When does class start/end?
Classes begin promptly at 9:00 am, and typically end at 5:00 pm.
Text Analytics and Natural Language Processing (NLP) with R
Accelebrate's Natural Language Processing (NLP) with R training course teaches attendees how to use R programming to explore and analyze text data. This class comprehensively covers methods for...
Read MoreAccelebrate's Natural Language Processing (NLP) with R training course teaches attendees how to use R programming to explore and analyze text data. This class comprehensively covers methods for ingesting text data from a variety of sources such as plain text files, pdfs, or the web, and then processing that data using the latest natural language processing and deep learning techniques.
Skills Gained
Students will be able to
- Import text data from a variety of source formats
- Tokenize text data to meaningful units
- Wrangle text data using specific textual functions
- Compute aggregating measures on tokenized data
- Translate between text data formats
- Complete a sentiment analysis
- Perform document classification
- Perform topic modeling
- Build a simple neural network appropriate for NLP modeling
Prerequisites
Students must have completed Accelebrate's Intro to R Programming training or have the equivalent experience. Students should have a working knowledge of the R language, RStudio, and the dplyr/tidyverse packages.
Course Details
Training Materials
All R Programming training students receive a copy of O’Reilly's Text Mining with R and related courseware.
Software Requirements
- A recent release of R 4.x
- IDE or text editor of your choice (RStudio recommended)
Outline
- Working with unstructured text data
- string methods
- regex
- reading in text files
- review of base (R/Python)
- Importing
- parsing data from a text file
- importing it into a tidy structure
- parsing data from a pdf
- From a “pile of pdfs”
- scraping data from the web
- Discussion of other methods
- OCR
- Handwriting recognition
- Managing Text Data 1
- a tidy text format
- Overview of text data formats
- tidy text
- token list
- Bag of words
- document term matrix or document frequency matrix (dfm/dt)
- corpus
- docvars
- associated formats
- stop words
- Sentiment lexica
- word vectors / models
- Managing Text Data 2
- tokenizing text
- units of tokenization
- tokens
- lemma
- stems
- n-grams
- sentences
- Tweets
- Tf-idf
- Log-odds (tidylo)
- Sentiment Analysis
- Sentiment lexica
- Sentiment analysis with inner_join
- Analyzing by other units
- Valence shifting
- VADER
- Document Classification
- Text similarity - stringiest
- Cosine
- Edit distance
- Machine Learning for document classification
- Naive Bayes model
- Text similarity - stringiest
- Topic Modeling / Document Clustering
- LDA
- stm
- Text and Deep Learning
- Deep learning introduction
- Architecture of neural networks
- Tensorflow + keras
- Word vectors
- word2vec
- Text2vec
- GloVe
- Spacy
- Combining Deep Learning and NLP
- CNN
- RNN
- LSTM
- Named Entity Recognition (NER)
- Part of Speech tagging (POS)
- Dependency Parsing
- Conclusion