NLP for Scientific Papers is a project focused on analysis of scientific articles using natural language processing techniques. The goal is to extract key insights from research papers, explore word frequency patterns, and apply LDA modeling to identify underlying topics, with visualizations of the results. The project currently supports both unigrams and bigrams analysis.
Languages and packages used:
- R (pdftools, tm, textstem, dplyr, tidytext, ggplot2, topicmodels, tidyverse, textmineR, Matrix, slam, wordcloud, RColorBrewer, wordcloud2, widyr, ggraph, igraph, tibble)
Repository structure:
- loading_files.R - data loading from PDF files
- main.R - NPL model implementation
- preprocessing.R - data preprocessing
- LDA.R - LDA model, with visualization
- tf-idf.R - visualization of high tf-idf words
- README.md