Adherence Model for Granular Data Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in accurately monitoring and controlling adherence parameters for data items with time-varying values, especially when direct observations are not available and data items are fungible, leading to rough estimates and limited information processing.
Innovation Solution
The development of an adherence model that interpolates data values using contextual information, builds tracking data structures, and applies customizable mappings to calculate adherence, allowing for fine-grained analytics and alerts when adherence thresholds are met.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If systems estimate data item values at high level granularity with limited information, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent introduces an adherence model as an intermediary component that sits between the limited observational data and the required adherence determination. This model uses contextual information and event data as intermediate representations to infer adherence status without requiring direct observation of all data item values, thus resolving the contradiction between system complexity and measurement precision.
Solution Approach 2:
The patent replaces direct mechanical observation of data item values with an informational/inferential system. Instead of directly measuring all data values, the system substitutes this with a computational model that processes contextual information and event data to determine adherence, achieving high precision without proportionally increasing system complexity.
2Measurement precision
If systems track adherence parameters in real-time with fine-grained granularity, then measurement precision is improved, but loss of information increases due to data processing requirements
Solution Approach 1:
The patent extracts only the essential information needed for adherence determination from the full dataset. By identifying and processing only the critical event data and contextual information relevant to adherence events, the system achieves fine-grained adherence tracking while minimizing the processing of unnecessary data, thus reducing information loss.
Solution Approach 2:
The patent performs preliminary processing of data by pre-identifying adherence events and contextual information before full analysis. This preliminary action filters and prepares only the necessary data for adherence determination, reducing the overall data processing burden while maintaining fine-grained tracking capability.
3Adaptability or versatility
If systems monitor time-varying data items without direct observation, then adaptability is improved, but measurement precision deteriorates
Solution Approach 1:
The patent implements a dynamic adherence model that adapts to time-varying data characteristics. The model processes events and contextual information as they occur, allowing it to handle changing data patterns and conditions in real-time while maintaining accurate adherence determination through continuous updates based on new information.
Solution Approach 2:
The patent incorporates feedback mechanisms where the adherence model continuously refines its understanding of data item values based on observed events and contextual information. This feedback loop allows the system to adapt to time-varying conditions while improving measurement precision by correcting and updating adherence determinations as new data becomes available.
Data Source
AI summary
Exemplary embodiments relate to systems for building a model of changes to data items when information the data items is limited or not directly observed. Exemplary embodiments allow properties of the data items to be inferred using a single data structure and creates a highly granular log of changes to the data item. Using this data structure, the time-varying nature of changes to the data item can be determined. The data structure may be used to identify characteristics associated with a regularly-performed action, to examine how adherence to the action affects a system, and to identify outcomes of non-adherence. Fungible data items may be mapped to a remediable condition or remedy class. This may be accomplished by automatically deriving conditions and remedial information from available information, matching the conditions to remedial classes or types via a customizable mapping, and then calculating adherence for the condition on the available information.