Using Text Data with EvalML

In this demo, we will show you how to use EvalML to build models which use text data.

import evalml
from evalml import AutoMLSearch

Dataset

We will be utilizing a dataset of SMS text messages, some of which are categorized as spam, and others which are not (“ham”). This dataset is originally from Kaggle, but modified to produce a slightly more even distribution of spam to ham.

from urllib.request import urlopen
import pandas as pd

input_data = urlopen('https://featurelabs-static.s3.amazonaws.com/spam_text_messages_modified.csv')
data = pd.read_csv(input_data)[:750]

X = data.drop(['Category'], axis=1)
y = data['Category']

display(X.head())

	Message
0	Free entry in 2 a wkly comp to win FA Cup fina...
1	FreeMsg Hey there darling it's been 3 week's n...
2	WINNER!! As a valued network customer you have...
3	Had your mobile 11 months or more? U R entitle...
4	SIX chances to win CASH! From 100 to 20,000 po...

The ham vs spam distribution of the data is 3:1, so any machine learning model must get above 75% accuracy in order to perform better than a trivial baseline model which simply classifies everything as ham.

y.value_counts(normalize=True)

spam    0.593333
ham     0.406667
Name: Category, dtype: float64

In order to properly utilize Woodwork’s ‘Natural Language’ typing, we need to pass this argument in during initialization. Otherwise, this will be treated as an ‘Unknown’ type and dropped in the search.

X.ww.init(logical_types={"Message": "NaturalLanguage"})

Search for 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.

X_train, X_holdout, y_train, y_holdout = evalml.preprocessing.split_data(X, y, problem_type='binary', test_size=0.2, random_seed=0)

EvalML uses Woodwork to automatically detect which columns are text columns, so you can run search normally, as you would if there was no text data. We can print out the logical type of the Message column and assert that it is indeed inferred as a natural language column.

X_train.ww

	Physical Type	Logical Type	Semantic Tag(s)
Column
Message	string	NaturalLanguage	[]

Because the spam/ham labels are binary, we will use AutoMLSearch(X_train=X_train, y_train=y_train, problem_type='binary'). When we call .search(), the search for the best pipeline will begin.

automl = AutoMLSearch(X_train=X_train, y_train=y_train,
                      problem_type='binary',
                      max_batches=1,
                      optimize_thresholds=True,
                      verbose=True)

automl.search()

Generating pipelines to search over...
8 pipelines ready for search.

*****************************
* Beginning pipeline search *
*****************************

Optimizing for Log Loss Binary.
Lower score is better.

Using SequentialEngine to train and score pipelines.
Searching up to 1 batches for a total of 9 pipelines.
Allowed model families: linear_model, linear_model, xgboost, lightgbm, catboost, random_forest, decision_tree, extra_trees

Evaluating Baseline Pipeline: Mode Baseline Binary Classification Pipeline
Mode Baseline Binary Classification Pipeline:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 14.046

*****************************
* Evaluating Batch Number 1 *
*****************************

Elastic Net Classifier w/ Label Encoder + Text Featurization Component + Imputer + Standard Scaler:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.350
Logistic Regression Classifier w/ Label Encoder + Text Featurization Component + Imputer + Standard Scaler:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.350
XGBoost Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.266
LightGBM Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.292
CatBoost Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.589
Random Forest Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.221
Decision Tree Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 3.386
        High coefficient of variation (cv >= 0.5) within cross validation scores.
        Decision Tree Classifier w/ Label Encoder + Text Featurization Component + Imputer may not perform as estimated on unseen data.
Extra Trees Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.292

Search finished after 00:31
Best pipeline: Random Forest Classifier w/ Label Encoder + Text Featurization Component + Imputer
Best pipeline Log Loss Binary: 0.221422

View rankings and select pipeline

Once the fitting process is done, we can see all of the pipelines that were searched.

