Spark GraphX library to detect causalities across time related events
@aamend / (1)
The rooster crows immediately before sunrise, the rooster causes the sun to rise.
Using GraphX to detect possible causes and effects between time related events. We observe a true causation signal by generating random correlations over same events at different time and back propagate these scores to their most connected events. Finally, we extract the most probable causes and effects together with a score of aggressiveness (how likely an event could explain downstream effects) and sensitivity (how likely an event results from an upstream cause).
Include this package in your Spark Applications using:
spark-shell, pyspark, or spark-submit
> $SPARK_HOME/bin/spark-shell --packages com.aamend.spark:pathogen:1.0
In your sbt build file, add:
libraryDependencies += "com.aamend.spark" % "pathogen" % "1.0"
MavenIn your pom.xml, add:
<dependencies> <!-- list of dependencies --> <dependency> <groupId>com.aamend.spark</groupId> <artifactId>pathogen</artifactId> <version>1.0</version> </dependency> </dependencies>
Version: 1.0 ( a60da6 | zip | jar ) / Date: 2017-07-14 / License: Apache-2.0 / Scala version: 2.11