The authors have provided simple scripts for repeating the experiments. The graphs in Fig. 2 and Fig. 3 can be obtained after a few clicks/commands with the correct settings. Here is a few points I would like to add in addition to the installation notes from the authors: 1. ODBC is required in using Access2MySQL. (downloadable from http://www.mysql.com/downloads/connector/odbc/) 2. There are quite a lot of databases from Medicare.com. Not all of them are involved in the experiment (except table 2). The attribute `NursingHomeName' is found in the most databases about `Nursing Home Compare', e.g., `Nursing Home Compare - About the Nursing Home' (15704 tuples, 15302 distinct values). Note that some scripts refer the source table as `vwnhc_msr'. In order to make the scripts work, it is easier to simply rename the table. The attribute `corp_name' is found in the database `Supplier Directory' (784418 tuples, 21781 distinct values). Note: the experiment in Table 2 will use all tables in the medical database. So, one has to load all databases from Medicare.com to mysql. 3. The scripts, by default, use `wwf' and `medical' as the database names. It is easier for one to use these names for databases in mysql when transferring the data (note: they are not the default database in using Access2MySQL). Otherwise, one may need to edit some of the scripts. The results in my run are all similar to the results shown in the paper. For workability, I ran the experiment on different data and different parameters. (i) different dataset: I used the attribute `species' from the table `species' in the wwf database for the first experiment (Fig 2/3). The procedures are easy. First, extract all distinct values in the attribute to a separate database. Then, create a setting (.con) file and modify the scripts accordingly. Finally, the graphs on the dataset can be obtained. (ii) different parameters: it is more difficult than to use a different dataset, as this involves modifications of codes in different parts. The authors have documented the way to modify some parameters, but not all (like number of typos), in the XML.