Large language modeling systems and methods for building, testing, and validating a predictive model
A computer system for building, simulating, and/or validating a predictive model. The computer system programmed to (i) analyze one or more predictive pricing sub-models to detect one or more issues; (ii) execute a GPT model on the one or more predictive pricing sub-models, the GPT model trained to identify differences between predicted pricing and actual pricing for certain predefined events; (iii) compare the one or more issues to one or more outputs of the GPT model; (iv) in response to the comparison, generate a new model software template including one or more code changes to the one or more predictive pricing sub-models based upon the comparison; (v) generate a simulation environment based upon a plurality of data; (vi) execute the new model software template in the simulation environment; and/or (vii) update the new model software template based upon one or more outputs of the execution.
1 . A computer system for building, simulating, and validating a predictive model, the system comprising at least one processor, and at least one memory in communication with the at least one processor, the at least one processor programmed to:
detect one or more issues with one or more predictive sub-models, the one or more predictive sub-models configured to, when executed using a trained large language model (LLM), output a predicted value associated with one or more products based on historical data;
execute a GPT model on the one or more predictive sub-models, the GPT model trained to output identified differences between the predicted value outputted by the one or more predictive sub-models and an actual value associated with the one or more products;
input the one or more detected issues and the output of the GPT model into a code LLM configured to output a new model software template including one or more computer executable instruction changes to the one or more predictive sub-models based upon a comparison of the one or more detected issues and the output of the GPT model;
generate a simulation environment based upon a plurality of record data, the plurality of record data including a plurality of historical known values associated with a plurality of products;
execute the new model software template in the simulation environment using the plurality of record data to generate one or more outputs relating to the new model software template; and
update the new model software template based upon the one or more outputs of the execution to generate one or more updated predictive sub-models, the one or more updated predictive sub-models including the one or more computer executable instructions changes.
2 . The computer system of claim 1 , wherein the at least one processor is further programmed to iteratively execute the new model software template in the simulation environment and update the new model software template based upon one or more outputs of the execution until one or more end conditions are met.
3 . The computer system of claim 1 , wherein the GPT model analyzes the one or more issues with the predictive sub-model over a predetermined period of time.
4 . The computer system of claim 1 , wherein an issue includes a consistent variance or difference between a predicted value for replacing the one or more products outputted from an existing predictive sub-model and an actual value for replacing the one or more products.
5 . The computer system of claim 1 , wherein an issue includes a trend exceeding one or more thresholds, wherein the trend includes differences between different predictive sub-models and actual values.
6 . The computer system of claim 1 , wherein the at least one processor is further programmed to communicate with one or more users via natural language processing.
7 . The computer system of claim 1 , wherein the at least one processor is further programmed to transmit an action item to a user via a client device based upon the detection of the one or more issues with the one or more sub-models.
8 . The computer system of claim 1 , wherein the at least one processor is further programmed to:
receive a prompt from a user via a client device to analyze the one or more issues by executing the GPT model on the one or more predictive sub-models; and
output from the GPT model a list of data elements identified as being related to the differences between the predicted value outputted by the one or more predictive sub-models for replacing the one or more products and actual value for replacing the one or more products.
9 . The computer system of claim 1 , wherein the at least one processor is further programmed to generate a production model based upon the updated model software template.
10 . The computer system of claim 1 , wherein the GPT model is trained for a specific jurisdiction.
11 . A computer-implemented method implemented by a computer device including at least one processor in communication with at least one memory, the computer-implemented method comprising:
detecting one or more issues with one or more predictive sub-models, the one or more predictive sub-models configured to, when executed using a trained large language model (LLM), output a predicted value associated with one or more products based on historical data;
executing a GPT model on the one or more predictive sub-models, the GPT model trained to output identified differences between the predicted value outputted by the one or more predictive sub-models and an actual value associated with the one or more products;
inputting the one or more detected issues and the output of the GPT model into a code LLM configured to output a new model software template including one or more computer executable instruction changes to the one or more predictive sub-models based upon a comparison of the one or more detected issues and the output of the GPT model;
generating a simulation environment based upon a plurality of record data, the plurality of record data including a plurality of historical known values associated with a plurality of products;
executing the new model software template in the simulation environment using the plurality of record data to generate one or more outputs relating to the new model software template; and
updating the new model software template based upon the one or more outputs of the execution to generate one or more updated predictive sub-models, the one or more updated predictive sub-models including the one or more computer executable instructions changes.
12 . At least one non-transitory computer-readable storage media having computer-executable instructions embodied thereon, wherein when executed by at least one processor of a computer system, the computer-executable instructions cause the processor to:
detect one or more issues with one or more predictive sub-models, the one or more predictive sub-models configured to, when executed using a trained large language model (LLM), output a predicted value associated with one or more products based on historical data;
execute a GPT model on the one or more predictive sub-models, the GPT model trained to output identified differences between the predicted value outputted by the one or more predictive sub-models and an actual value associated with the one or more products;
input the one or more detected issues and the output of the GPT model into a code LLM configured to output a new model software template including one or more computer executable instruction changes to the one or more predictive sub-models based upon a comparison of the one or more detected issues and the output of the GPT model;
generate a simulation environment based upon a plurality of record data, the plurality of record data including a plurality of historical known values associated with a plurality of products;
execute the new model software template in the simulation environment using the plurality of record data to generate one or more outputs relating to the new model software template; and
update the new model software template based upon the one or more outputs of the execution to generate one or more updated predictive sub-models, the one or more updated predictive sub-models including the one or more computer executable instructions changes.
13 . The computer-implemented method of claim 11 , further comprising iteratively executing the new model software template in the simulation environment and updating the new model software template based upon one or more outputs of the execution until one or more end conditions are met.
14 . The computer-implemented method of claim 11 , wherein the GPT model analyzes the one or more issues over a predetermined period of time.
15 . The computer-implemented method of claim 11 , wherein an issue includes a consistent variance or difference between a predicted value for replacing the one or more products outputted from an existing predictive sub-model and an actual value for replacing the one or more products.
16 . The computer-implemented method of claim 11 , wherein an issue includes a trend between different sub-models exceeding one or more thresholds.
17 . The computer-implemented method of claim 11 , further comprising communicating with one or more users via natural language processing.
18 . The computer-implemented method of claim 11 , further comprising transmitting an action item to a user via a client device based upon the detection of the one or more issues.
19 . The computer-implemented method of claim 11 , further comprising receiving a prompt from a user via a client device to analyze the one or more issues in view of the GPT model.
20 . The computer-implemented method of claim 11 , further comprising generating a production model based upon the updated model software template.
21 . The computer system of claim 1 , wherein the at least one processor is further programmed to, after generating the one or more updated predictive pricing sub-models, generate at least one predicted value using at least one of the one or more updated predictive pricing sub-models.