Empirical Study: What is the Best N-Gram Graphical Indexing Technique

Recent research shows that indexing techniques, such as n-gram models, are useful at a wide variety of software engineering tasks, e.g., data extraction, document indexing, etc. However, these models require some numerous parameters. Moreover, the different ways one can extract data essentially yield different models. In this paper, we focus on n-gram models and evaluate how the use different set of methods and n values impact the predicting ability of these models. Thus, we compare the use of multiple techniques and sets of different parameters (n values character gram and word gram) with the aim of identifying the most appropriate indexing technique. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.