Vehicle Item Graphics Using ML-Based Preference Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Online vehicle purchasing lacks personalized assistance from knowledgeable salespersons, leading to generic and unsophisticated presentation of vehicle items, which can result in sub-optimal selections due to large or incomplete item sets.
Innovation Solution
A system utilizing a machine learning model trained on user data and population data to generate graphics for vehicle items, filtering and displaying only those with high predicted likelihood of user selection based on score thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a salesperson assists in vehicle purchasing, then personalized recommendations and user guidance are improved, but device complexity and automation level deteriorate
Solution Approach 1:
The system enables self-service by using machine learning models to automatically generate personalized vehicle item recommendations without human salesperson intervention. The model processes user data, vehicle information, and interaction patterns to autonomously provide customized suggestions, replacing the need for manual salesperson assistance while maintaining personalized service quality.
Solution Approach 2:
The patent substitutes the mechanical system of human salesperson interaction with an automated computational system. The machine learning model processes information, generates recommendations, and adapts to user preferences through algorithms rather than human cognition and communication, achieving personalization through computational methods instead of human judgment.
2Quantity of substance
If all vehicle items are displayed to users, then completeness of information is improved, but user decision-making difficulty and interface complexity worsen
Solution Approach 1:
The system extracts and highlights only the most relevant vehicle items from the complete set based on machine learning predictions. Instead of displaying all available items, the model identifies and presents the subset of items with highest predicted user interest, removing irrelevant options that would clutter the interface and complicate user decision-making while maintaining access to the full item set when needed.
Solution Approach 2:
The patent applies local quality by providing different levels of information presentation to different users based on their specific characteristics and preferences. The machine learning model analyzes individual user data to customize which items are highlighted, how they are presented, and what additional information is provided, creating a tailored viewing experience for each user rather than a uniform presentation for all users.
3Device complexity
If generic electronically generated menus are used, then device complexity is reduced, but user guidance and personalization deteriorate
Solution Approach 1:
The system transitions from static generic menus to dynamic personalized presentations. The machine learning model continuously adapts the menu content, ordering, and presentation based on real-time analysis of user data, vehicle selection, and interaction patterns. This dynamic adjustment provides personalized guidance while maintaining a relatively simple underlying system architecture that processes user inputs and generates customized outputs.
Data Source
AI summary
A computer-implemented method of generating a graphic for a vehicle item may include: causing a user device to display a user interface indicative of one or more vehicles; receiving, from the user device, vehicle selection information, the vehicle selection information indicative of a vehicle selected by a user; obtaining, from a database, user data corresponding to the user; generating, using a machine learning model, a first score corresponding to a first vehicle item based on the user data; determining whether the first score exceeds a first predetermined score threshold; generating, in response to a determination that the first score exceeds the first predetermined score threshold, a first graphic indicative of the first vehicle item; and causing the user device to display the first graphic via the user interface.