Bangla Text Dataset and Exploratory Analysis for Online Harassment Detection
Harassment
Scarcity
Exploratory analysis
Exploratory research
DOI:
10.48550/arxiv.2102.02478
Publication Date:
2021-01-01
AUTHORS (6)
ABSTRACT
Being the seventh most spoken language in world, use of Bangla online has increased recent times. Hence, it become very important to analyze text data maintain a safe and harassment-free place. The that been made accessible this article gathered marked from comments people public posts by celebrities, government officials, athletes on Facebook. total amount collected is 44001. dataset compiled with aim developing ability machines differentiate whether comment bully expression or not help Natural Language Processing what extent improper if an inappropriate comment. are labeled different categories harassment. Exploratory analysis perspectives also included paper have detailed overview. Due scarcity collection categorized Bengali comments, can significant role for research detecting words, identifying bullies, etc. publicly available at https://data.mendeley.com/datasets/9xjx8twk8p.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES ()
CITATIONS ()
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....