automl.rankings

	id	pipeline_name	search_order	mean_cv_score	standard_deviation_cv_score	validation_score	percent_better_than_baseline	high_variance_cv	parameters
0	6	Random Forest Classifier w/ Label Encoder + Te...	6	0.221422	0.040958	0.221587	98.423568	False	{'Imputer': {'categorical_impute_strategy': 'm...
1	3	XGBoost Classifier w/ Label Encoder + Text Fea...	3	0.266164	0.106501	0.242896	98.105025	False	{'Imputer': {'categorical_impute_strategy': 'm...
2	4	LightGBM Classifier w/ Label Encoder + Text Fe...	4	0.291768	0.114862	0.291521	97.922737	False	{'Imputer': {'categorical_impute_strategy': 'm...
3	8	Extra Trees Classifier w/ Label Encoder + Text...	8	0.292373	0.029893	0.325764	97.918427	False	{'Imputer': {'categorical_impute_strategy': 'm...
4	2	Logistic Regression Classifier w/ Label Encode...	2	0.350340	0.074833	0.349271	97.505728	False	{'Imputer': {'categorical_impute_strategy': 'm...
5	1	Elastic Net Classifier w/ Label Encoder + Text...	1	0.350471	0.074886	0.349437	97.504795	False	{'Imputer': {'categorical_impute_strategy': 'm...
6	5	CatBoost Classifier w/ Label Encoder + Text Fe...	5	0.588944	0.004016	0.592259	95.806967	False	{'Imputer': {'categorical_impute_strategy': 'm...
7	7	Decision Tree Classifier w/ Label Encoder + Te...	7	3.385551	0.672118	3.708759	75.896294	True	{'Imputer': {'categorical_impute_strategy': 'm...
8	0	Mode Baseline Binary Classification Pipeline	0	14.045769	0.099705	13.988204	0.000000	False	{'Baseline Classifier': {'strategy': 'mode'}}

To select the best pipeline we can call automl.best_pipeline.

best_pipeline = automl.best_pipeline

Describe pipeline

You can get more details about any pipeline, including how it performed on other objective functions.

automl.describe_pipeline(automl.rankings.iloc[0]["id"])

**************************************************************************************
* Random Forest Classifier w/ Label Encoder + Text Featurization Component + Imputer *
**************************************************************************************

Problem Type: binary
Model Family: Random Forest

Pipeline Steps
==============
1. Label Encoder
2. Text Featurization Component
3. Imputer
         * categorical_impute_strategy : most_frequent
         * numeric_impute_strategy : mean
         * categorical_fill_value : None
         * numeric_fill_value : None
4. Random Forest Classifier
         * n_estimators : 100
         * max_depth : 6
         * n_jobs : -1

Training
========
Training for binary problems.
Total training time (including CV): 3.7 seconds

Cross Validation
----------------
             Log Loss Binary  MCC Binary  Gini   AUC  Precision    F1  Balanced Accuracy Binary  Accuracy Binary # Training # Validation
0                      0.222       0.817 0.950 0.975      0.862 0.893                     0.913            0.910        400          200
1                      0.180       0.875 0.970 0.985      0.937 0.925                     0.936            0.940        400          200
2                      0.262       0.783 0.925 0.963      0.918 0.865                     0.883            0.895        400          200
mean                   0.221       0.825 0.948 0.974      0.906 0.894                     0.910            0.915          -            -
std                    0.041       0.047 0.023 0.011      0.039 0.030                     0.026            0.023          -            -
coef of var            0.185       0.057 0.024 0.012      0.043 0.034                     0.029            0.025          -            -

best_pipeline.graph()

Notice above that there is a Text Featurization Component as the first step in the pipeline. AutoMLSearch uses the woodwork accessor to recognize that 'Message' is a text column, and converts this text into numerical values that can be handled by the estimator.

Evaluate on holdout

Now, we can score the pipeline on the holdout data using the core objectives for binary classification problems.

scores = best_pipeline.score(X_holdout, y_holdout,  objectives=evalml.objectives.get_core_objectives('binary'))
print(f'Accuracy Binary: {scores["Accuracy Binary"]}')

Accuracy Binary: 0.96

As you can see, this model performs relatively well on this dataset, even on unseen data.

What does the Text Featurization Component do?

Machine learning models cannot handle non-numeric data. Any text must be broken down into numeric features that provide useful information about that text. The Text Featurization component first normalizes your text by removing any punctuation and other non-alphanumeric characters and converting any capital letters to lowercase. From there, it passes the text into featuretools’ nlp_primitives dfs search, resulting in several informative features that replace the original column in your dataset: Diversity Score, Mean Characters per Word, Polarity Score, and LSA (Latent Semantic Analysis).

Diversity Score is the ratio of unique words to total words.

Mean Characters per Word is the average number of letters in each word.

Polarity Score is a prediction of how “polarized” the text is, on a scale from -1 (extremely negative) to 1 (extremely positive).

Latent Semantic Analysis is an abstract representation of how important each word is with respect to the entire text, reduced down into two values per text. While the other text features are each a single column, this feature adds two columns to your data, LSA(column_name)[0] and LSA(column_name)[1].

