Spreadsheet Table Detection via Cell Feature Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing table detection methods in spreadsheets face challenges in accuracy due to varying user preferences and complex table structures, such as omitted titles or sub-headers, and the presence of empty cells, which complicates the identification of table regions.
Innovation Solution
A method that extracts attributes from spreadsheet cells, determines features based on these attributes, and clusters cells into candidate regions to identify tables, using a mapping relationship trained on sample spreadsheets to adjust boundaries and update errors, ensuring accurate detection without altering existing data storage styles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional table detection methods are used, then the detection process is simple, but the detection accuracy is low due to varying user preferences and complex table structures
Solution Approach 1:
The patent segments the table detection process into multiple distinct modules: feature extraction module that identifies cell attributes, feature determination module that processes these attributes into meaningful features, candidate region division module that groups cells into potential table areas, and candidate table determination module that identifies final table structures. This segmentation allows each module to specialize in specific aspects of detection, improving overall accuracy while maintaining manageable complexity through modular design.
Solution Approach 2:
The patent performs preliminary feature extraction and feature determination before actual table detection. By pre-processing the spreadsheet data to extract cell attributes (such as content type, formatting, position) and determine meaningful features (such as header identification, data patterns) in advance, the system prepares structured information that facilitates more accurate subsequent table region identification and reduces complexity during the actual detection phase.
2Adaptability or versatility
If multiple tables are inserted with different sizes and locations, then user preferences are accommodated, but the detection of table regions becomes more difficult
Solution Approach 1:
The patent applies local quality by extracting and determining features specific to each individual cell rather than treating the entire spreadsheet uniformly. Each cell's attributes (content, formatting, position) and features (header status, data patterns) are analyzed locally, allowing the system to adapt to varying table structures, sizes, and locations according to different user preferences while maintaining consistent detection methodology across diverse scenarios.
3Ease of operation
If empty cells are present in tables, then flexible data entry is enabled, but the identification of table boundaries becomes ambiguous
Solution Approach 1:
The patent implements a universal feature extraction and determination mechanism that handles both empty and non-empty cells uniformly. The feature extraction module identifies attributes of all cells regardless of content, and the feature determination module processes these attributes to determine features such as table boundaries, headers, and data regions. This universal approach allows the system to maintain accurate table boundary identification even when empty cells are present, as the detection relies on structural and contextual features rather than merely cell content presence.
Data Source
AI summary
The subject matter described herein relates to table detection in a spreadsheet. According to implementations of the subject matter described herein, there is proposed a solution for determining a table in a spreadsheet. In the solution, respective multiple attributes of multiple cells comprised in the spreadsheet may be extracted. Respective features of the multiple cells may be determined based on the extracted multiple attributes. The multiple cells may be divided into at least one candidate area based on the features. At least one candidate table in the spreadsheet may be determined based on the at least one candidate area. By means of the solution, respective features of the multiple cells comprised in the spreadsheet may be determined based on the respective attributes of the multiple cells, and further, a candidate region where a table might exist may be determined based on the respective features of the multiple cells.