Confidence-Boosted Automated Valuation System for Residential Properties

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current automated valuation models for residential properties lack the ability to confidently and autonomously determine home values, often requiring human intervention and extensive computational resources, and struggle to model confidence in predictions, leading to inefficiencies and inaccuracies in valuations.

Innovation Solution

The confidence-boosted automated valuation system learns from errors in predicted values to generate confidence scores and rank homes based on predicted pricing error, filtering out uncertain valuations and reducing computational burdens by automating the process of selecting confident predictions for presentation to buyers or sellers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated valuation models are used to determine home values, then productivity and efficiency are improved, but reliability and accuracy of valuations deteriorate due to inability to model confidence in predictions

Engineering Contradiction:

Improvevaluation processing efficiencyVSAvoidconfidence in valuation accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system uses feedback from actual home sale prices to train and refine the automated valuation model. By continuously comparing predicted values with actual sale prices and adjusting the model parameters accordingly, the system improves its ability to accurately model confidence in predictions while maintaining high productivity

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system changes the parameter space by introducing confidence scores and error rate predictions as additional output dimensions. This allows the model to not only provide a single valuation estimate but also quantify its uncertainty, thereby improving reliability while maintaining efficiency

Inventive Principle:

Principle #35Parameter changes

2Reliability

If conventional appraisal methods are used, then reliability of valuations is improved through professional judgment, but productivity deteriorates due to time-consuming manual processes

Engineering Contradiction:

Improveaccuracy of home valuationVSAvoidvaluation processing speed

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system replaces the mechanical manual appraisal process with an automated machine learning model that processes home attributes and generates valuations instantly. This substitution maintains reliability by using sophisticated algorithms that replicate and enhance professional judgment while dramatically improving productivity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The automated valuation model performs self-service by automatically analyzing home attributes, comparing them with comparable properties, and generating valuations without requiring manual intervention. The system serves itself by continuously learning from actual sale data to improve its accuracy over time

Inventive Principle:

Principle #25Self-service

3Reliability

If multiple automated valuation models are used to improve accuracy, then reliability is improved, but device complexity and computational resources required increase

Engineering Contradiction:

Improveaccuracy of predicted valuesVSAvoidcomputational resource requirements

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system merges multiple automated valuation models into a unified framework that processes home attributes through a single integrated model architecture. This consolidation maintains high reliability by incorporating the strengths of multiple approaches while reducing device complexity and computational resource requirements through shared processing layers

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS20220172255A1Confident processing of valuations from distributed models systems and methods

Publication Date: 2022.06.02 MFTB HOLDCO INC

AI summary

Confidence-boosted automated valuation systems and methods for performing confident processing of valuations from automated valuation models are disclosed. The confidence-boosted automated valuation system uses home/model data, actual values, and predicted values of homes to train a confidence model to produce confidence scores. The system can partition the homes into confident bins each containing homes with similar confidence scores. To generate a confidence score for a predicted value of a subject home, the confidence-boosted automated valuation system can apply the trained confidence model to the subject home. Using the generated confidence score, the confidence-boosted automated valuation system can identify a confidence bin the subject home falls in. The confidence-boosted automated valuation system can compute a predicted error in the predicted value of the subject home using the confidence bin. Based on the predicted error, the confidence-boosted automated valuation system can determine whether the predicted value is a confident home value.