Let’s see what this looks like with our spam/ham example.

best_pipeline.input_feature_names

{'Label Encoder': ['Message'],
 'Text Featurization Component': ['Message'],
 'Imputer': ['DIVERSITY_SCORE(Message)',
  'MEAN_CHARACTERS_PER_WORD(Message)',
  'POLARITY_SCORE(Message)',
  'LSA(Message)[0]',
  'LSA(Message)[1]'],
 'Random Forest Classifier': ['DIVERSITY_SCORE(Message)',
  'MEAN_CHARACTERS_PER_WORD(Message)',
  'POLARITY_SCORE(Message)',
  'LSA(Message)[0]',
  'LSA(Message)[1]']}

Here, the Text Featurization component takes in a single “Message” column, but then the next component in the pipeline, the Imputer, receives five columns of input. These five columns are the result of featurizing the text-type “Message” column. Most importantly, these featurized columns are what ends up passed in to the estimator.

If the dataset had any non-text columns, those would be left alone by this process. If the dataset had more than one text column, each would be broken into these five feature columns independently.

The features, more directly

Rather than just checking the new column names, let’s examine the output of this component directly. We can see this by running the component on its own.

text_featurizer = evalml.pipelines.components.TextFeaturizer()
X_featurized = text_featurizer.fit_transform(X_train)

Now we can compare the input data to the output from the text featurizer:

X_train.head()

	Message
296	Sunshine Hols. To claim ur med holiday send a ...
652	Yup ü not comin :-(
526	Hello hun how ru? Its here by the way. Im good...
571	I tagged MY friends that you seemed to count a...
472	What happened to our yo date?

X_featurized.head()

	DIVERSITY_SCORE(Message)	MEAN_CHARACTERS_PER_WORD(Message)	POLARITY_SCORE(Message)	LSA(Message)[0]	LSA(Message)[1]
296	1.0	4.344828	0.003	0.150556	-0.072443
652	1.0	3.000000	0.000	0.017340	-0.005411
526	1.0	3.363636	0.162	0.169954	0.022670
571	0.8	4.083333	0.681	0.144713	0.036799
472	1.0	3.833333	0.000	0.109373	-0.042754

These numeric values now represent important information about the original text that the estimator at the end of the pipeline can successfully use to make predictions.

Why encode text this way?

To demonstrate the importance of text-specific modeling, let’s train a model with the same dataset, without letting AutoMLSearch detect the text column. We can change this by explicitly setting the data type of the 'Message' column in Woodwork to Categorical using the utility method infer_feature_types.

from evalml.utils import infer_feature_types
X = infer_feature_types(X, {'Message': 'Categorical'})
X_train, X_holdout, y_train, y_holdout = evalml.preprocessing.split_data(X, y, problem_type='binary', test_size=0.2, random_seed=0)

automl_no_text = AutoMLSearch(X_train=X_train, y_train=y_train,
                              problem_type='binary',
                              max_batches=1,
                              optimize_thresholds=True,
                              verbose=True)

automl_no_text.search()

Generating pipelines to search over...
8 pipelines ready for search.

*****************************
* Beginning pipeline search *
*****************************

Optimizing for Log Loss Binary.
Lower score is better.

Using SequentialEngine to train and score pipelines.
Searching up to 1 batches for a total of 9 pipelines.
Allowed model families: linear_model, linear_model, xgboost, lightgbm, catboost, random_forest, decision_tree, extra_trees

Evaluating Baseline Pipeline: Mode Baseline Binary Classification Pipeline
Mode Baseline Binary Classification Pipeline:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 14.046

*****************************
* Evaluating Batch Number 1 *
*****************************

Elastic Net Classifier w/ Label Encoder + Text Featurization Component + Imputer + Standard Scaler:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.350
Logistic Regression Classifier w/ Label Encoder + Text Featurization Component + Imputer + Standard Scaler:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.350
XGBoost Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.266
LightGBM Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.292
CatBoost Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.589
Random Forest Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.221
Decision Tree Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 3.386
        High coefficient of variation (cv >= 0.5) within cross validation scores.
        Decision Tree Classifier w/ Label Encoder + Text Featurization Component + Imputer may not perform as estimated on unseen data.
Extra Trees Classifier w/ Label Encoder + Text Featurization Component + Imputer:
        Starting cross validation
        Finished cross validation - mean Log Loss Binary: 0.292

Search finished after 00:28
Best pipeline: Random Forest Classifier w/ Label Encoder + Text Featurization Component + Imputer
Best pipeline Log Loss Binary: 0.221422

