Dictionary Reducer for Business Name Categorization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current energy management systems face challenges in efficiently categorizing businesses for effective energy management messaging, especially when business types are unknown, requiring extensive computing resources and labor-intensive processes.
Innovation Solution
An energy management system utilizing a neural network, predictive model, and dictionary reducer to calculate weights for single-word terms and part of speech from business names, predicting accurate business categories with high confidence, and eliminating unessential terms for efficient processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional business name categorization methods are used, then comprehensive business category classification can be achieved, but extensive computing resources and processing time are required
Solution Approach 1:
The patent segments the business name categorization process into multiple stages: initial filtering using a reduced dictionary, neural network-based weight calculation for candidate terms, and predictive model-based category assignment. This segmentation allows the system to process business names efficiently by breaking down the complex classification task into manageable steps, each optimized for specific purposes.
Solution Approach 2:
The patent performs preliminary action by pre-processing business names through a reduced dictionary to identify candidate terms before applying the full neural network and predictive model. This preliminary filtering step reduces the computational burden by eliminating obviously irrelevant terms early in the process, allowing resources to be focused on more discriminating analysis.
2Measurement precision
If comprehensive business name analysis is performed, then accurate business category prediction can be achieved, but extensive computing resources are required
Solution Approach 1:
The patent extracts only the most relevant terms from business names using a reduced dictionary, separating these candidate terms from the rest of the business name text. This extraction process allows the system to focus computational resources on analyzing only the discriminative terms that actually contribute to accurate category prediction, rather than processing entire business names uniformly.
Solution Approach 2:
The patent applies local quality by calculating neural network weights specifically for candidate terms identified in the reduced dictionary, rather than uniformly processing all terms in business names. Each candidate term receives customized weight calculation based on its relevance to specific business categories, optimizing computational efficiency while maintaining prediction accuracy.
3Measurement precision
If manual analyst validation is used for business categorization, then high accuracy can be achieved, but labor-intensive processes and time consumption occur
Solution Approach 1:
The patent implements self-service by enabling the system to automatically validate and categorize business names without requiring manual analyst intervention. The neural network and predictive model work together to self-correct and refine category assignments based on the calculated weights of candidate terms, achieving high accuracy through automated iterative processing rather than human review.
Solution Approach 2:
The patent incorporates feedback mechanisms where the predictive model uses the neural network-calculated weights to refine business category predictions. The system continuously adjusts its categorization decisions based on feedback from weight calculations and prediction outcomes, enabling automated high-accuracy classification without manual validation while maintaining adaptability to different business naming patterns.
Data Source
AI summary
An energy management system includes a neural network, a predictive model, and a dictionary reducer. The network iteratively calculates weights, resulting in a final set, for each of single-word terms and part of speech terms of training data business names, each of the weights indicative of a likelihood of correlating a business category. The predictive employs sets of the weights to predict a first corresponding one of the plurality of business categories for each of the training data business names until employment of the final set accurately predicts a correct business category for the each of the training data business names, and subsequently employs the final set of the weights to predict a second corresponding one of the plurality of business categories for each of a plurality of operational business names. The dictionary reducer eliminates unessential terms taken to determine the plurality of single-word terms and part of speech terms.