Dynamic Data Trimming for Print Setup Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing input assisting systems for print setup information, particularly for dynamic print products like periodicals, face challenges in accurately suggesting suitable setting values due to low evaluation values from historical data, leading to inappropriate candidate suggestions.
Innovation Solution
The system accumulates setting values associated with time and date information, uses machine learning to create rules that associate operator inputs with candidate values, and performs data trimming to enhance the accuracy of candidate suggestions by adjusting evaluation criteria and data sets based on historical trends.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If historical data is used to determine candidate setting values, then the system can provide automated suggestions, but the accuracy of suggestions decreases for dynamic print products like periodicals
Solution Approach 1:
The system dynamically adjusts the data set used for candidate suggestion by trimming historical data based on the product type. For dynamic print products like periodicals, it uses only recent historical data (e.g., last 6 months) rather than all historical data, allowing the suggestion accuracy to adapt to seasonal and temporal changes in print product specifications
Solution Approach 2:
The system changes the parameter of time range for historical data based on product characteristics. By introducing a trim period parameter that varies according to product type (shorter for dynamic products like periodicals, longer for static products), it optimizes the balance between automation and accuracy for different print products
2Quantity of substance
If all historical setting data is accumulated, then more data is available for analysis, but the relevance of data decreases for products with seasonal variations
Solution Approach 1:
The system extracts only the relevant portion of historical data by trimming old records based on a specified period. This extraction principle removes outdated data that no longer reflects current product specifications, ensuring that only relevant historical data is used for generating candidate suggestions
Solution Approach 2:
The data retention period is dynamically adjusted based on product type characteristics. For periodicals and magazines with frequent specification changes, a shorter trim period is applied, while for books and catalogs with stable specifications, a longer period is used, optimizing data relevance for each product category
Data Source
AI summary
Provided are an input assisting method, a non-transitory computer-readable recording medium and a setup-information input system including an information input apparatus. A hardware processor of the apparatus determines at least one candidate value for at least one input field in a setup screen, by using a first data set created from a database of setting values specified for past jobs, creates a rule for the at least one candidate value, and calculates an evaluation value. On finding no candidate value for which a sufficient evaluation value was calculated, the hardware processor creates a second data set from the first data set, creates a rule for at least one candidate value determined for the at least one input field, by using the second data set, and when detecting an operator's operation on a certain input field, indicates a candidate value determined according to the rule, in the certain input field.