Embeddings and topic vectors for MOOC lectures dataset
Published: 6 December 2019| Version 1 | DOI: 10.17632/xknjp8pxbj.1
Contributors:
Zenun Kastrati, , Description
This dataset is comprised of word embeddings and document topic distribution vectors generated from transcripts of 12032 video lectures from 200 courses that were collected from Coursera learning platform. Two well-known natural language processing techniques, namely Word2Vec and Latent Dirichlet Allocation (LDA) implemented in the Gensim package in Python are used to generate word embeddings and topic vectors, respectively.
Files
Institutions
Linneuniversitet
Categories
Natural Language Processing, Machine Learning, e-Learning