Interface Type Classification via Object Hierarchy Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automatically classifying arbitrary interfaces from various interface providers into their specific types is challenging due to variations in source code, making it difficult for software applications to interact with them effectively.
Innovation Solution
A system that determines the type of an interface by generating a category dictionary from object hierarchy paths, using machine learning algorithms like recurrent neural networks and convolutional neural networks to classify interfaces based on feature vectors, enabling specific interactions with identified interface types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional interface classification methods are used, then implementation is simple, but classification accuracy deteriorates due to source code variations across different interface providers
Solution Approach 1:
The patent introduces an intermediary processing layer that extracts object hierarchy paths from interface source code and transforms them into standardized feature vectors. This intermediary representation layer mediates between the raw source code variations and the classification algorithm, enabling accurate classification across different interface providers while maintaining system modularity and manageability
Solution Approach 2:
The patent transforms interface source code into a different parameter representation - specifically, converting code structure into object hierarchy paths and then into numerical feature vectors. This parameter transformation enables the classification system to work with standardized numerical data rather than raw code, improving accuracy while keeping the system complexity manageable through automated transformation processes
2Productivity
If manual interface classification is performed, then accuracy can be maintained, but processing speed deteriorates and automation is reduced
Solution Approach 1:
The system implements self-service classification by automatically extracting object hierarchy paths from interface source code and generating feature vectors without human intervention. The machine learning model autonomously performs classification based on the extracted features, achieving both high processing speed and accuracy while enabling full automation of interface interaction workflows
Solution Approach 2:
The patent replaces manual mechanical classification processes with an automated machine learning system. The system uses algorithmic processing to substitute human analysts, achieving faster processing speeds while maintaining or improving accuracy through consistent application of trained classification models across all interface classification tasks
3Reliability
If interface-specific operations are implemented without accurate classification, then automation can proceed, but reliability deteriorates due to incorrect interface type identification
Solution Approach 1:
The patent performs preliminary classification of interfaces into specific types before executing interface-specific operations. By pre-identifying the interface type using the trained machine learning model, the system ensures that subsequent operations are appropriate for the specific interface type, thereby improving reliability while keeping the overall system complexity manageable through a clear two-stage process
Solution Approach 2:
The patent segments the interface interaction process into distinct stages: feature extraction, classification, and type-specific operation execution. This segmentation allows each component to be optimized independently, improving overall reliability while managing complexity through modular design where the classification system is a self-contained module
Data Source
AI summary
A first path associated with a first interface of a first provider is determined, with the first interface being of a first type. A second path associated with a second interface of the first provider is determined, with the second interface being of a second type. The first path is determined to be a mismatch to the second path. A third path corresponding to an additional interface of the first type is obtained from a second provider. A dictionary is generated based on the third path matching the first path. Vectors corresponding to the first and second types are generated based on the dictionary. A machine learning algorithm is trained based on the vectors with the first and second types as a ground truth values.