Purpose: Flattering emails are crucial in tempting authors to submit papers to predatory journals. Although there is ample literature regarding the questionable practices of predatory journals, the nature and detection of spam emails need more attention. Current research provides insight into fallacious calls for papers from potential predatory journals and develops a toolkit in this regard.
Methods: In this study, we analyzed three datasets of calls for papers from potential predatory journals and legitimate journals using a text mining approach and R programming language.
Results: Overall, most potential predatory journals use similar language and templates in their calls for papers. Importantly, these journals praise themselves in glorious terms involving positive words that may be rarely seen in emails from legitimate journals. Based on these findings, we developed a lexicon for detecting unsolicited calls for papers from potential predatory journals.
Conclusion: We conclude that calls for papers from potential predatory journals and legitimate journals are different, and it can help to distinguish them. By providing an educational plan and easily usable tools, we can deal with predatory journals better than previously.
Keywords: Academic ethics; Calls for papers; Data science; Journal publishing; Predatory journal; Sentiment analysis.
©2023 The Authors.