Like before, we can look at the rankings and pick the best pipeline.

automl_no_text.rankings

	id	pipeline_name	search_order	mean_cv_score	standard_deviation_cv_score	validation_score	percent_better_than_baseline	high_variance_cv	parameters
0	6	Random Forest Classifier w/ Label Encoder + Te...	6	0.221422	0.040958	0.221587	98.423568	False	{'Imputer': {'categorical_impute_strategy': 'm...
1	3	XGBoost Classifier w/ Label Encoder + Text Fea...	3	0.266164	0.106501	0.242896	98.105025	False	{'Imputer': {'categorical_impute_strategy': 'm...
2	4	LightGBM Classifier w/ Label Encoder + Text Fe...	4	0.291768	0.114862	0.291521	97.922737	False	{'Imputer': {'categorical_impute_strategy': 'm...
3	8	Extra Trees Classifier w/ Label Encoder + Text...	8	0.292373	0.029893	0.325764	97.918427	False	{'Imputer': {'categorical_impute_strategy': 'm...
4	2	Logistic Regression Classifier w/ Label Encode...	2	0.350340	0.074833	0.349271	97.505728	False	{'Imputer': {'categorical_impute_strategy': 'm...
5	1	Elastic Net Classifier w/ Label Encoder + Text...	1	0.350471	0.074886	0.349437	97.504795	False	{'Imputer': {'categorical_impute_strategy': 'm...
6	5	CatBoost Classifier w/ Label Encoder + Text Fe...	5	0.588944	0.004016	0.592259	95.806967	False	{'Imputer': {'categorical_impute_strategy': 'm...
7	7	Decision Tree Classifier w/ Label Encoder + Te...	7	3.385551	0.672118	3.708759	75.896294	True	{'Imputer': {'categorical_impute_strategy': 'm...
8	0	Mode Baseline Binary Classification Pipeline	0	14.045769	0.099705	13.988204	0.000000	False	{'Baseline Classifier': {'strategy': 'mode'}}

best_pipeline_no_text = automl_no_text.best_pipeline

Here, changing the data type of the text column removed the Text Featurization Component from the pipeline.

best_pipeline_no_text.graph()

automl_no_text.describe_pipeline(automl_no_text.rankings.iloc[0]["id"])

**************************************************************************************
* Random Forest Classifier w/ Label Encoder + Text Featurization Component + Imputer *
**************************************************************************************

Problem Type: binary
Model Family: Random Forest

Pipeline Steps
==============
1. Label Encoder
2. Text Featurization Component
3. Imputer
         * categorical_impute_strategy : most_frequent
         * numeric_impute_strategy : mean
         * categorical_fill_value : None
         * numeric_fill_value : None
4. Random Forest Classifier
         * n_estimators : 100
         * max_depth : 6
         * n_jobs : -1

Training
========
Training for binary problems.
Total training time (including CV): 3.7 seconds

Cross Validation
----------------
             Log Loss Binary  MCC Binary  Gini   AUC  Precision    F1  Balanced Accuracy Binary  Accuracy Binary # Training # Validation
0                      0.222       0.817 0.950 0.975      0.862 0.893                     0.913            0.910        400          200
1                      0.180       0.875 0.970 0.985      0.937 0.925                     0.936            0.940        400          200
2                      0.262       0.783 0.925 0.963      0.918 0.865                     0.883            0.895        400          200
mean                   0.221       0.825 0.948 0.974      0.906 0.894                     0.910            0.915          -            -
std                    0.041       0.047 0.023 0.011      0.039 0.030                     0.026            0.023          -            -
coef of var            0.185       0.057 0.024 0.012      0.043 0.034                     0.029            0.025          -            -

# get standard performance metrics on holdout data
scores = best_pipeline_no_text.score(X_holdout, y_holdout, objectives=evalml.objectives.get_core_objectives('binary'))
print(f'Accuracy Binary: {scores["Accuracy Binary"]}')

Accuracy Binary: 0.96

Without the Text Featurization Component, the 'Message' column was treated as a categorical column, and therefore the conversion of this text to numerical features happened in the One Hot Encoder. The best pipeline encoded the top 10 most frequent “categories” of these texts, meaning 10 text messages were one-hot encoded and all the others were dropped. Clearly, this removed almost all of the information from the dataset, as we can see the best_pipeline_no_text performs very similarly to randomly guessing “ham” in every case.

previous

Using the Cost-Benefit Matrix Objective

next

User Guide