Building a Lead Scoring Model with EvalML¶
In this demo, we will build an optimized lead scoring model using EvalML. To optimize the pipeline, we will set up an objective function to maximize the revenue generated with true positives while taking into account the cost of false positives. At the end of this demo, we also show you how introducing the right objective during the training is significantly better than using a generic machine learning metric like AUC.
[1]:
import evalml
from evalml import AutoMLSearch
from evalml.objectives import LeadScoring
Configure LeadScoring¶
To optimize the pipelines toward the specific business needs of this model, you can set your own assumptions for how much value is gained through true positives and the cost associated with false positives. These parameters are
true_positive
- dollar amount to be gained with a successful leadfalse_positive
- dollar amount to be lost with an unsuccessful lead
Using these parameters, EvalML builds a pileline that will maximize the amount of revenue per lead generated.
[2]:
lead_scoring_objective = LeadScoring(
true_positives=100,
false_positives=-5
)
Dataset¶
We will be utilizing a dataset detailing a customer’s job, country, state, zip, online action, the dollar amount of that action and whether they were a successful lead.
[3]:
from urllib.request import urlopen
import pandas as pd
import woodwork as ww
customers_data = urlopen('https://featurelabs-static.s3.amazonaws.com/lead_scoring_ml_apps/customers.csv')
interactions_data = urlopen('https://featurelabs-static.s3.amazonaws.com/lead_scoring_ml_apps/interactions.csv')
leads_data = urlopen('https://featurelabs-static.s3.amazonaws.com/lead_scoring_ml_apps/previous_leads.csv')
customers = pd.read_csv(customers_data)
interactions = pd.read_csv(interactions_data)
leads = pd.read_csv(leads_data)
X = customers.merge(interactions, on='customer_id').merge(leads, on='customer_id')
y = X['label']
X = X.drop(['customer_id', 'date_registered', 'birthday','phone', 'email',
'owner', 'company', 'id', 'time_x',
'session', 'referrer', 'time_y', 'label', 'country'], axis=1)
display(X.head())
job | state | zip | action | amount | |
---|---|---|---|---|---|
0 | Engineer, mining | NY | 60091.0 | page_view | NaN |
1 | Psychologist, forensic | CA | NaN | purchase | 135.23 |
2 | Psychologist, forensic | CA | NaN | page_view | NaN |
3 | Air cabin crew | NaN | 60091.0 | download | NaN |
4 | Air cabin crew | NaN | 60091.0 | page_view | NaN |
We will convert our data into Woodwork data structures. Doing so enables us to have more control over the types passed to and inferred by AutoML.
[4]:
X.ww.init(semantic_tags={'job': 'category'}, logical_types={'job': 'Categorical'})
y = ww.init_series(y)
X.ww
[4]:
Physical Type | Logical Type | Semantic Tag(s) | |
---|---|---|---|
Column | |||
job | category | Categorical | ['category'] |
state | category | Categorical | ['category'] |
zip | float64 | Double | ['numeric'] |
action | category | Categorical | ['category'] |
amount | float64 | Double | ['numeric'] |
Search for the best pipeline¶
In order to validate the results of the pipeline creation and optimization process, we will save some of our data as a holdout set.
EvalML natively supports one-hot encoding and imputation so the above NaN
and categorical values will be taken care of.
[5]:
X_train, X_holdout, y_train, y_holdout = evalml.preprocessing.split_data(X, y, problem_type='binary', test_size=0.2, random_seed=0)
X.ww
[5]:
Physical Type | Logical Type | Semantic Tag(s) | |
---|---|---|---|
Column | |||
job | category | Categorical | ['category'] |
state | category | Categorical | ['category'] |
zip | float64 | Double | ['numeric'] |
action | category | Categorical | ['category'] |
amount | float64 | Double | ['numeric'] |
Because the lead scoring labels are binary, we will use set the problem type to “binary”. When we call .search()
, the search for the best pipeline will begin.
[6]:
automl = AutoMLSearch(X_train=X_train, y_train=y_train,
problem_type='binary',
objective=lead_scoring_objective,
additional_objectives=['auc'],
allowed_model_families=["catboost", "random_forest", "linear_model"],
max_batches=1,
verbose=True)
automl.search()
Generating pipelines to search over...
4 pipelines ready for search.
*****************************
* Beginning pipeline search *
*****************************
Optimizing for Lead Scoring.
Greater score is better.
Using SequentialEngine to train and score pipelines.
Searching up to 1 batches for a total of 5 pipelines.
