Using longer n-grams

So far you have created features based on individual words in each of the texts. This can be quite powerful when used in a machine learning model but you may be concerned that by looking at words individually a lot of the context is being ignored. To deal with this when creating models you can use n-grams which are sequence of n words grouped together. For example:

bigrams: Sequences of two consecutive words
trigrams: Sequences of two consecutive words

These can be automatically created in your dataset by specifying the ngram_range argument as a tuple (n1, n2) where all n-grams in the n1 to n2 range are included.

This exercise is part of the course

Feature Engineering for Machine Learning in Python

Exercise instructions

Import CountVectorizer from sklearn.feature_extraction.text.
Instantiate CountVectorizer while considering only trigrams.
Fit the vectorizer and apply it to the text_clean column in one step.
Print the feature names generated by the vectorizer.

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

# Import CountVectorizer
from sklearn.feature_extraction.text import ____

# Instantiate a trigram vectorizer
cv_trigram_vec = CountVectorizer(max_features=100, 
                                 stop_words='english', 
                                 ____)

# Fit and apply trigram vectorizer
cv_trigram = ____(speech_df['text_clean'])

# Print the trigram features
print(cv_trigram_vec.____)

Edit and Run Code

This exercise is part of the course

Feature Engineering for Machine Learning in Python

IntermediateSkill Level

4.8+

Start Course for Free

In this chapter, you will explore what feature engineering is and how to get started with applying it to real-world data. You will load, explore and visualize a survey response dataset, and in doing so you will learn about its underlying data types and why they have an influence on how you should engineer your features. Using the pandas package you will create new features from both categorical and continuous columns.

Exercise 1: Why generate features?Exercise 2: Getting to know your data Exercise 3: Selecting specific data types Exercise 4: Dealing with categorical features Exercise 5: One-hot encoding and dummy variables Exercise 6: Dealing with uncommon categories Exercise 7: Numeric variables Exercise 8: Binarizing columns Exercise 9: Binning values

This chapter introduces you to the reality of messy and incomplete data. You will learn how to find where your data has missing values and explore multiple approaches on how to deal with them. You will also use string manipulation techniques to deal with unwanted characters in your dataset.

Exercise 1: Why do missing values exist?Exercise 2: How sparse is my data?Exercise 3: Finding the missing values Exercise 4: Dealing with missing values (I)Exercise 5: Listwise deletion Exercise 6: Replacing missing values with constants Exercise 7: Dealing with missing values (II)Exercise 8: Filling continuous missing values Exercise 9: Imputing values in predictive models Exercise 10: Dealing with other data issues Exercise 11: Dealing with stray characters (I)Exercise 12: Dealing with stray characters (II)Exercise 13: Method chaining

In this chapter, you will focus on analyzing the underlying distribution of your data and whether it will impact your machine learning pipeline. You will learn how to deal with skewed data and situations where outliers may be negatively impacting your analysis.

Exercise 1: Data distributions Exercise 2: What does your data look like? (I)Exercise 3: What does your data look like? (II)Exercise 4: When don't you have to transform your data?Exercise 5: Scaling and transformations Exercise 6: Normalization Exercise 7: Standardization Exercise 8: Log transformation Exercise 9: When can you use normalization?Exercise 10: Removing outliers Exercise 11: Percentage based outlier removal Exercise 12: Statistical outlier removal Exercise 13: Scaling and transforming new data Exercise 14: Train and testing transformations (I)Exercise 15: Train and testing transformations (II)

Finally, in this chapter, you will work with unstructured text data, understanding ways in which you can engineer columnar features out of a text corpus. You will compare how different approaches may impact how much context is being extracted from a text, and how to balance the need for context, without too many features being created.

Exercise 1: Encoding text Exercise 2: Cleaning up your text Exercise 3: High level text features Exercise 4: Word counts Exercise 5: Counting words (I)Exercise 6: Counting words (II)Exercise 7: Limiting your features Exercise 8: Text to DataFrame Exercise 9: Term frequency-inverse document frequency Exercise 10: Tf-idf Exercise 11: Inspecting Tf-idf values Exercise 12: Transforming unseen data Exercise 13: N-grams Exercise 14: Using longer n-grams

Current Exercise

Exercise 15: Finding the most common words Exercise 16: Wrap-up