.keyword fields are automatically created for analyzed string fields. The problem with analyzed strings was (in the older versions) that if you visualize them then each word gets counted once. So for example if you have a field saying "not good" and you have 3 matches then in the visualization you would find 3 matches for "not" and 3 for "good". To solve this problem one had to map the field in elasticsearch as not analyzed before indexing.
To solve this problem the .keyword field is created automatically which is a not analyzed string and it is the only field which can be used visualizations (in searches the normal field which is analyzed will be used).
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.