AI Product Features Map for Support Level Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining correlations among terms in product descriptions, such as software tools and support levels, are inefficient due to the scattered nature of information across multiple documents, leading to high computational resource usage and slow data collection processes.
Innovation Solution
Implementing an AI multiclass model to predict the class of content and subsequently using AI binary class models to determine support levels, reducing the number of model instances needed and enhancing processing efficiency by selecting relevant content for analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional methods are used to determine correlations among terms in product descriptions, then comprehensive data collection is achieved, but computational resource usage is high and processing speed is slow
Solution Approach 1:
The patent segments the data collection and analysis process into distinct phases: content identification, class prediction using multiclass models, and support level prediction using binary class models. This segmentation allows for optimized resource allocation at each stage, preventing unnecessary computational expenditure while maintaining comprehensive data collection.
Solution Approach 2:
The patent performs preliminary classification of content into relevant categories before conducting detailed analysis. By using multiclass models to identify and filter relevant content first, the system prepares data in advance for subsequent binary class analysis, reducing overall computational resource requirements while maintaining processing completeness.
2Measurement precision
If multiple AI model instances are used to analyze all content, then accurate predictions are achieved, but the number of computational resources required increases significantly
Solution Approach 1:
The patent divides the prediction task into two separate model types: multiclass models for content classification and binary class models for support level prediction. This segmentation maintains prediction accuracy by specializing each model type for its specific task while reducing the total number of model instances needed compared to using multiple general-purpose models.
Solution Approach 2:
The multiclass model acts as an intermediary between content identification and binary class prediction. It filters and categorizes content before passing it to binary class models, enabling accurate predictions with fewer overall model instances by eliminating the need for every binary model to process every piece of content.
3Reliability
If all content is processed through all AI models, then complete analysis is achieved, but processing time increases significantly
Solution Approach 1:
The patent segments the processing workflow so that content first passes through multiclass models for classification, then only relevant content proceeds to binary class models for support level analysis. This segmented approach maintains complete and reliable analysis while significantly reducing processing time by avoiding redundant model evaluations on all content.
Solution Approach 2:
The multiclass classification serves as a preliminary action that filters content before binary class analysis. By pre-processing content to identify relevant categories and characteristics, the system ensures complete analysis coverage while reducing processing time by preparing data in advance for subsequent specialized analysis.
Data Source
AI summary
Content for products can be identified. For each identified content, at least one class to which the first content pertains can be predicted using an artificial intelligence multiclass model. For each identified first content that corresponds to the at least one class, a support level of the product with regard to at least one class can be predicted using artificial intelligence binary class models. For each identified product, data indicating the product and the support level of the product with regard to the at least one class can be added to a data table. A product features map based on the data indicating the products and a support level of each product with regard to the at least one class can be generated and output.