Skip to content

Instantly share code, notes, and snippets.

Show Gist options
  • Save carlthome/6d2fce69124c9724bdd247c4e0500d11 to your computer and use it in GitHub Desktop.
Save carlthome/6d2fce69124c9724bdd247c4e0500d11 to your computer and use it in GitHub Desktop.
Predicting a grocery store chain's customer satisfaction index.
{
"cells": [
{
"metadata": {},
"cell_type": "markdown",
"source": "# Predicting a grocery store chain's customer satisfaction index\n**Given survey responses and transaction data from a Swedish chain of grocery stores, we'd like to predict customers' satisfaction, called the _Customer Satisfaction Index (CSI)_.**\n\nThis was my contribution to [Nepa](http://nepa.com/)'s first machine learning Modelathon (a data science hackathon) that was held the 29th of September, 2016 in Stockholm, Sweden."
},
{
"metadata": {},
"cell_type": "markdown",
"source": "## Load and clean data\nOur data is roughly 150000 samples of 33 features, where each sample is a customer's information such as their age, how close they are to their grocery store, how many times they've visited recently, how much they like to buy of certain product groups, and so on."
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"ExecuteTime": {
"start_time": "2016-09-29T23:03:07.086469",
"end_time": "2016-09-29T23:03:07.397485"
}
},
"cell_type": "code",
"source": "import os\n\nfrom IPython.display import display\nimport matplotlib.pyplot as plt\nimport numpy as np\nimport pandas as pd\n\n%matplotlib inline\nplt.xkcd();",
"execution_count": 1,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"scrolled": true,
"trusted": true,
"ExecuteTime": {
"start_time": "2016-09-29T23:03:07.398600",
"end_time": "2016-09-29T23:03:09.665157"
}
},
"cell_type": "code",
"source": "# Load data.\ndf = pd.read_csv(os.path.join('Data', 'ModelathonData.csv'), encoding='latin_1')\ntest = pd.read_csv(os.path.join('Data', 'FinalTestLabels.csv'), encoding='latin_1')\ntest.index = df[df['CSI'] == 0].index\ndf.update(test)\n\n# DistanceToStore contains junk after a tab character. Strip that out.\ndf['DistanceToStore'] = df['DistanceToStore'].str.split('\\t').apply(lambda x: x[0]).astype('float')\n\n# InterviewDate is a string, let's parse it to datetime.\nfrom datetime import datetime\ndf['InterviewDate'] = df['InterviewDate'].apply(lambda s: datetime.strptime(s, '%m/%d/%Y %H:%M').timestamp())\n\n# Feature selection. Note: we want to predict CSI so it cannot be a feature.\nlabels = df['CSI']\nfeatures = df[[x for x in df.columns \n if x.startswith('ShareBoughtAt') or x.endswith('Id')] + [\n 'Gender', \n 'CSIaverage', \n 'InterviewDate', \n 'BirthYear', \n 'PurchasingPower', \n 'VisitsLast3Months', \n 'AmountLast3Months',\n 'StoreSquaremeters',\n 'DistanceToStore']]",
"execution_count": 2,
"outputs": []
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"ExecuteTime": {
"start_time": "2016-09-29T23:03:09.666279",
"end_time": "2016-09-29T23:03:09.933406"
}
},
"cell_type": "code",
"source": "from sklearn.cross_validation import train_test_split\nfrom sklearn.preprocessing import StandardScaler\n\n# Split out test data.\nX_test, y_test = features.iloc[test.index], labels.iloc[test.index]\n\n# Whatever remains is our training data.\nX_train, y_train = map(lambda x: x.iloc[~x.index.isin(test.index)], \n [features, labels])\n\n# Normalize training data and split into test and validation sets.\nscaler = StandardScaler()\nX_train = scaler.fit_transform(X_train)\nX_train, X_val, y_train, y_val = train_test_split(X_train, \n y_train, \n test_size=0.1)",
"execution_count": 3,
"outputs": []
},
{
"metadata": {},
"cell_type": "markdown",
"source": "## Train models\nLet's try out some different models and see what performs best. For performance reasons we'll train every model in its own process."
},
{
"metadata": {
"collapsed": false,
"trusted": true,
"ExecuteTime": {
"start_time": "2016-09-29T23:03:09.934444",
"end_time": "2016-09-29T23:03:45.156302"
}
},
"cell_type": "code",
"source": "from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier\n\nclassifiers = [RandomForestClassifier(),\n AdaBoostClassifier()]\n\n# Also try XGBoost if it's installed.\ntry:\n from xgboost.sklearn import XGBClassifier\n classifiers.append(XGBClassifier())\nexcept:\n pass\n\ndef train(clf):\n clf.fit(X_train, y_train)\n score = clf.score(X_val, y_val)\n print(clf, score)\n return clf\n\nfrom multiprocessing import Pool\nwith Pool(len(classifiers)) as p:\n classifiers = p.map(train, classifiers)",
"execution_count": 4,
"outputs": [
{
"output_type": "stream",