Allowed model families: linear_model, linear_model, catboost, random_forest
Evaluating Baseline Pipeline: Mode Baseline Binary Classification Pipeline
Mode Baseline Binary Classification Pipeline:
Starting cross validation
Finished cross validation - mean Lead Scoring: 0.000
*****************************
* Evaluating Batch Number 1 *
*****************************
Elastic Net Classifier w/ Label Encoder + Imputer + One Hot Encoder + Oversampler + Standard Scaler:
Starting cross validation
Finished cross validation - mean Lead Scoring: 0.028
Logistic Regression Classifier w/ Label Encoder + Imputer + One Hot Encoder + Oversampler + Standard Scaler:
Starting cross validation
Finished cross validation - mean Lead Scoring: 0.028
CatBoost Classifier w/ Label Encoder + Imputer + Oversampler:
Starting cross validation
Finished cross validation - mean Lead Scoring: 0.358
Random Forest Classifier w/ Label Encoder + Imputer + One Hot Encoder + Oversampler:
Starting cross validation
Finished cross validation - mean Lead Scoring: 0.031
Search finished after 00:11
Best pipeline: CatBoost Classifier w/ Label Encoder + Imputer + Oversampler
Best pipeline Lead Scoring: 0.358194
View rankings and select pipeline¶
Once the fitting process is done, we can see all of the pipelines that were searched, ranked by their score on the lead scoring objective we defined.
[7]:
automl.rankings
[7]:
id | pipeline_name | search_order | mean_cv_score | standard_deviation_cv_score | validation_score | percent_better_than_baseline | high_variance_cv | parameters | |
---|---|---|---|---|---|---|---|---|---|
0 | 3 | CatBoost Classifier w/ Label Encoder + Imputer... | 3 | 0.358194 | 0.328811 | -0.016129 | inf | False | {'Imputer': {'categorical_impute_strategy': 'm... |
1 | 4 | Random Forest Classifier w/ Label Encoder + Im... | 4 | 0.031193 | 0.041485 | -0.016129 | inf | False | {'Imputer': {'categorical_impute_strategy': 'm... |
2 | 1 | Elastic Net Classifier w/ Label Encoder + Impu... | 1 | 0.027979 | 0.065303 | -0.012903 | inf | False | {'Imputer': {'categorical_impute_strategy': 'm... |
3 | 2 | Logistic Regression Classifier w/ Label Encode... | 2 | 0.027979 | 0.065303 | -0.012903 | inf | False | {'Imputer': {'categorical_impute_strategy': 'm... |
4 | 0 | Mode Baseline Binary Classification Pipeline | 0 | 0.000000 | 0.000000 | 0.000000 | 0.0 | False | {'Baseline Classifier': {'strategy': 'mode'}} |
To select the best pipeline we can call automl.best_pipeline
.
[8]:
best_pipeline = automl.best_pipeline
Describe pipeline¶
You can get more details about any pipeline, including how it performed on other objective functions by calling .describe_pipeline()
and specifying the id
of the pipeline.
[9]:
automl.describe_pipeline(automl.rankings.iloc[0]["id"])
****************************************************************
* CatBoost Classifier w/ Label Encoder + Imputer + Oversampler *
****************************************************************
Problem Type: binary
Model Family: CatBoost
Pipeline Steps
==============
1. Label Encoder
2. Imputer
* categorical_impute_strategy : most_frequent
* numeric_impute_strategy : mean
* categorical_fill_value : None
* numeric_fill_value : None
3. Oversampler
* sampling_ratio : 0.25
* k_neighbors_default : 5
* n_jobs : -1
* sampling_ratio_dict : None
* categorical_features : [0, 1, 3]
* k_neighbors : 5
4. CatBoost Classifier
* n_estimators : 10
* eta : 0.03
* max_depth : 6
* bootstrap_type : None
* silent : True
* allow_writing_files : False
* n_jobs : -1
Training
========
Training for binary problems.
Objective to optimize binary classification pipeline thresholds for: <evalml.objectives.lead_scoring.LeadScoring object at 0x7fae6e80e580>
Total training time (including CV): 1.2 seconds
Cross Validation
----------------
Lead Scoring AUC # Training # Validation
0 -0.016 0.869 3,099 1,550
1 0.490 0.887 3,099 1,550
2 0.600 0.889 3,100 1,549
mean 0.358 0.882 - -
std 0.329 0.011 - -
coef of var 0.918 0.012 - -
Evaluate on hold out¶
Finally, since the best pipeline was trained on all of the training data, we evaluate it on the holdout dataset.
[10]:
best_pipeline_score = best_pipeline.score(X_holdout, y_holdout, objectives=["auc", lead_scoring_objective])
best_pipeline_score
[10]:
OrderedDict([('AUC', 0.8585599879117558),
('Lead Scoring', 0.8469475494411006)])
Why optimize for a problem-specific objective?¶
To demonstrate the importance of optimizing for the right objective, let’s search for another pipeline using AUC, a common machine learning metric. After that, we will score the holdout data using the lead scoring objective to see how the best pipelines compare.
