LectureBank Dataset
Branch: master
Clone or download
Alex-Fabbri Merge pull request #1 from Mistobaan/patch-1
add link to the arxiv paper
Latest commit d9d4f21 Feb 14, 2019
Type Name Latest commit message Commit time
Failed to load latest commit information.
imgs add image Nov 12, 2018
208topics.csv add data Nov 11, 2018
download_all.py Update download_all.py Nov 12, 2018
lecturebank.tsv add data Nov 11, 2018
prerequisite_annotation.csv add data Nov 11, 2018
vocabulary.txt add data Nov 11, 2018


LectureBank: a corpus for NLP Education and Prerequisite Chain Learning

This is the github page for our paper What Should I Learn First: Introducing LectureBank for NLP Education and Prerequisite Chain Learning in the proceedings AAAI 2019.

Code for replicating the results will be uploaded soon. Stay tuned.

An example of prerequisite relations from lecture slides depicted as a directed graph

The list of descriptions can be found in the following:

LectureBank Dataset

LectureBank Dataset is a manually-collected dataset of lecture slides. We collected 1352 online lecture files from 60 courses covering 5 different domains, including Natural Language Processing (nlp), Machine Learning (ml), Artificial Intelligence (ai), Deep Learning (dl) and Information Retrieval (ir). In addition, we release the corresponding annotations for each slide file to the taxonomy described below. We also provide an additional vocabulary list of size 1221 extracted from the corpus.


Each line identifies a lecture file. Format:

(ID, Title, URL, Topic_ID, Year, Author, Domain, Venue)

  • ID: Id of each line.
  • Title: File tile.
  • URL: Online URL.
  • Topic_ID: Classified taxonomy Topic ID, referring topics from taxonomy.tsv.
  • Year: Year of the course.
  • Author: The author name(s).
  • Domain: The domain (nlp, ir, dl, ml, ai).
  • Venue: Name of the university, or GitHub.


The scripts of downloading the resources from the urls of lecturebank.tsv. After running the scripts, all the resources will be downloaded into data_lecturebank/ folder (change the base_path if you want), organized by the Domain (for example, nlp, ir). The code is in python3, and you will need to install wget to run it. Run with: python3 download_all.py. It may take an hour or less for the resources to be downloaded.

Due to the change of the links by the owner, some of the URLs may have broken.


Contains taxonomy topics and corresponding IDs referred by lecturebank.tsv.


Contains the 208 topics which we annotated, format:

(ID, Topic, Wiki_Page_URL)


Contains the prerequisite chain annotation for each possible pair from the 208 topics. Format:

(Source_Topic_ID, Target_Topic_ID, If_prerequisite)


Contains 1221 vocabulary terms combined from taxonomy, 208 topics and terms extracted from LectureBank.