I am starting to play with ML, i have created a job to look for rare user name to soure country. Many people login from the UK, the issue is ingesting a new company leads to high scores for the new user names, organizations.name is populated for all data
Job is Rare user.name, partition source.geo.country_name and user.name + source.geo.county_name as the influencers
Is there anything within the job i can set to deal with this?
Beleive i have resolved my issue with the filters. Although when cloning the job it fails on the validation, just says bad. Ive had to create the ML job with a small amount of data, open, add the filter then expand its dataset.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.