"text": "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n max_depth=None, max_features='auto', max_leaf_nodes=None,\n min_samples_leaf=1, min_samples_split=2,\n min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,\n oob_score=False, random_state=None, verbose=0,\n warm_start=False) 0.21704\nAdaBoostClassifier(algorithm='SAMME.R', base_estimator=None,\n learning_rate=1.0, n_estimators=50, random_state=None) 0.24608\nXGBClassifier(base_score=0.5, colsample_bylevel=1, colsample_bytree=1,\n gamma=0, learning_rate=0.1, max_delta_step=0, max_depth=3,\n min_child_weight=1, missing=nan, n_estimators=100, nthread=-1,\n objective='multi:softprob', reg_alpha=0, reg_lambda=1,\n scale_pos_weight=1, seed=0, silent=True, subsample=1) 0.26752\n",
"name": "stdout"
}
]
},
{
"metadata": {},
"cell_type": "markdown",
"source": "## Test models\nLet's see how well the trained models perform on unseen customer data."
},
{
"metadata": {
"collapsed": false,
"scrolled": false,
"trusted": true,
"ExecuteTime": {
"start_time": "2016-09-29T23:03:45.157496",
"end_time": "2016-09-29T23:03:47.099978"
}
},
"cell_type": "code",
"source": "import matplotlib.mlab as mlab\n\nX = scaler.transform(X_test)\npredictions = [y_test]\nlabels = ['Test data']\nfor i, clf in enumerate(classifiers):\n predictions.append(clf.predict(X))\n labels.append('{} ({:0.2f}% accuracy)'.format(type(clf).__name__, \n 100*clf.score(X, y_test)))\ncolors = [c['color'] for c in plt.rcParams['axes.prop_cycle']][:len(predictions)]\nplt.figure()\nplt.hist(predictions, color=colors)\nfor i, p in enumerate(predictions):\n x = np.linspace(p.min(), p.max(), 100)\n plt.plot(x, y_test.size * mlab.normpdf(x, np.mean(p), np.std(p)), color=colors[i])\nplt.title('CSI distributions')\nplt.xlim([0, 10])\nplt.ylim([0, y_test.size])\nplt.legend(labels, loc='upper left', prop={'size': 9})\nplt.xlabel('Customer Satisfaction Index')\nplt.ylabel('Customers')\nplt.tight_layout()",
"execution_count": 5,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": "<matplotlib.figure.Figure at 0x7f2f720a4320>",
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAkEAAAF2CAYAAABpk0wlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKnyVAAAAAElFTkSuQmCC\n"
},
"metadata": {}
}
]
},
{
"metadata": {},
"cell_type": "markdown",
"source": "## Conclusions\nUnfortunately our classes are imbalanced, and we have a lot of high CSI scores, and very few low ones, meaning almost every customer that got asked how satisfied they were actually seemed pretty content. That's good news for the grocery stores, but it's a problem for us, as we need our model to be familiar with not only the happy customers, but also (and maybe even especially!) the unhappy ones.\n\nDue to this, the more complex models suffer from overfitting. Even though the random forest got the lowest accuracy (when we require each sample to be labeled with the exact CSI), its distribution is smoothest and closer to the actual test data's. Assuming we're okay with guessing CSI +- a few points, a random forest is likely our best choice for doing predictions, at least as far as classifiers go. To be absolutely sure this is the case though we should also not rely too heavily on histograms and also plot confusion matrices first.\n\nFuture experiments should look into applying ordinal regression instead of standard classification. As CSI is ordinal we'd like our model to take into account that guessing CSI = 6, when the real label should have been seven, would be a lot better than just guessing CSI = 1, even though both guesses are equally wrong in a classification sense. Whether this delimitation makes sense depends on the user of the prediction system, of course, but it seems intuitive enough."
}
],
"metadata": {
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"pygments_lexer": "ipython3",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"version": "3.5.2",
"file_extension": ".py"
},
"anaconda-cloud": {},
"hide_input": false,
"kernelspec": {
"name": "conda-root-py",
"display_name": "Python [conda root]",
"language": "python"
},
"gist": {
"id": "",
"data": {
"description": "Predicting a grocery store chain's customer satisfaction index.",
"public": true
}
}
},
"nbformat": 4,
"nbformat_minor": 1
}
@Hopper1906
Copy link

Hello CarlTHome, I love the design of the CSI Distributions graph. How did you make it look like that?

@carlthome
Copy link
Author

Hi @Hopper1906! It's in the first cell. Run plt.xkcd() and the style should become like this.

@Hopper1906
Copy link

Wow, Thank you for replying Carl!

Really like that. So much less monotonous than a normal blah graph!
https://www.geeksforgeeks.org/matplotlib-pyplot-xkcd-in-python/

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment