentiment analysis tools\n",
    "analyzer_es = create_analyzer(task=\"sentiment\", lang=\"es\")\n",
    "analyzer_en = create_analyzer(task=\"sentiment\", lang=\"en\")\n",
    "\n",
    "#Get discoveries dataset:\n",
    "#For example lets take an spanish prompt\n",
    "discoveries_es = dataset.discoveries_df[dataset.discoveries_df['language'] == 'es'].reset_index(drop=True)\n",
    "discoveries_en = dataset.discoveries_df[dataset.discoveries_df['language'] == 'en'].reset_index(drop=True)\n",
    "discoveries_en.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see how in english we have prompts in other languages too from people who used the app on their own language, it would be an interesting task to clean prompts based on language without supervision, but more on that later..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Analyzing Spanish Prompt\n",
      "Spanish model:  AnalyzerOutput(output=NEU, probas={NEU: 0.604, POS: 0.350, NEG: 0.046})\n",
      "English model:  AnalyzerOutput(output=NEU, probas={NEU: 0.946, POS: 0.049, NEG: 0.004})\n",
      "Analyzing English Prompt\n",
      "Spanish model:  AnalyzerOutput(output=POS, probas={POS: 0.547, NEU: 0.400, NEG: 0.053})\n",
      "English model:  AnalyzerOutput(output=POS, probas={POS: 0.865, NEU: 0.133, NEG: 0.001})\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "print(\"Analyzing Spanish Prompt\")\n",
    "print(\"Spanish model: \" ,analyzer_es.predict(discoveries_es.loc[0].prompt)) \n",
    "print(\"English model: \" ,analyzer_en.predict(discoveries_es.loc[0].prompt)) \n",
    "\n",
    "print(\"Analyzing English Prompt\")\n",
    "print(\"Spanish model: \" ,analyzer_es.predict(discoveries_en.loc[0].prompt)) \n",
    "print(\"English model: \" ,analyzer_en.predict(discoveries_en.loc[0].prompt)) \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see how there is a discrepancy in how models interpret sentiment if they are not tuned for the input language.\n",
    "\n",
    "In this section we will focus on using the spanish language analyzer, as we have more spanish prompts, and we have not yet removed other languages from the english dataset (there are some prompts in the spanish subset in other languages but the landing page defaults to english so that's were other language users prompted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm import tqdm\n",
    "\n",
    "tqdm.pandas()\n",
    "\n",
    "#Lets get a sentiment score for each prompt. We will only keep id and prompt for this\n",
    "ddf_es = discoveries_es[['id','prompt']]\n",
    "\n",
    "def getSentiment(prompt):\n",
    "    sentiment = analyzer_es.predict(prompt)\n",
    "    return sentiment.probas\n",
    "\n",
    "# Apply getSentiment function to each prompt and create new columns\n",
    "ddf_es['sentiment'] = ddf_es['prompt'].progress_apply(getSentiment) #As it takes a while this is useful to check eta\n",
    "\n",
    "# Extract sentiment scores into separate columns\n",
    "ddf_es['neg'] = ddf_es['sentiment'].apply(lambda x: x['NEG'])\n",
    "ddf_es['neu'] = ddf_es['sentiment'].apply(lambda x: x['NEU'])\n",
    "ddf_es['pos'] = ddf_es['sentiment'].apply(lambda x: x['POS'])\n",
    "\n",
    "# Drop the temporary 'sentiment' column\n",
    "ddf_es = ddf_es.drop('sentiment', axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we have sentiment, lets find out how this relates with the destinations suggested"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#First lets create a new dataframe adding to each discovery the sentiment of the prompts.\n",
    "# Merge the dataframes based on id and parent_id\n",
    "combined_es_df = es_df.merge(ddf_es,how='left',left_on='parent_id',right_on='id')\n",
    "\n",
    "#Remove redundant columns\n",
    "combined_es_df = combined_es_df.drop(columns='id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Now we keep only the countries from the top 10\n",
    "combined_es_df10 = combined_es_df[combined_es_df['countrycode'].isin(ranking_df['CC'])]\n",
    "\n",
    "#And get some sentiment statistics per country.\n",
    "cesdf10_stats = combined_es_df10.groupby('countrycode').agg({'neg':['mean','std'],'neu':['mean','std'],'pos':['mean','std']})\n",
    "#Sort index so it matches the ranking one\n",
    "cesdf10_stats = cesdf10_stats.reset_index().set_index('countrycode').loc[ranking_df['CC']].reset_index()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ranking and Sentiment Comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visual comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x1200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "#And now lets do some plotting.\n",
    "plt.rcParams['axes.facecolor'] = 'white'\n",
    "# Plotting\n",
    "# Assuming you have your data in pandas DataFrames: ranking_df and sentiment_df\n",
    "\n",
    "fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(14, 12), sharex=True)\n",
    "\n",
    "# Upper panel: Rank difference chart (your existing code)\n",
    "colors = plt.cm.Reds(np.abs(ranking_df['es_rank_diff']) / np.abs(ranking_df['es_rank_diff']).max())\n",
    "ax1.bar(ranking_df['Country'], ranking_df['es_rank_diff'], color=colors)\n",
    "ax1.set_title('Rank difference: Spanish Vooyai suggestions vs. real world visits')\n",
    "ax1.set_ylabel('Rank Difference')\n",
    "ax1.axhline(y=0, color='black', linestyle='-', linewidth=0.5)\n",
    "\n",
    "# Lower panel: Sentiment chart\n",
    "x = np.arange(len(ranking_df['Country']))\n",
    "width = 0.27\n",
    "error_kw=dict(lw=1, capsize=3, capthick=1)\n",
    "\n",
    "# Assuming sentiment_df has columns: Country, pos_mean, neu_mean, neg_mean, pos_std, neu_std, neg_std\n",
    "ax2.bar(x - width, -cesdf10_stats[('pos','mean')], width, label='Positive', yerr=cesdf10_stats[('pos','std')],error_kw=error_kw)\n",
    "ax2.bar(x,- cesdf10_stats[('neu','mean')], width, label='Neutral', yerr=cesdf10_stats[('neu','std')],error_kw=error_kw)\n",
    "ax2.bar(x + width, -cesdf10_stats[('neg','mean')], width, label='Negative', yerr=cesdf10_stats[('neg','std')],error_kw=error_kw)\n",
    "\n",
    "ax2.set_ylabel('Sentiment Score')\n",
    "ax2.set_title('User Sentiment for Recommended Countries')\n",
    "ax2.set_xticks(x)\n",
    "ax2.set_xticklabels(ranking_df['Country'], rotation=45, ha='right')\n",
    "ax2.legend()\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the graphs it may seem that Turkey tends to get suggested more with neutral prompts, while positive prompts tend to not suggest Turkey compared to the other countries. Let's further explore this.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Statistical tests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's use the countries with small rank differences for comparison: France, Spain, USA, Italy, México and Germany.\n",
    "countries_to_compare = ['FR', 'ES', 'US', 'IT','MX','DE']\n",
    "comparison_data = combined_es_df10[combined_es_df10['countrycode'].isin(countries_to_compare)]\n",
    "\n",
    "# Test if sentiments are significantly different between comparison countries and Turkey\n",
    "turkey_data = combined_es_df10[combined_es_df10['countrycode'] == 'TR']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualize the distribution of scores for Turkey and other countries:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize the distribution of sentiments for Turkey and combined selected countries\n",
    "import seaborn as sns\n",
    "\n",
    "plt.figure(figsize=(12, 6))\n",
    "\n",
    "# Prepare data for Turkey\n",
    "turkey_melted = pd.melt(turkey_data, id_vars=['countrycode'], value_vars=['pos', 'neu', 'neg'], var_name='sentiment', value_name='score')\n",
    "turkey_melted['group'] = 'Turkey'\n",
    "\n",
    "# Prepare data for other countries (combined)\n",
    "other_countries_melted = pd.melt(comparison_data, id_vars=['countrycode'], value_vars=['pos', 'neu', 'neg'], var_name='sentiment', value_name='score')\n",
    "other_countries_melted['group'] = 'Other Countries'\n",
    "\n",
    "# Combine the data\n",
    "combined_melted = pd.concat([turkey_melted, other_countries_melted])\n",
    "\n",
    "# Create the boxplot\n",
    "sns.boxplot(x='group', y='score', hue='sentiment', data=combined_melted)\n",
    "plt.title(\"Distribution of Sentiments for Turkey vs Other Selected Countries\")\n",
    "plt.ylabel(\"Sentiment Score\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see how turkey has lower positive scores and more neutral scores than other countries, although it also has less negatives.\n",
    "\n",
    "Lets perform some statistical tests to see if this difference in distribution is significant."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Statistical test for pos sentiment (Comparison countries vs Turkey):\n",
      "U-statistic: 84250.0 (n1=2871, n2=50)\n",
      "P-value: 0.03486196964697912\n",
      "\n",
      "Statistical test for neu sentiment (Comparison countries vs Turkey):\n",
      "U-statistic: 49677.0 (n1=2871, n2=50)\n",
      "P-value: 0.00018577888521714334\n",
      "\n",
      "Statistical test for neg sentiment (Comparison countries vs Turkey):\n",
      "U-statistic: 75715.0 (n1=2871, n2=50)\n",
      "P-value: 0.5051978305129257\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Perform statistical tests\n",
