Data Aggregation for dblp Monitoring

VM Arguments

For the XML parser, the entity expansion limit given by the VM is not enough, so you need to override it in the VM arguments. A reasonable choice would be so set no limit at all, thus -DentityExpansionLimit=0.

It is also advised to provide the program with enough memory, e.g. -Xmx9g or more.

Dependencies

This project has some dependencies on other projects that are dblp-internal and not publicly available.
That is, to run the application, you need to have the following projects on the build path:

• apache-log-analysis
• dblp-citations
• dblp-commons
• mmdb
• streams

Ask dblp developers for access.

Setting up the Database

To create the database with the parameters and input files specified, AggregationApplication needs to be executed.

A PostgreSQL database is used to store the aggregated data.

The database parameters have to be specified in database.properties file. The database.properties.default file serves as a template.

host=hostname
port=port
dbname=dbname
user=user
password=password

Input Data

The following files are needed for the program to build the database:

• bht.xml
• dblp.xml
• hdblp.xml
• streams.xml
• dblp.dtd
• folder containing auxiliary data (mainly citation info) mapped from OAG/MAG to dblp

The paths to these files have to be specified in setup.properties file. The setup.properties.default file serves as a template.

xml.bht=path/to/bht.xml
xml.dblp=path/to/dblp.xml
xml.hdblp=path/to/hdblp.xml
xml.streams=path/to/streams.xml

dtd.dblp=path/to/dblp.dtd

path.oaga=path/to/oag-mag-aux-root-dir

output.basedir=path/to/output/basedir

Take care that all files containing dblp data are from the same date.