Machine-Learning Field Mapping for Dynamic Web Submissions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automatic submission methods do not utilize machine learning models to determine suggested associations for fields on webpages, leading to repetitive and error-prone manual submissions, which waste time and resources.
Innovation Solution
A system utilizing machine learning models to determine suggested associations between column data in tabular data and fields on webpages by comparing data types and field names, allowing for automated and dynamic submission processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If machine learning models are used to determine suggested associations for fields on webpages, then submission accuracy and automation capability are improved, but system complexity increases
Solution Approach 1:
The patent introduces machine learning models as intermediary components that mediate between tabular data and webpage fields. These models process column data and field information to generate suggested associations, acting as a bridge that enables accurate automated submissions without requiring direct complex mappings between data sources and targets.
Solution Approach 2:
The patent replaces manual mechanical mapping processes with automated machine learning-based association determination. Instead of manually configuring field mappings or using rigid rules, the system uses ML models to dynamically determine associations based on data types and field names, substituting mechanical configuration with intelligent automation.
2Ease of operation
If manual submissions are performed repeatedly, then flexibility and control are maintained, but time consumption and error rates increase
Solution Approach 1:
The system enables self-service automation where the machine learning models automatically determine field associations and generate submissions without requiring repeated manual user intervention. The system serves itself by autonomously processing tabular data, determining mappings, and executing submissions, thereby eliminating time-consuming repetitive manual tasks while maintaining operational effectiveness.
3Device complexity
If automatic submission methods are used without machine learning, then system simplicity is maintained, but association accuracy and adaptability to data changes deteriorate
Solution Approach 1:
The patent employs machine learning models that dynamically adjust association parameters based on input data characteristics. The models analyze data types, field names, and contextual information to determine optimal associations, allowing the system to adapt to changing data structures and maintain high association accuracy without requiring complex pre-configured rules.
Data Source
AI summary
Disclosed embodiments may include a method for dynamically automating submissions. The method may include receiving tabular data and one or more associations of columns from the tabular data to fields on a webpage for one or more submissions, and extracting column data from the tabular data, determining, using a first machine learning model, one or more suggested associations for the fields on the webpage based on data types for the fields, and determining, using a second machine learning model, one or more suggested associations for the fields based on column names of the extracted column data. In response to determining one or more suggested associations, sending to a user device the one or more suggested associations, receiving, from the user device, accepted suggested associations; and transmitting one or more submissions for each row in the tabular data using the one or more suggested associations and the accepted suggested associations.