    "from scipy import stats\n",
    "\n",
    "# Perform Mann-Whitney U test so we dont assume normality.\n",
    "sentiments = ['pos', 'neu', 'neg']\n",
    "for sentiment in sentiments:\n",
    "    u_stat, p_value = stats.mannwhitneyu(comparison_data[sentiment], turkey_data[sentiment])\n",
    "    print(f\"Statistical test for {sentiment} sentiment (Comparison countries vs Turkey):\")\n",
    "    n1, n2 = len(comparison_data), len(turkey_data)\n",
    "    print(f\"U-statistic: {u_stat} (n1={n1}, n2={n2})\")\n",
    "    print(f\"P-value: {p_value}\")\n",
    "    print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From this test we can extract some early conclusions:\n",
    "\n",
    "-  The U-value is higher than half of the maximum (143550) suggesting that other countries tend to have slightly higher positive sentiments than turkey.\n",
    "-  On the other hand, the U-value for neutral is notably smaller than half of the maximum suggesting that the other countries get more neutral sentiment scores.\n",
    "-  The U-value for negative is close to half of the maximum (143550/2 = 71775) suggesting that the distribution is similar, although it has a high p-value so this is not statistically significant.\n",
    "\n",
    "Still, we can observe a huge discrepancy in sample size (2871 vs 50), which could affect the reliability of these results, so lets do some additional tests, using bootstrapping."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The idea is simple, sample multiple subsets from both groups that are as big as the smaller group, and compare them using again the Mann-Whitney U test. Then, keep the distribution of U-statistics and p-values of all comparisons, and calculate the mean of them. We do this for each sentiment. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Bootstrapping for pos sentiment:\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10000/10000 [00:04<00:00, 2186.21it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Bootstrapping for neu sentiment:\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10000/10000 [00:04<00:00, 2173.46it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Bootstrapping for neg sentiment:\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10000/10000 [00:04<00:00, 2158.91it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Pos Sentiment:\n",
      "Mean U-statistic: 1467.35\n",
      "Median U-statistic: 1467.50\n",
      "Proportion of significant results (p < 0.05): 32.05%\n",
      "\n",
      "Neu Sentiment:\n",
      "Mean U-statistic: 863.34\n",
      "Median U-statistic: 860.00\n",
      "Proportion of significant results (p < 0.05): 77.26%\n",
      "\n",
      "Neg Sentiment:\n",
      "Mean U-statistic: 1319.14\n",
      "Median U-statistic: 1318.00\n",
      "Proportion of significant results (p < 0.05): 8.10%\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from scipy import stats\n",
    "from tqdm import tqdm\n",
    "\n",
    "def bootstrap_mann_whitney(group1, group2, n_iterations=10000):\n",
    "    n_small = len(group2)  # Turkey sample size\n",
    "    u_stats = []\n",
    "    p_values = []\n",
    "    \n",
    "    for _ in tqdm(range(n_iterations)):\n",
    "        # Resample Turkey data with replacement\n",
    "        boot_turkey = np.random.choice(group2, size=n_small, replace=True)\n",
    "        \n",
    "        # Sample equal size from comparison countries\n",
    "        boot_comparison = np.random.choice(group1, size=n_small, replace=False)\n",
    "        \n",
    "        # Perform Mann-Whitney U test\n",
    "        u_stat, p_value = stats.mannwhitneyu(boot_comparison, boot_turkey)\n",
    "        \n",
    "        u_stats.append(u_stat)\n",
    "        p_values.append(p_value)\n",
    "    \n",
    "    return np.array(u_stats), np.array(p_values)\n",
    "\n",
    "# Perform bootstrapping for each sentiment\n",
    "sentiments = ['pos', 'neu', 'neg']\n",
    "results = {}\n",
    "\n",
    "for sentiment in sentiments:\n",
    "    print(f\"\\nBootstrapping for {sentiment} sentiment:\")\n",
    "    u_stats, p_values = bootstrap_mann_whitney(comparison_data[sentiment], turkey_data[sentiment])\n",
    "    results[sentiment] = {'u_stats': u_stats, 'p_values': p_values}\n",
    "\n",
    "# Analyze results\n",
    "fig, axes = plt.subplots(1, 3, figsize=(20, 5))\n",
    "fig.suptitle('Distribution of p-values for Sentiments')\n",
    "\n",
    "for i, sentiment in enumerate(sentiments):\n",
    "    u_stats = results[sentiment]['u_stats']\n",
    "    p_values = results[sentiment]['p_values']\n",
    "    \n",
    "    print(f\"\\n{sentiment.capitalize()} Sentiment:\")\n",
