MIS372 Lecture Notes - Lecture 10: Natural-Language Processing, Sentiment Analysis, Natural Language
1 Oct 2018
School
Department
Course
Professor
Document Summary
Lecture 10: text mining applications sentiment analysis. How it answers questions: searches for text documents matching the question keywords: Uses stemming and more: applies natural language processing (nlp) techniques on relevant text documents: Data instances or data points or rows: do not have a fixed structure: Text mining: general procedure: to analyse text data: Convert textual (unstructured) data to more structured format. Apply any descriptive analysis method, rule-based method, and/or predictive modelling technique on the text data in the new structured format. It is to break down text into smaller chunks. N words (e. g. phrases with 2 words) These are also called n-grams: stopword removal: Stopwords are words that do not add to the meaning of text. Examples: after, before, the, a/an, of, for, from, in, out. There are standard stopword lists (per language) that can be used as references: stemming: Converts each word into its root form. Remove different variations of the same word from text.
