- Patent Title: Predicting the value of an asset using machine-learning techniques
-
Application No.: US17494688Application Date: 2021-10-05
-
Publication No.: US11887168B2Publication Date: 2024-01-30
- Inventor: Don Goonetilleke
- Applicant: Alt Platform Inc.
- Applicant Address: US CA San Francisco
- Assignee: ALT PLATFORM INC.
- Current Assignee: ALT PLATFORM INC.
- Current Assignee Address: US CA San Francisco
- Agency: Fenwick & West LLP
- Main IPC: G06Q30/02
- IPC: G06Q30/02 ; G06Q30/0601 ; G06N20/00 ; G06F16/951 ; G06F18/23

Abstract:
A system and a method are disclosed for a platform that determines the classification of a physical asset received from a provider. For a physical asset, the platform extracts intrinsic data from the asset and compiles extrinsic data by web scraping data associated with the asset. The platform generates a principal component vector comprising the intrinsic data and extrinsic data and inputs the principal component vector into a machine learning model. The machine-learning model is trained using training data that includes an aggregation of principal components of a physical asset as labeled by a historical classification and a time associated with the entry. The machine-learning model outputs the classification of the physical asset corresponding to a present time. The platform provides the classification to a user in visual association with a representation of the physical asset.
Information query