Accéder au contenu principal

cours

Feature Engineering for NLP in Python

Avancé

Updated 12/2024

Learn techniques to extract useful information from text and process them into a format suitable for machine learning.

Commencer le cours gratuitement

Inclus gratuitementPremium or Teams

PythonMachine learning4 heures15 vidéos52 exercices4,200 XP25,235Déclaration de réalisation

Créez votre compte gratuit

Google LinkedIn Facebook

ou

En continuant, vous acceptez nos Conditions d'utilisation, notre Politique de confidentialité et le fait que vos données sont stockées aux États-Unis.

Formation de 2 personnes ou plus ?

Essayer DataCamp for Business

Apprécié par les apprenants de milliers d’entreprises

Description du cours

In this course, you will learn techniques that will allow you to extract useful information from text and process them into a format suitable for applying ML models. More specifically, you will learn about POS tagging, named entity recognition, readability scores, the n-gram and tf-idf models, and how to implement them using scikit-learn and spaCy. You will also learn to compute how similar two documents are to each other. In the process, you will predict the sentiment of movie reviews and build movie and Ted Talk recommenders. Following the course, you will be able to engineer critical features out of any text and solve some of the most challenging problems in data science!

Conditions préalables

Introduction to Natural Language Processing in Python Supervised Learning with scikit-learn

1

Basic features and readability scores

Commencer le chapitre

Introduction to NLP feature engineering

Data format for ML algorithms

One-hot encoding

Basic feature extraction

Character count of Russian tweets

Word count of TED talks

Hashtags and mentions in Russian tweets

Readability tests

Readability of 'The Myth of Sisyphus'

Readability of various publications

2

Text preprocessing, POS tagging and NER

Commencer le chapitre

Tokenization and Lemmatization

Identifying lemmas

Tokenizing the Gettysburg Address

Lemmatizing the Gettysburg address

Text cleaning

Cleaning a blog post

Cleaning TED talks in a dataframe

Part-of-speech tagging

POS tagging in Lord of the Flies

Counting nouns in a piece of text

Noun usage in fake news

Named entity recognition

Named entities in a sentence

Identifying people mentioned in a news article

3

N-Gram models

Commencer le chapitre

Building a bag of words model

Word vectors with a given vocabulary

BoW model for movie taglines

Analyzing dimensionality and preprocessing

Mapping feature indices with feature names

Building a BoW Naive Bayes classifier

BoW vectors for movie reviews

Predicting the sentiment of a movie review

Building n-gram models

n-gram models for movie tag lines

Higher order n-grams for sentiment analysis

Comparing performance of n-gram models

4

TF-IDF and similarity scores

Commencer le chapitre

Building tf-idf document vectors

tf-idf weight of commonly occurring words

tf-idf vectors for TED talks

Cosine similarity

Range of cosine scores

Computing dot product

Cosine similarity matrix of a corpus

Building a plot line based recommender

Comparing linear_kernel and cosine_similarity

Plot recommendation engine

The recommender function

TED talk recommender

Beyond n-grams: word embeddings

Generating word vectors

Computing similarity of Pink Floyd songs

Congratulations!

Feature Engineering for NLP in Python

Cours
terminé

Earn Déclaration de réalisation

Ajoutez ces informations d’identification à votre profil LinkedIn, à votre CV ou à votre CV
Partagez-le sur les réseaux sociaux et dans votre évaluation de performance

Inclus avecPremium or Teams

S'inscrire maintenant

Inscrivez-vous 15 millions d’apprenants et commencer Feature Engineering for NLP in Python Aujourd’hui!

Créez votre compte gratuit

Google LinkedIn Facebook

ou

En continuant, vous acceptez nos Conditions d'utilisation, notre Politique de confidentialité et le fait que vos données sont stockées aux États-Unis.