    "    print(f\"Mean U-statistic: {np.mean(u_stats):.2f}\")\n",
    "    print(f\"Median U-statistic: {np.median(u_stats):.2f}\")\n",
    "    print(f\"Proportion of significant results (p < 0.05): {np.mean(p_values < 0.05):.2%}\")\n",
    "\n",
    "    # Visualize distribution of p-values\n",
    "    axes[i].hist(p_values, bins=50, edgecolor='black')\n",
    "    axes[i].set_title(f'{sentiment.capitalize()} Sentiment')\n",
    "    axes[i].set_xlabel('p-value')\n",
    "    axes[i].set_ylabel('Frequency')\n",
    "    axes[i].axvline(0.05, color='red', linestyle='--', label='p = 0.05')\n",
    "    axes[i].legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we now analyze these results:\n",
    "  - We can seee how for positive sentiment, the U statistic is higher than half of the maximum (now is 50*50 = 2500), implying that other countries tend to have more positive sentiment, still now we can see how only about one third of the bootstrapping tests had a significant p-value, so its is not consistently different.\n",
    "  - On the other hand, for Neutral sentiment we have a much slower U statistic, and 77% of tests had a significance smaller than 0.05, suggesting that indeed Turkey has more neutral scores with statistical significance\n",
    "  - Negative sentiment has a value near half the maximum, but a very small significance, so there is no consistent difference in negative sentiment between turkey and other countries.\n",
    "\n",
    "Turkey consistently shows higher neutral scores than other countries.\n",
    "\n",
    "Lets see if thats also the case for UK"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "202"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Test if sentiments are significantly different between comparison countries and Turkey\n",
    "uk_data = combined_es_df10[combined_es_df10['countrycode'] == 'GR']\n",
    "len(uk_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Statistical test for pos sentiment (Comparison countries vs Turkey):\n",
      "U-statistic: 304146.5 (n1=2871, n2=202)\n",
      "P-value: 0.24483841936331963\n",
      "\n",
      "Statistical test for neu sentiment (Comparison countries vs Turkey):\n",
      "U-statistic: 266696.5 (n1=2871, n2=202)\n",
      "P-value: 0.05619773085933091\n",
      "\n",
      "Statistical test for neg sentiment (Comparison countries vs Turkey):\n",
      "U-statistic: 276083.5 (n1=2871, n2=202)\n",
      "P-value: 0.2545575089688714\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Perform Mann-Whitney U test so we dont assume normality.\n",
    "sentiments = ['pos', 'neu', 'neg']\n",
    "for sentiment in sentiments:\n",
    "    u_stat, p_value = stats.mannwhitneyu(comparison_data[sentiment], uk_data[sentiment])\n",
    "    print(f\"Statistical test for {sentiment} sentiment (Comparison countries vs Turkey):\")\n",
    "    n1, n2 = len(comparison_data), len(uk_data)\n",
    "    print(f\"U-statistic: {u_stat} (n1={n1}, n2={n2})\")\n",
    "    print(f\"P-value: {p_value}\")\n",
    "    print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Al values are close to half the maximum (~290k), with only neutral sentiment more notably below, and a p-value bigger than (albeit close to) 0.05, suggesting there may be again a different distribution for neutral sentiment.\n",
    "\n",
    "So we again perfom bootstrapping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Bootstrapping for pos sentiment:\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10000/10000 [00:05<00:00, 1976.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Bootstrapping for neu sentiment:\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10000/10000 [00:05<00:00, 1999.26it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Bootstrapping for neg sentiment:\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10000/10000 [00:04<00:00, 2003.24it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Pos Sentiment:\n",
      "Mean U-statistic: 21406.21\n",
      "Median U-statistic: 21401.00\n",
      "Proportion of significant results (p < 0.05): 13.53%\n",
      "\n",
      "Neu Sentiment:\n",
      "Mean U-statistic: 18785.65\n",
      "Median U-statistic: 18796.00\n",
      "Proportion of significant results (p < 0.05): 27.77%\n",
      "\n",
      "Neg Sentiment:\n",
      "Mean U-statistic: 19419.58\n",
      "Median U-statistic: 19412.75\n",
      "Proportion of significant results (p < 0.05): 13.05%\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Perform bootstrapping for each sentiment\n",
    "sentiments = ['pos', 'neu', 'neg']\n",
    "results = {}\n",
    "\n",
    "for sentiment in sentiments:\n",
