Best代写-最专业靠谱代写IT | CS | 留学生作业 | 编程代写Java | Python |C/C++ | PHP | Matlab | Assignment Project Homework代写

机器学习代写|Patronizing and Condescending Language Detection

机器学习代写|Patronizing and Condescending Language Detection


This task is based on the paper Don’t Patronize Me! An annotated Dataset with Patronizing and Condescending Language Towards Vulnerable Communities (Perez-Almendros et al., 2020).

The aim of this task is to identify PCL, and to categorize the linguistic techniques used to express it, specifically in news stories, when referring to vulnerable communities.

What is PCL?

Somebody is patronizing or condescending when their language denotes a superior attitude towards others, talks down to them, or describes them or their situation in a charitable way, raising a feeling of pity and compassion.

Patronizing and Condescending Language (PCL) is often involuntary and unconscious, and the authors using such language are usually trying to help communities in need by e.g., raising awareness, moving the audience to action or standing for the rights of the under-represented. On the other hand, due to its subtlety, subjectivity and the (generally) good intentions behind its use, the audience is often unaware of this diminishing treatment. But PCL can potentially be very harmful, as it feeds stereotypes, routinizes discrimination and drives to greater exclusion.

PCL detection is difficult both for humans and NLP systems, due to its subtle nature, its subjectivity and the fair amount of world knowledge and commonsense reasoning required to understand this kind of language. With this task, we expect to push the boundaries of this new challenge in the NLP community.

About the task:

The minimum linguistic unit in this task is the paragraph (i.e., sentences in context). Paragraphs are extracted from news articles from a range of outlets. These paragraphs may contain one or more instances of PCL targeting a set of predefined vulnerable communities. This challenge is divided into two subtasks.

· Subtask 1: Binary classification.

Given a paragraph, a system must predict whether or not it contains any form of PCL.

  • In the last week of April, 1,100 migrants died or went missing off Libya in nine separate incidents.– NO PCL
  • People across Australia ordered pizzas to be delivered on Saturday night, with the ample leftovers donated to local homeless shelters.– PCL