Text Analytics and Natural Language Processing (NLP) with R

Master basic text manipulation in R using paste for concatenation with separators or collapse, extract substrings, replace segments, and split strings, with packages like stringr and stringi.

Explore how to use regular expressions in R to detect, extract, and replace matching strings, using grep, escape sequences, and basic regex concepts.

Learn to find, replace, and detect regular expression matches in R using grep on string vectors, with pipe workflows, escaping, quantifiers, and character classes to build robust search patterns.

Learn to use the pipe operator in R to simplify command structure and write clear, readable code with magrittr and tidyverse, including examples with log, sign, pi, and dot placeholders.

use the pipe operator in r to chain commands for selecting columns and filtering by continent, then mutate and summarize gdp per capita and life expectancy.

Explore how the readtext package in R reads formats like txt, json, csv/tsv, pdf, and doc/docx, auto-detects formats, builds a corpus, and cleans page numbers using stringi and regex.

Master basic text pre-processing in R, including string manipulation, abbreviation, word and sentence tokenization, word length, stemming with snowball, and frequency analysis to prepare text data for analytics.

Master basic string manipulation in r by loading the stringr package, defining lines, using the combine function, and handling backslashes with write_lines, paste, and cat.

Explore tokenization in text analytics using R, creating tokens from words, phrases, or sentences, and generate n-grams such as bi-grams and skip-grams while handling punctuation and stop words from tweets.

Discover stemming, the process of reducing inflected words to their root forms to improve text analytics accuracy. The lecture demonstrates applying R-based stemming with different algorithms and dictionaries.

Learn how lemmatization links inflected forms to lemmas using dictionaries and morphological analysis, and compare its performance to stemming in text analytics with R.

Explore stop words in text data and learn how to remove them to improve text analytics in R, using English and multilingual lexicons such as snowball and smart.

Learn how to check scraping permission via robots.txt and path allowed, analyze site rules with examples, and use httr to make requests, all while practicing ethical and responsible scraping.

Learn to web scrape pages using selector gadget and R, extract reviews from sites like TripAdvisor and Indeed, and clean text for basic analysis.

Learn to scrape Indeed to extract job titles, company names, locations, and descriptions for the City of Toronto, loop through pages, build a data frame, and save and clean results.

Learn to scrape song lyrics from a website with r: load libraries, parse pages, extract titles and links, fetch lyrics, and compile a data frame with title, link, and lyrics.

Learn to scrape TripAdvisor hotel reviews with R, build a corpus, clean text, and perform sentiment analysis using positive and negative lexicons, culminating in visualizations of review sentiment.

Discover corpus analytics in r by working with a corpus, performing tokenization, frequency analysis, and n-grams, and exploring sentiment analysis or word clouds with quanteda, text2vec, tidytext, and tm.

Create a word cloud from English tweets through corpus analytics in R, cleaning and stemming text, filtering stop words, and visualizing top terms with color and frequency controls.

Learn to generate word clouds in R with a custom function, including excluding words, rendering word frequencies, and visualizing results with a frequency table and a bar plot.

Analyze a Gutenberg Project novel with R using the corpus package, load and tokenize text, extract chapters, clean punctuation, and compute chapter-level token, type, and sentence statistics.

Analyze lexical diversity and word usage with a linear model in corpus analytics part 2, applying stop words exclusion, by-grams, and text locate in R on the Wizard of Oz.

Load 170 documents into a corpus, perform pre-processing (punctuation and digit removal, lowercasing, stop-word removal, stemming), tokenize, then analyze term frequencies and tf-idf with visualizations.

Explore corpus analytics of multiple documents in R, computing IDF metrics, extracting key terms, performing sentiment analysis with tidytext lexicons, and visualizing by-grams and word networks.

Learn to use the quanteda package to work with corpora. Explore creating, summarizing, subsetting, and querying corpora with kwic and regular expressions.

Master quanteda part 2: search keywords with wildcards and multiword phrases; build document feature matrices with DSM, apply stop word removal and stemming, and explore word clouds and clustering.

Explore quanteda features for term similarity, document scaling, and topic modeling in R, including wordfish and LDA, to analyze a corpus.

Explore the core concepts of natural language processing, including natural language generation and understanding, and see how NLP with R enables text analytics, sentiment analysis, and topic modeling.

Analyze song lyrics with sentiment analysis in natural language processing, loading datasets, cleaning tokens, and applying Bing and NRC lexicons to classify words by sentiment and emotion.

Compare the being lexicon, the NRC lexicon, and the African lexicon to analyze positive, negative, and neutral words in lyrics, showing more negative than positive terms.

Load and prepare the fake news dataset, extract the title, compare bias between Trump and Clinton groups across fake, conspiracy, and B.S. categories, analyze punctuation, and introduce sentiment analysis.

Explore document similarity with cosine similarity, the technique behind search engines like Google, applied to a tweets dataset to rank top 15 results using a document-term matrix and query vector.

Learn to perform named entity recognition to locate and classify entities such as persons, organizations, locations, times, and monetary values in text using R and openNLP.

Explore part of speech tagging in R using lexical, rule-based, probabilistic, and deep learning methods (hidden Markov models) to tag tokens in a corpus.

Explore the spacyr wrapper for spaCy in R, enabling non-destructive tokenization, part-of-speech tagging, named-entity recognition, and noun-phrase extraction with multilingual models.

Explore how to load and use SpacyR for natural language processing in R, including German and Spanish language models, tokenization, named-entity recognition, and integration with tidytext for multilingual text analysis.

Explore the UDPipe package for language-agnostic tokenization, tagging, lemmatization, and dependency parsing, plus visualizations, keyword extraction, and topic modeling across languages.

Explore rapid keyword extraction with the UDPipe package using rake, PMI, and text rank to identify keywords, phrases, and word networks, then visualize core occurrences and correlations.

Explore trends in natural language processing, from ubiquitous chat boards and sentiment analysis to advanced text analytics and deep learning approaches in NLP.

Master stylometry with the R style package to analyze writing style through preprocessing, feature extraction, and machine learning, including GUI and visualization of results for forensic and historical texts.

Thank you for joining this course on text analytics and natural language processing with R; the next course on advanced NLP invites you, and I wish you success.