About this resource
<p>It has 4 central portions, which show Bangla text. One contains only Bangla text(12179-unsuspicious, 7822-suspicious). Another one contains Bangla and English mixed(12725-unsuspicious, 7219-suspicious). Another one contains politically suspicious content(167-unsuspicious, 132-suspicious). Lastly, another contains the <a class="user-mention" href="https://www.kaggle.com/name" rel="noopener">@name</a> mentioned comment(6145-suspicious, 53855-unsuspicious). Finally, a CSV file contains all categorical Bangla Data. It contains 1,00,100+ data.</p> <p><strong>COLLECTION METHODOLOGY</strong></p> <p>Suspicious tweets- <a href="https://www.kaggle.com/datasets/syedabbasraza/suspicious-tweets" rel="noopener">https://www.kaggle.com/datasets/syedabbasraza/suspicious-tweets</a> Suspicious Tweets - <a href="https://www.kaggle.com/datasets/munkialbright/suspicious-tweets" rel="noopener">https://www.kaggle.com/datasets/munkialbright/suspicious-tweets</a> Suspicious Communication on Social Platforms - <a href="https://www.kaggle.com/datasets/syedabbasraza/suspicious-communication-on-social-platforms" rel="noopener">https://www.kaggle.com/datasets/syedabbasraza/suspicious-communication-on-social-platforms</a></p> <p>Others are collected manually from Facebook comments. After collecting the Bangla comments, check dataset comment was understandable or not. Then step by step, each Excel file is converted into a datagram. then change the column name to the desired one('Detect' and 'Bangla Text'). I also drop some columns if needed. The files are saved in an Excel file because the CSV file can not contain Bangla text appropriately.</p> <p>The 5 XLSX file are "suspicious_content(bangla)", "suspicious_content(bangla + english)", "suspicious_content(political)", "suspicious_content(including mention)" and "suspicious_content(all)". All the Excel files have only two columns, 'Detect' and 'Bangla Text'.</p> <p>I used a script to convert the English text into Bangla in Kaggle. You will be able to see the dataset creation process in this link: <a href="https://www.kaggle.com/code/meherunnesashraboni/suspicious" rel="noopener">https://www.kaggle.com/code/meherunnesashraboni/suspicious</a></p> <p> </p> <p> </p>
Files and access
Resources are read online through DARE. Direct file links are not published; access is checked against DSpace when a document is opened. Authorised librarian workflows are unaffected.
This record contains bibliographic metadata, but no files are currently attached.
