Category Name Extraction for E-Commerce Product Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual maintenance of category names on electronic commerce sites is time-consuming and labor-intensive, especially when new products do not fit existing categories or when certain names become more common, requiring frequent updates.
Innovation Solution
A category name extraction device that specifies common words across multiple categories, extracts candidate category names from item information, and outputs them for automatic setting, reducing manual effort by identifying appropriate words for categorization based on frequency, search data, and product relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual maintenance of category names is performed, then category classification accuracy can be maintained, but time consumption and labor effort increase significantly
Solution Approach 1:
The system enables automatic category name extraction and maintenance through self-service mechanisms. The extraction unit automatically identifies candidate category names from item information without requiring manual intervention, while the setting unit autonomously updates category names based on extraction results, replacing manual maintenance operations with automated self-maintenance processes
Solution Approach 2:
The patent replaces the mechanical manual maintenance process with an automated information processing system. The specifying unit, extraction unit, and setting unit work together to automatically extract category names from item information and update category structures, substituting human manual operations with automated computational processes that analyze text data and perform classification updates
2Reliability
If manual maintenance of category names is performed, then category structures can be updated accurately, but labor effort and operational complexity increase
Solution Approach 1:
The system performs self-service by automatically extracting category names from item information and updating category structures without manual intervention. The extraction unit identifies candidate names, and the setting unit automatically applies updates, enabling the system to maintain and update itself autonomously without requiring operational personnel
Solution Approach 2:
The patent replaces manual category maintenance operations with an automated information processing system. The specifying unit identifies target items, the extraction unit retrieves category names from item information, and the setting unit updates category structures automatically, substituting complex manual operational processes with streamlined automated procedures
3Productivity
If automatic category name extraction is implemented, then maintenance time is reduced, but system complexity increases
Solution Approach 1:
The system segments the category maintenance process into three distinct functional units: the specifying unit that identifies target item information, the extraction unit that retrieves category names, and the setting unit that updates category structures. This segmentation divides the complex maintenance task into manageable modular components, each with a specific function, reducing overall system complexity while maintaining high productivity
Solution Approach 2:
The extraction unit serves multiple functions by extracting category names from various types of item information across different categories. The system universally applies the same extraction and setting processes to diverse data inputs, enabling a single automated system to handle multiple category maintenance tasks simultaneously, improving productivity without proportionally increasing complexity
Data Source
AI summary
A category name extraction device includes a specifying means configured to specify a word contained in a plurality of item information respectively belonging to a plurality of categories in parallel structure, qualifying or being qualified by a name of a category where each item information belongs, and being in common to a plurality of item information belonging to a plurality of different categories, as a reference word, an extraction means configured to extract a word contained in a phrase contained in item information belonging to any of the plurality of categories, qualifying or being qualified by the reference word, and being different from names of the plurality of categories, as a candidate category name, and an output means configured to output the candidate category name extracted by the extraction means.