[11]:
automl_auc = evalml.AutoMLSearch(X_train=X_train, y_train=y_train,
problem_type='binary',
objective='auc',
additional_objectives=[lead_scoring_objective],
allowed_model_families=["catboost", "random_forest", "linear_model"],
max_batches=1,
verbose=True)
automl_auc.search()
Generating pipelines to search over...
4 pipelines ready for search.
*****************************
* Beginning pipeline search *
*****************************
Optimizing for AUC.
Greater score is better.
Using SequentialEngine to train and score pipelines.
Searching up to 1 batches for a total of 5 pipelines.
Allowed model families: linear_model, linear_model, catboost, random_forest
Evaluating Baseline Pipeline: Mode Baseline Binary Classification Pipeline
Mode Baseline Binary Classification Pipeline:
Starting cross validation
Finished cross validation - mean AUC: 0.500
*****************************
* Evaluating Batch Number 1 *
*****************************
Elastic Net Classifier w/ Label Encoder + Imputer + One Hot Encoder + Oversampler + Standard Scaler:
Starting cross validation
Finished cross validation - mean AUC: 0.672
Logistic Regression Classifier w/ Label Encoder + Imputer + One Hot Encoder + Oversampler + Standard Scaler:
Starting cross validation
Finished cross validation - mean AUC: 0.672
CatBoost Classifier w/ Label Encoder + Imputer + Oversampler:
Starting cross validation
Finished cross validation - mean AUC: 0.882
Random Forest Classifier w/ Label Encoder + Imputer + One Hot Encoder + Oversampler:
Starting cross validation
Finished cross validation - mean AUC: 0.662
Search finished after 00:07
Best pipeline: CatBoost Classifier w/ Label Encoder + Imputer + Oversampler
Best pipeline AUC: 0.881709
[12]:
automl_auc.rankings
[12]:
id | pipeline_name | search_order | mean_cv_score | standard_deviation_cv_score | validation_score | percent_better_than_baseline | high_variance_cv | parameters | |
---|---|---|---|---|---|---|---|---|---|
0 | 3 | CatBoost Classifier w/ Label Encoder + Imputer... | 3 | 0.881709 | 0.010861 | 0.869201 | 38.170947 | False | {'Imputer': {'categorical_impute_strategy': 'm... |
1 | 1 | Elastic Net Classifier w/ Label Encoder + Impu... | 1 | 0.671885 | 0.050182 | 0.727798 | 17.188511 | False | {'Imputer': {'categorical_impute_strategy': 'm... |
2 | 2 | Logistic Regression Classifier w/ Label Encode... | 2 | 0.671517 | 0.048732 | 0.725816 | 17.151664 | False | {'Imputer': {'categorical_impute_strategy': 'm... |
3 | 4 | Random Forest Classifier w/ Label Encoder + Im... | 4 | 0.662322 | 0.050673 | 0.718588 | 16.232241 | False | {'Imputer': {'categorical_impute_strategy': 'm... |
4 | 0 | Mode Baseline Binary Classification Pipeline | 0 | 0.500000 | 0.000000 | 0.500000 | 0.000000 | False | {'Baseline Classifier': {'strategy': 'mode'}} |
Like before, we can look at the rankings and pick the best pipeline.
[13]:
best_pipeline_auc = automl_auc.best_pipeline
[14]:
# get the auc and lead scoring score on holdout data
best_pipeline_auc_score = best_pipeline_auc.score(X_holdout, y_holdout, objectives=["auc", lead_scoring_objective])
best_pipeline_auc_score
[14]:
OrderedDict([('AUC', 0.8585599879117558),
('Lead Scoring', 0.08168529664660361)])
[15]:
assert best_pipeline_score['Lead Scoring'] > best_pipeline_auc_score['Lead Scoring']
assert best_pipeline_auc_score['Lead Scoring'] >= 0
When we optimize for AUC, we can see that the AUC score from this pipeline is similar to the AUC score from the pipeline optimized for lead scoring. However, the revenue per lead is much smaller per lead when optimized for AUC and was much larger when optimized for lead scoring. As a result, we would have a huge gain on the amount of revenue if we optimized for lead scoring.
This happens because optimizing for AUC does not take into account the user-specified true_positive
(dollar amount to be gained with a successful lead) and false_positive
(dollar amount to be lost with an unsuccessful lead) values. Thus, the best pipelines may produce the highest AUC but may not actually generate the most revenue through lead scoring.
This example highlights how performance in the real world can diverge greatly from machine learning metrics.