HKUST-KnowComp/atomic-conceptualization
Code and data for the paper Acquiring and Modelling Abstract Commonsense Knowledge via Conceptualization
This project helps researchers and developers working with natural language understand the abstract common-sense knowledge behind everyday events. It takes text describing events (like "PersonX wants to leave") and identifies the underlying abstract concepts and their relationships (e.g., "leave" -> "exit"). The primary output is a structured representation of these conceptualizations and abstract common-sense triples.
No commits in the last 6 months.
Use this if you need to extract and model high-level, abstract common-sense knowledge from event descriptions for tasks like improving AI understanding or generating more human-like responses.
Not ideal if you're looking for a direct, user-friendly application for general text analysis rather than a research framework for abstract common-sense knowledge.
Stars
23
Forks
1
Language
Python
License
—
Category
Last pushed
Nov 21, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/HKUST-KnowComp/atomic-conceptualization"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ymcui/cmrc2018
A Span-Extraction Dataset for Chinese Machine Reading Comprehension (CMRC 2018)
princeton-nlp/DensePhrases
[ACL 2021] Learning Dense Representations of Phrases at Scale; EMNLP'2021: Phrase Retrieval...
thunlp/MultiRD
Code and data of the AAAI-20 paper "Multi-channel Reverse Dictionary Model"
IndexFziQ/KMRC-Papers
A list of recent papers regarding knowledge-based machine reading comprehension.
danqi/rc-cnn-dailymail
CNN/Daily Mail Reading Comprehension Task