Academic Dishonesty or Academic Integrity? Using Natural Language Processing (NLP) Techniques to Investigate Positive Integrity in Academic Integrity Research

Abstract Is academic integrity research presented from a positive standpoint? This paper uses Natural Language Processing (NLP) techniques to explore data set of 8,507 papers published between 1904 and 2019. Two main are used linguistically examine titles: (1) bigram (word pair) analysis (2) sentiment analysis. The sees the three bigrams in titles as being “academic integrity” (2.38%), dishonesty” (2.06%) “plagiarism detection” (1.05%). When only highly cited considered, negative dominate bigrams. For example, 100 most all time times more likely have included their than integrity”. Similarly, outperforming papers. history is seen place field at disadvantage due portrayals integrity. Despite this, shows that change towards possible. by ten prolific researchers found use terminology cases not. suggests an approach for emerging model themselves after.