MFP Data Transmission Destination Prediction Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing systems for designating a data transmission destination in MFPs are labor-intensive and inconvenient, especially when regular transmission is required, as users must manually input the destination each time, and in some cases, pre-registration is not possible.
Innovation Solution
A management system that accumulates transmission information, generates training data using machine learning to predict a data transmission destination, and provides a prediction process using a learning model, allowing for automatic prediction and easier data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual input of transmission destination is required, then transmission accuracy is ensured, but operation labor increases
Solution Approach 1:
The system automatically designates transmission destinations by analyzing document content and comparing it with registered document groups, enabling the system to serve itself without manual intervention. The MFP autonomously determines the appropriate transmission destination based on the document being processed, eliminating the need for users to manually input destinations while maintaining accuracy through the registered document group database.
Solution Approach 2:
Document groups are registered in advance with their associated transmission destinations stored in the system. By pre-configuring the relationships between document types and transmission destinations, the system prepares the necessary information beforehand, allowing automatic designation during actual transmission operations without requiring real-time manual input.
2Loss of time
If transmission destination is registered beforehand, then operation time is reduced, but adaptability decreases
Solution Approach 1:
The system dynamically selects transmission destinations based on the specific document being processed. Instead of using fixed pre-registered destinations, the MFP analyzes the current document's content, compares it with registered document groups, and automatically determines the most appropriate transmission destination. This dynamic approach allows the system to adapt to different documents while still operating automatically.
Solution Approach 2:
The system serves multiple functions by combining automatic document recognition, classification against registered document groups, and dynamic transmission destination selection. A single automatic designation mechanism handles various document types and transmission scenarios, making the system both time-efficient and adaptable to different situations.
3Ease of operation
If pre-registration of transmission destination is implemented, then convenience is improved, but device complexity increases
Solution Approach 1:
The system segments the transmission destination selection process into distinct functional modules: document analysis, comparison with registered document groups, and destination determination. By dividing the complex task into separate processing stages, the system manages complexity while maintaining convenience through automatic operation.
Solution Approach 2:
Registered document groups serve as an intermediary database that mediates between the document being processed and the transmission destination. Instead of directly complex destination selection logic, the system uses the document group database as a reference intermediary, simplifying the overall system architecture while enabling automatic and convenient destination designation.
Data Source
AI summary
A management system for managing a data transmission destination comprises an accumulation unit configured to accumulate, when data transmission is performed, transmission information of the data transmission as collected data; a generation unit configured to generate training data including a pair of ground truth data containing transmission destination information, and input data containing an item other than the transmission destination information, from the collected data accumulated in the accumulation unit; a processing unit configured to generate, by machine learning, a learning model to be used to predict a transmission destination, by using the training data generated by the generation unit; and a providing unit configured to provide a prediction process using the learning model in response to a request.