Using artificial intelligence to design a product

Abstract

In an embodiment, a method for optimizing computer machine learning includes receiving an optimization goal. The optimization goal is used to search a database of base option candidates (BOC) to identify matching BOCs that at least in part matches the goal. A selection of a selected base option among the matching BOCs is received. Machine learning prediction model(s) are selected based at least in part on the goal to determine prediction values associated with alternative features for the selected base option, where the model(s) were trained using training data to at least identify weight values associated with the alternative features for models. Based on the prediction values, at least a portion of the alternative features is sorted to generate an ordered list. The ordered list is provided for use in manufacturing an alternative version of the selected base option with the alternative feature(s) in the ordered list.

What is claimed is: 
     
         1 . (canceled) 
     
     
         2 . A method, comprising:
 determining one or more components of an optimization goal;   selecting one or more eligible base options based on the optimization goal, wherein the one or more components are associated with one or more target segments and the one or more selected base options are included in the one or more target segments;   evaluating the one or more eligible base options with respect to each of the one or more components of the optimization goal, wherein evaluating the one or more eligible base options includes evaluating a divergence between actual performance of the one or more eligible base options and predicted performance of the one or more base options as a product by a trained model;   identifying one or more candidate base options based on the evaluation;   selecting, based on the one or more target segments associated with the optimization goal, one or more trained models from a plurality of trained models to evaluate alternative features for the one or more identified candidate base options; and   providing an ordered list for use in manufacturing an alternative version of the one or more identified candidate base options with one or more of the alternative features.   
     
     
         3 . The method of  claim 2 , wherein the one or more components are associated with an optimization type. 
     
     
         4 . The method of  claim 3 , wherein the optimization type is one of a sales metric, an inventory metric, a variety metric, a style rating, a size rating, a fit rating, a quality rating, retention, personalization, style grouping, or price value rating. 
     
     
         5 . The method of  claim 2 , wherein the one or more target segments include a target business line, a target product type, a client segment, or seasonality. 
     
     
         6 . The method of  claim 2 , wherein the optimization goal indicates the one or more target segments from a plurality of target segments. 
     
     
         7 . The method of  claim 2 , wherein evaluating the one or more eligible base options includes ranking the one or more eligible base options. 
     
     
         8 . The method of  claim 2 , wherein the one or more identified candidate base options include one or more attributes not accounted for by the trained model. 
     
     
         9 . The method of  claim 2 , wherein each of the one or more components of the optimization goal is associated with a corresponding score indicator. 
     
     
         10 . The method of  claim 9 , wherein the one or more eligible base options are ranked based on corresponding score indicators for each of the one or more components of the optimization goal. 
     
     
         11 . The method of  claim 2 , wherein the one or more identified candidate base options correspond to the one or more eligible base options having a corresponding score above a threshold score. 
     
     
         12 . The method of  claim 2 , wherein the one or more identified candidate base options correspond to a top pre-defined number of the one or more eligible base options. 
     
     
         13 . The method of  claim 2 , further comprising evaluating the alternative features for the one or more identified candidate base options. 
     
     
         14 . The method of  claim 13 , wherein evaluating the alternative features for the one or more identified candidate base options includes weighing prediction results of the plurality of trained models and combining the prediction results to determine an overall prediction value for the alternative features. 
     
     
         15 . The method of  claim 2 , wherein a first alternative feature of the one or more alternative features is added to a first candidate base option of the one or more candidate base options. 
     
     
         16 . The method of  claim 2 , wherein a first alternative feature of the one or more alternative features replaces a first feature of a first candidate base option of the one or more candidate base options. 
     
     
         17 . A system, comprising:
 a processor configured to:
 determine one or more components of an optimization goal; 
 select one or more eligible base options based on the optimization goal, wherein the one or more components are associated with one or more target segments and the one or more selected base options are included in the one or more target segments; 
 evaluate the one or more eligible base options with respect to each of the one or more components of the optimization goal, wherein evaluating the one or more eligible base options includes evaluating a divergence between actual performance of the one or more eligible base options and predicted performance of the one or more base options as a product by a trained model; 
 identify one or more candidate base options based on the evaluation; 
 select, based on the one or more segments associated with the optimization goal, one or more trained models from a plurality of trained models to evaluate alternative features for the one or more identified candidate base options; and 
 provide an ordered list for use in manufacturing an alternative version of the one or more identified candidate base options with one or more of the alternative features; and 
   a memory coupled to the processor and configured to provide the processor with instructions.   
     
     
         18 . The system of  claim 17 , wherein each of the one or more components of the optimization goal is associated with a corresponding score indicator. 
     
     
         19 . The system of  claim 18 , wherein the one or more eligible base options are ranked based on corresponding score indicators for each of the one or more components of the optimization goal. 
     
     
         20 . The system of  claim 17 , wherein the one or more identified candidate base options correspond to the one or more eligible base options having a corresponding score above a threshold score. 
     
     
         21 . A computer program product for optimizing computer machine learning, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:
 determining one or more components of an optimization goal;   selecting one or more eligible base options based on the optimization goal, wherein the one or more components are associated with one or more target segments and the one or more selected base options are included in the one or more target segments;   evaluating the one or more eligible base options with respect to each of the one or more components of the optimization goal, wherein evaluating the one or more eligible base options includes evaluating a divergence between actual performance of the one or more eligible base options and predicted performance of the one or more base options as a product by a trained model;   identifying one or more candidate base options based on the evaluation;   selecting, based on the one or more segments associated with the optimization goal, one or more trained models from a plurality of trained models to evaluate alternative features for the one or more identified candidate base options; and   providing an ordered list for use in manufacturing an alternative version of the one or more identified candidate base options with one or more of the alternative features.