Effective Filters and Linear Time Verification for Tree Similarity Joins

A Link is not Enough – Reproducibility of Data

“…In this section, we showcase how missing details in the description of the data preparation process may affect the experimental results. We motivate the discussion with our experience gained while working on similarity join algorithms for sets [44] and trees [38]. As part of the experimental evaluation in these works, we also compared to previous solutions.…”

“…DBLP [6] stores bibliographic data in XML format and includes, among others, authors, titles, and venues of computer science publications. Due to its availability and intuitiveness, the DBLP dataset has been used in many works for experimental purposes, e.g., as a collection of sets [44, 45], as a collection of trees [37, 38, 46], as a large hierarchical document [34, 40], and as a coauthor network graph [42, 49]. In this section, we show the impact of differences in the data preparation process that converts raw DBLP XML data into the desired input format.…”

“…We now take a look at a specific example in which we convert the DBLP dataset into trees using two different approaches and perform a tree similarity join [38]. Conversion A is used in our data preparation process [38] and Conversion B is used by Tang et al [46]1. Conversion A is illustrated in Figure 1.…”

“…We now go back to our use case, and review the data reproducibility levels of five related works on tree similarity joins [38] and five works on set similarity joins [44].…”

“…Two recent research experiences [38, 44] required us to compare to a number of related works. As detailed in Section 7, in most cases the data was not even replicable.…”