Drillbit Plagiarism Checker: Filters and exclusions || Hindi || 2024

The core takeaway is that a “plagiarism checker” workflow can be made compliant and more accurate by applying targeted exclusions—specifically for quoted material, references/bibliography, generic terms, and common knowledge—then generating a similarity report that stays under a required threshold (often “below 10”). The process matters because universities and UGC-style rules typically treat similarity differently from plagiarism: similarity can include legitimate overlap, so filtering out non-substantive sections helps committees judge the real risk.

After aligning with UGC and university regulations (which may vary slightly by institution), the workflow focuses on what can be excluded. Exclusions can include quoted content, reference lists/bibliography, table of contents, acknowledgements, and generic symbols or generic terms. The transcript also stresses that “common knowledge” and repeated technical phrases may be excluded only after checking with the supervisor/guide and the relevant committee—so the exclusions are defensible when the report is shared.

Practically, the checker is used by logging in, creating a folder (project) and then uploading a document for analysis. During folder creation, filter options become available and persist for that project. The example walkthrough uses a demo project and a sample paper, then demonstrates enabling specific exclusions such as “exclude bibliography.” It also shows how adding exclusion phrases (e.g., recurring terms like ECG, or database names/phrases inserted in the text) affects what gets filtered.

The tool then asks which sources to compare against, grouped into categories such as general public publishing, publisher/internet sources, and web repositories (with multiple source categories selectable). Similarity is reported as a similarity word/index, with a stated requirement to keep it under 10. The transcript draws a distinction between “plagiarism” and “similarity index,” noting that they are not identical and that the threshold is a policy requirement rather than a direct measure of wrongdoing.

Once the report is generated, the user can inspect match sources and—crucially—verify that excluded items appear as excluded in the report. The settings section shows whether exclusions like code/quotes, bibliography, and similarity-word filtering were enabled. There’s also a “duplicate phrases” control: the user can set a range (for example, selecting from 3 up to 14 words) so repeated phrases can be excluded, which is useful when a term or phrase keeps pushing similarity upward.

The transcript warns that exclusions must be applied correctly at the project/folder level. If the same file is re-submitted without updating the folder’s exclusion settings, the filtered phrases may not be removed in the next run. After generating the report, the user should include only the relevant pages/sections when submitting to the committee, and the tool can provide a plagiarism certificate printout with options to select the document type (thesis/dissertation/article). The overall message: keep similarity below the required limit by using guideline-approved filters, but always coordinate exclusions with the guide and ensure the report transparently reflects what was excluded.