    "    print(f\"\\nBootstrapping for {sentiment} sentiment:\")\n",
    "    u_stats, p_values = bootstrap_mann_whitney(comparison_data[sentiment], uk_data[sentiment])\n",
    "    results[sentiment] = {'u_stats': u_stats, 'p_values': p_values}\n",
    "\n",
    "# Analyze results\n",
    "fig, axes = plt.subplots(1, 3, figsize=(20, 5))\n",
    "fig.suptitle('Distribution of p-values for Sentiments')\n",
    "\n",
    "for i, sentiment in enumerate(sentiments):\n",
    "    u_stats = results[sentiment]['u_stats']\n",
    "    p_values = results[sentiment]['p_values']\n",
    "    \n",
    "    print(f\"\\n{sentiment.capitalize()} Sentiment:\")\n",
    "    print(f\"Mean U-statistic: {np.mean(u_stats):.2f}\")\n",
    "    print(f\"Median U-statistic: {np.median(u_stats):.2f}\")\n",
    "    print(f\"Proportion of significant results (p < 0.05): {np.mean(p_values < 0.05):.2%}\")\n",
    "\n",
    "    # Visualize distribution of p-values\n",
    "    axes[i].hist(p_values, bins=50, edgecolor='black')\n",
    "    axes[i].set_title(f'{sentiment.capitalize()} Sentiment')\n",
    "    axes[i].set_xlabel('p-value')\n",
    "    axes[i].set_ylabel('Frequency')\n",
    "    axes[i].axvline(0.05, color='red', linestyle='--', label='p = 0.05')\n",
    "    axes[i].legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And we can see how with bootstrapping the proportion of significant results is low with only 28.7% of them being lower than 0.05, so we cannot conclude the distribution is significantly different to that of other countries for UK"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### To close this sentiment section, lets calculate correlations between rank and sentiment scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Now lets see if there is some correlation with this.\n",
    "from scipy.stats import spearmanr\n",
    "\n",
    "# Merge ranking information with the combined dataset\n",
    "full_data = combined_es_df10.merge(ranking_df[['Country', 'es_rank_diff']], \n",
    "                                   left_on='countrycode', \n",
    "                                   right_on=ranking_df['CC'],how='left')\n",
    "\n",
    "# Function to format and print correlation results\n",
    "def print_correlations(correlations):\n",
    "    for corr_type, sentiments in correlations.items():\n",
    "        print('\\n')\n",
    "        for sentiment, (statistic, pvalue) in sentiments.items():\n",
    "            print(f\"{corr_type:<8} {sentiment:<8} Statistic: {statistic:8.4f}   P-value:{pvalue:8.4f}\")\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlations:\n",
      "\n",
      "\n",
      "Spearman Positive Statistic:  -0.0336   P-value:  0.0561\n",
      "Spearman Negative Statistic:  -0.0661   P-value:  0.0002\n",
      "Spearman Neutral  Statistic:   0.0749   P-value:  0.0000\n"
     ]
    }
   ],
   "source": [
    "# Calculate correlations\n",
    "correlations_original = {\n",
    "    'Spearman': {\n",
    "        'Positive': spearmanr(np.abs(full_data['es_rank_diff']), full_data['pos']),\n",
    "        'Negative': spearmanr(np.abs(full_data['es_rank_diff']), full_data['neg']),\n",
    "        'Neutral': spearmanr(np.abs(full_data['es_rank_diff']), full_data['neu'])\n",
    "    },\n",
    "}\n",
    "# Print results for both\n",
    "print(\"Correlations:\")\n",
    "print_correlations(correlations_original)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The correlations are very weak across all sentiments, with the strongest being only 0.0749 for neutral sentiment. \n",
    "\n",
    "Negative sentiment is negatively correlated, suggesting that a higher negative score results in a slower rank, which is the opposite of our initial guess. Still, it is a very weak correlation and probably not meaningful.\n",
    "\n",
    "Neutral sentiment on the other hand its correlated, suggesting that a h igher neutral score results in a higher rank, but again, this is a very weak correlation and probably not meaningful."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Overall, as we have not found strong evidence that support our initial hypothesis (prompt sentiment affects the results) we have to assume it is not true.**\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The next step will be trying to find out if the suggestion difference is built into the prompts, for example users asking for named places not in these countries, maybe with some Named Entity Recognition (NER)."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "vooyai",
   "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": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}