Abstract: Embodiments leverage local data available from an on-premises Extract, Transfer, and Load (ETL) job, in order to efficiently perform remote processing (e.g., as implemented on the cloud). Connectivity data (e.g., target setup) and ETL logic (e.g., configuring data flattening, pivot transform, and/or data quality transform) is stored locally in a non-transitory computer readable storage medium. In response to receiving data transformed on-premises, the transformed data, connectivity data, and ETL logic are forwarded to a remote location for processing. Some embodiments may also forward the transformed data on to its original target on-premises (e.g., via a local database loader). Particular embodiments may provide hidden, Representational State Transfer (REST)-based loader(s) that duplicate output of the local ETL job. Embodiments conserve developer effort by allowing preparatory local ETL data that is already available on-premises, to be read and used for processing in a remote cloud intelligence system.

Abstract: A processing tool is described that can evaluate a plurality of fields in a plurality of records and assign a scoring measure to each field based on the evaluation of each field. The scoring measures can provide an indication as to the uniqueness in content within each field, which can assist with determining one or more fields that are most useful for performing a search to identify similar records.

Abstract: A processing tool is described that can evaluate a plurality of fields in a plurality of records and assign a scoring measure to each field based on the evaluation of each field. The scoring measures can provide an indication as to the uniqueness in content within each field, which can assist with determining one or more fields that are most useful for performing a search to identify similar records.