Knowledge Graph Assessment for Online Platform Intervention Impact
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods are insufficient to reliably evaluate the impact of interventions in interactive computing environments, making it difficult to determine proper actions for improving performance without significant experience.
Innovation Solution
An assessment model evaluates the interactive computing environment using a knowledge graph and relational hops, computing ranked target entities and paths to inform modifications that enhance user interactions and performance metrics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing methods are used to evaluate intervention impacts, then the process is simple, but the evaluation reliability is insufficient
Solution Approach 1:
The patent introduces a knowledge graph as an intermediary structure that mediates between the intervention and the performance evaluation. The knowledge graph contains entities representing performance metrics, user behaviors, and content elements, along with relationships between them. This intermediary structure enables reliable evaluation by providing a formalized framework for tracking causal relationships, rather than relying on simple empirical observation alone.
Solution Approach 2:
The patent replaces manual, experience-based evaluation methods with an automated computational system. The assessment model uses algorithms to traverse the knowledge graph, compute impact scores, and generate recommendations automatically. This substitution of mechanical (manual) processes with computational mechanisms significantly improves evaluation reliability while maintaining systematic approach.
2Measurement precision
If experiential knowledge base is used, then the system is easy to implement, but the measurement precision is insufficient
Solution Approach 1:
The patent segments the evaluation system into distinct components: a knowledge graph module for storing structured information, an assessment model module for computation, and a recommendation module for output. The knowledge graph itself is segmented into entities (performance metrics, user behaviors, content) and relationships (causal links). This segmentation enables precise measurement by allowing detailed tracking of specific causal paths without requiring the entire system to be overly complex.
Solution Approach 2:
The patent transitions from flat, two-dimensional empirical data collection to a multi-dimensional knowledge graph structure. The knowledge graph adds temporal, causal, and hierarchical dimensions to the data organization. Entities are linked through multiple relationship types (causes, effects, intermediate factors), enabling precise measurement of intervention impacts by analyzing data across multiple dimensions simultaneously.
3Reliability
If comprehensive impact evaluation is performed, then the performance improvement is accurate, but the time consumption increases
Solution Approach 1:
The patent performs preliminary action by pre-building and pre-populating the knowledge graph with relationships between performance metrics, user behaviors, and content elements before actual intervention evaluation occurs. Causal relationships are established in advance through knowledge engineering. When an intervention is evaluated, the system can quickly traverse the pre-structured knowledge graph rather than analyzing raw data from scratch, significantly reducing evaluation time while maintaining accuracy.
Solution Approach 2:
The patent extracts only the relevant portions of the knowledge graph necessary for evaluating a specific intervention. Instead of analyzing all possible relationships in the comprehensive knowledge graph, the system identifies and processes only the causal paths and entities relevant to the current intervention being evaluated. This extraction approach maintains evaluation accuracy for the specific question at hand while reducing time consumption by avoiding unnecessary analysis of unrelated data.
Data Source
AI summary
A method includes accessing a subject entity and a subject relation of a focal platform and accessing a knowledge graph representative of control performance data. Further, the method includes computing a set of ranked target entities that cause the subject entity based on the subject relation or are an effect of the subject entity based on the subject relation. Computing the set of ranked target entities is performed using relational hops from the subject entity within the knowledge graph performed using the subject relation and reward functions. The method also includes transmitting the set of ranked target entities to the focal platform. The set of ranked target entities is usable for modifying a user interface of an interactive computing environment provided by the focal platform.