Complete Guide to NLP with R
5h 5mAdvanced2024-08-01
Authors
Mark Niemann-Ross
Technologist experienced in hardware, software, and science fiction
Course details
Natural Language Processing is to words as Computer Vision is to pictures! Learn NLP with the R programming language. In this course, experienced technologist Mark Niemann-Ross shows you how to use the R programming language to implement natural language processing algorithms. R is uniquely adept at manipulating matrices and producing statistics, both of which are core to NLP. Learn about frameworks that you can use with NLP, as well as the importance of corpora and sources. Find out how to work with NLP metadata and preprocess text in preparation for NLP. Explore creating structured data, applying statistics to text, and performing sentiment analysis, and then dive into visualizing NLP. Discover ways to use tidytext and quanteda R for NLP. Build your understanding of corpora, tokens, and document-feature matrix (DFM). Plus, go over analysis and visualization.
Skills covered
RStatisticsNatural Language Processing (NLP)AdvancedArtificial Intelligence (AI)Programming LanguagesData ScienceOpen SourceSoftware Development
Concepts
Introduction
- Welcome to natural language processing with R
- Skills and tools you need to be successful in this course
Up and Running with tm
- What is tm and why do you need it
- Real-world NLP with tm
- Real-world NLP with quanteda
- Real-world NLP with tidytext
Corpora and Sources
- Understanding corpora and sources
- Examining corpora
- Examining sources
- Custom sources
- Combining and subsetting corpora
Working with NLP Metadata
- Working with document metadata
- Make useful metadata
- Finding and filtering based on metadata
Preprocessing Text in Preparation for NLP
- Transformations
- Stop words
- Stemming
- Lemmatization
- Tokenization
- N-grams
- Part of speech tagging
Create Structured Data
- Understanding the document-term matrix
- Create the document-term matrix
- Weighting the document-term matrix
- Focus the document-term matrix
Apply Statistics to Text
- Word and document frequency
- Hierarchical clustering
- Associated terms
Sentiment Analysis
- What is sentiment analysis
- Real-world example of sentiment analysis
- Sentiment datasets
- Sentiment tools
Visualizing Natural Language Processing
- Plotting text mining
- Plotting Zipf s and Heap s Law
- Word clouds
Conclusion
- Your next steps in NLP
Introduction to NLP Tidytext R
- Welcome to natural language processing with R
- Skills you need to be successful in this course
Use of Tidytext for NLP
- How to think like tidytext
- An example - Calculate the most popular terms in a document
- Tokenizing with unnest tokens( )
- Stopwords, punctuation, whitespace, and numbers
- Stemming and lemmatization
- Term frequency with bind tf idf( )
- Sentiment analysis with sentiments( )
- Parts of speech with parts of speech( )
- Import and export from other NLP packages
Conclusion
- Next steps
Introduction to NLP with Quanteda R
- Welcome to natural language processing with R
- Skills and tools you need
Getting Started with Quanteda
- Introduction to quanteda
- Install quanteda
Understanding Corpora
- Create a quanteda corpus
- Create metadata with docvars
- Corpus subsets and groups
- Reshape and segment a corpus
- Remove lines from a corpus
Understanding Tokens
- Corpus and tokens
- Remove tokens and stopwords
- Group tokens
- Stemming with tokens
Understanding Document-Feature Matrix (DFM)
- Corpus, tokens, and DFM
- Create and modify a DFM
- Real-world analysis with DFM
Analysis and Visualization
- The quanteda textstats package
- Real-world text statistics with textstats
- Understand the quanteda sentiment package
- Real-world sentiment analysis with quanteda sentiment
- Visualization with textplots
- Use dplyr with quanteda
Conclusion
- Your next steps in NLP
Capstone Project
- Project introduction
- Project explanation