Tabular Data Processing via Text Conversion and ML Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques fail to provide a suitable method for accurately and efficiently processing data from various computing systems for reporting and analysis due to differences in data formatting, categories, and recipients.
Innovation Solution
A computer-implemented method using machine learning models to convert tabular data into a text-based language, perform pre-processing based on user inputs and data characteristics, and apply specific models for customized processing and visualization, including generating reports and dashboards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional data processing methods are used, then data from different systems can be processed, but data formatting differences and category inconsistencies cause problems with data sharing, analysis, and reporting
Solution Approach 1:
The patent introduces an intermediary processing layer that converts data from various computing systems into a standardized format. This intermediary layer includes data conversion components that translate different data formats, categories, and structures into a common representation, enabling seamless data sharing and analysis across systems without requiring complex custom integration for each system pair.
Solution Approach 2:
The patent creates a universal data processing framework that handles multiple data types, formats, and sources through a single standardized interface. The system is designed to process tabular data from diverse computing systems (human resources, customer management, accounting) using the same conversion and analysis mechanisms, eliminating the need for separate processing pipelines for each data source.
2Measurement precision
If customized processing is implemented for different recipients and contexts, then data accuracy and relevance improve, but processing time and system complexity increase
Solution Approach 1:
The patent applies preliminary action by pre-processing and standardizing data into a universal format before analysis. The data conversion layer prepares data in advance by normalizing formats, categories, and structures, so that when analysis is needed, the data is already in the correct form. This eliminates the need for time-consuming custom processing for each analysis request while maintaining high accuracy.
3Reliability
If multiple machine learning models are applied for different pre-processing operations, then processing accuracy improves, but system complexity and computational requirements increase
Solution Approach 1:
The patent segments the data processing system into distinct functional layers: data conversion, pre-processing, analysis, and visualization. Each layer has specific machine learning models trained for particular tasks (e.g., one model for format conversion, another for anomaly detection, another for pattern recognition). This segmentation allows each model to be optimized for its specific function while maintaining overall system reliability, and enables independent training and deployment of individual models.
Data Source
AI summary
A method may include receiving tabular data and receiving a first user input of a description of the tabular data and a second user input of a recipient of the tabular data or a report to be generated based on the tabular data. The method may include converting the tabular data to a text-based language to form converted tabular data and performing, using one or more first machine learning models, pre-processing of the converted tabular data. The method may include applying one or more second machine learning models to the converted tabular data based on the description, the recipient, a context of the converted tabular data, or a characteristic of the converted tabular data. The method may include performing one or more actions based on a result of applying the one or more second machine learning models.