PERSONALIZED INTERNAL SERVICES VIA A PRIVACY-CONSTRAINED CONTENT PROMOTION PLATFORM

Smart overview of the Invention

The patent application describes a method for delivering personalized content within applications while maintaining user privacy. A base machine learning model is deployed in a cloud environment specific to a tenant organization. This model is customized based on user interactions and content categories relevant to the organization. The custom model selects content categories, tailoring content delivery to users without exposing personal details to external content providers.

Background

Traditional methods of personalized content delivery often compromise user privacy by tracking activities across various platforms. Companies typically use cookies and device information to tailor advertisements, leading to privacy concerns. To mitigate this, some organizations avoid using targeted ads in internal tools, ensuring employees can work without intrusive monitoring. This patent proposes a solution that respects privacy while still providing personalized content.

Implementation

The method involves loading a base machine learning model into a private cloud environment, where it is trained to develop a custom model for the tenant organization. This model analyzes user interactions with applications to select relevant content categories. Engagement with these content items is tracked privately by the tenant, cloud provider, and content provider, each maintaining specific privacy controls.

Feedback and Adaptation

User engagement with content items generates feedback that refines the custom model. This feedback loop enhances the personalization process without compromising privacy. The system also supports asset transfers and communications between content providers, cloud services, and tenant organizations, ensuring seamless integration and operation within the privacy-constrained framework.

Advanced Features

The system includes additional features such as location-based preferences and enriched content delivery based on user inputs. Users can opt-in for specific content promotions, allowing further customization of their experience. The patent also outlines a system architecture involving data processors and storage media that execute these methods, highlighting its comprehensive approach to private, personalized content delivery.