Mitigating shortage of labeled data using clustering-based active learning with diversity exploration

1 code implementation • 6 Jul 2022 • Xuyang Yan, Shabnam Nazmi, Biniam Gebru, Mohd Anwar, Abdollah Homaifar, Mrinmoy Sarkar, Kishor Datta Gupta

In this paper, we proposed a new clustering-based active learning framework, namely Active Learning using a Clustering-based Sampling (ALCS), to address the shortage of labeled data.

Active Learning Clustering

Surveillance of COVID-19 Pandemic using Social Media: A Reddit Study in North Carolina

no code implementations • 7 Jun 2021 • Christopher Whitfield, Yang Liu, Mohd Anwar

We observed that 'mask', 'flu', and 'testing' are the most prevalent named-entities for "Personal Protective Equipment", "symptoms", and "testing" categories, respectively.

named-entity-recognition Named Entity Recognition +1

Music Embedding: A Tool for Incorporating Music Theory into Computational Music Applications

1 code implementation • 24 Apr 2021 • SeyyedPooya HekmatiAthar, Mohd Anwar

Advancements in the digital technologies have enabled researchers to develop a variety of Computational Music applications.