Abstract
This disclosure describes techniques that enable group-level actions to be performed on entities identified as belonging to a group. First, the common characteristics of a given group of entities are automatically identified. The unique characteristics of a group that differentiate the group from the general population are then identified. An identified group of entities is expanded into the general population of entities. The described techniques find application in domains such as trend identification, and can be used for advertising, route planning, e-commerce, social media, policy enforcement in user-generated content (UGC), administering UGC appeals, etc. The techniques enable fast understanding of large datasets at scale and obviate the need for SQL querying skills to understand data. Non-technical users are presented with human-understandable interpretations of large datasets, enabling them to focus on business requirements.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Vijayakumar, Karthik and Bhakat, Sukalyan, "Deriving Human-Interpretable Insights from Tabular Data", Technical Disclosure Commons, (April 24, 2024)
https://www.tdcommons.org/dpubs_series/6902