This chapter will introduce some basic NLP concepts, such as word tokenization and regular expressions to help parse text. You'll also learn how to handle non-English text and more difficult tokenization you might find.

Introduction to regular expressions

Which pattern?

Practicing regular expressions: re.split() and re.findall()

Introduction to tokenization

Word tokenization with NLTK

More regex with re.search()

Advanced tokenization with NLTK and regex

Choosing a tokenizer

Regex with NLTK tokenization

Non-ascii tokenization

Charting word length with NLTK

Charting practice

Regular expressions & word tokenization

This chapter will introduce you to topic identification, which you can apply to any text you encounter in the wild. Using basic NLP models, you will identify topics from texts based on term frequencies. You'll experiment and compare two simple methods: bag-of-words and Tf-idf using NLTK, and a new library Gensim.

Word counts with bag-of-words

Bag-of-words picker

Building a Counter with bag-of-words

Simple text preprocessing

Text preprocessing steps

Text preprocessing practice

Introduction to gensim

What are word vectors?

Creating and querying a corpus with gensim

Gensim bag-of-words

Tf-idf with gensim

What is tf-idf?

Tf-idf with Wikipedia

Simple topic identification

This chapter will introduce a slightly more advanced topic: named-entity recognition. You'll learn how to identify the who, what, and where of your texts using pre-trained models on English and non-English text. You'll also learn how to use some new libraries, polyglot and spaCy, to add to your NLP toolbox.

Named Entity Recognition

NER with NLTK

Stanford library with NLTK

Introduction to SpaCy

Comparing NLTK with spaCy NER

spaCy NER Categories

Multilingual NER with polyglot

French NER with polyglot I

French NER with polyglot II

Spanish NER with polyglot

Named-entity recognition

You'll apply the basics of what you've learned along with some supervised machine learning to build a "fake news" detector. You'll begin by learning the basics of supervised machine learning, and then move forward by choosing a few important features and testing ideas to identify and classify fake news articles.

Fake News mit überwachten Lernverfahren und NLP klassifizieren

Möglichen Merkmale

Trainieren und Testen

Wortzählvektoren mit scikit-learn erstellen

CountVectorizer für Textklassifizierung

TfidfVectorizer für Textklassifikation

Vektoren analysieren

Ein Klassifikationsmodell mit scikit-learn trainieren und testen

Modelle zur Textklassifikation

Das „Fake News“-Modell mit CountVectorizer trainieren und testen

Training und Testen des „Fake News“-Modells mit TfidfVectorizer

Einfaches NLP, komplexe Probleme

Das Modell verbessern

Dein Modell verbessern

Dein Modell untersuchen

Building a "fake news" classifier

English stopwords

Monty Python and the Holy Grail

News articles

Wikipedia articles

In diesem Kurs lernst du die Grundlagen der natürlichen Sprachverarbeitung (NLP) kennen: wie man Wörter erkennt und trennt, wie man Themen in einem Text extrahiert und wie man einen eigenen Fake-News-Klassifikator baut. Außerdem lernst du, grundlegende Bibliotheken wie NLTK zu nutzen sowie Bibliotheken, die Deep Learning einsetzen, um gängige NLP-Probleme zu lösen. Dieser Kurs legt das Fundament, um Text zu verarbeiten und zu parsen, während du mit deinem Python-Lernen vorankommst.

Python Toolbox

Lerne grundlegende NLP-Techniken mit Python und wende sie an, um Erkenntnisse aus Textdaten zu gewinnen.

Einführung in Natural Language Processing mit Python

Lerne grundlegende Techniken der natürlichen Sprachverarbeitung mit Python und wie du sie anwendest, um Erkenntnisse aus realen Textdaten zu gewinnen.

Founder, kjamistan

Training and testing a classification model with scikit-learn

Naive Bayes classifier

Naive Bayes with scikit-learn

Confusion matrix

Let's practice!

Building a "fake news" classifier - Ein Klassifikationsmodell mit scikit-learn trainieren und testen

Ein Klassifikationsmodell mit scikit-learn trainieren und testen

Create Your Free Account