1-In that case you need to encode the data in that language format , ex if the data is in japanease then u need to encode in UTF-8 

REATE OR REPLACE TEMP VIEW japanese_data

AS SELECT * FROM

csv.`path/to/japanese_data.csv`

OPTIONS ('encoding'='UTF-8')

also you can use various libraries and tools for natural language processing (NLP) in Databricks.