English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Joint Reasoning for Multi-Faceted Commonsense Knowledge

Chalier, Y., Razniewski, S., & Weikum, G. (2020). Joint Reasoning for Multi-Faceted Commonsense Knowledge. Retrieved from http://arxiv.org/abs/2001.04170.

Item is

Files

show Files
hide Files
:
arXiv:2001.04170.pdf (Preprint), 1005KB
Name:
arXiv:2001.04170.pdf
Description:
File downloaded from arXiv at 2020-01-20 11:41
OA-Status:
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
-
Copyright Info:
-

Locators

show

Creators

show
hide
 Creators:
Chalier, Yohan1, Author
Razniewski, Simon2, Author           
Weikum, Gerhard2, Author           
Affiliations:
1External Organizations, ou_persistent22              
2Databases and Information Systems, MPI for Informatics, Max Planck Society, ou_24018              

Content

show
hide
Free keywords: Computer Science, Computation and Language, cs.CL,Computer Science, Artificial Intelligence, cs.AI,Computer Science, Information Retrieval, cs.IR
 Abstract: Commonsense knowledge (CSK) supports a variety of AI applications, from visual understanding to chatbots. Prior works on acquiring CSK, such as ConceptNet, have compiled statements that associate concepts, like everyday objects or activities, with properties that hold for most or some instances of the concept. Each concept is treated in isolation from other concepts, and the only quantitative measure (or ranking) of properties is a confidence score that the statement is valid. This paper aims to overcome these limitations by introducing a multi-faceted model of CSK statements and methods for joint reasoning over sets of inter-related statements. Our model captures four different dimensions of CSK statements: plausibility, typicality, remarkability and salience, with scoring and ranking along each dimension. For example, hyenas drinking water is typical but not salient, whereas hyenas eating carcasses is salient. For reasoning and ranking, we develop a method with soft constraints, to couple the inference over concepts that are related in in a taxonomic hierarchy. The reasoning is cast into an integer linear programming (ILP), and we leverage the theory of reduction costs of a relaxed LP to compute informative rankings. This methodology is applied to several large CSK collections. Our evaluation shows that we can consolidate these inputs into much cleaner and more expressive knowledge. Results are available at https://dice.mpi-inf.mpg.de.

Details

show
hide
Language(s): eng - English
 Dates: 2020-01-132020
 Publication Status: Published online
 Pages: 11 p.
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: arXiv: 2001.04170
URI: http://arxiv.org/abs/2001.04170
BibTex Citekey: Chalier_arXiv2001.04170
 Degree: -

Event

show

Legal Case

show

Project information

show

Source

show