Vehicular Trip Validation Using Device Interaction Baselines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to accurately classify vehicular trips as personal or non-personal use based on driving data, which is crucial for insurance and business reporting.
Innovation Solution
A computer-implemented method and system that classifies vehicular trips by training a classification model on historic device interaction data from work and personal trips, identifying representative features, and comparing them to baseline features to determine trip purpose.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a classification model is trained on historic device interaction data to classify vehicular trips, then classification accuracy is improved, but device complexity increases
Solution Approach 1:
The system segments the classification task by separating historic device interaction data into distinct work trip data and personal trip data sets. The classification model then processes unlabeled trip data by comparing it against these segmented baseline features, enabling accurate classification without requiring complex real-time analysis of all trip characteristics simultaneously.
Solution Approach 2:
The system performs preliminary action by pre-training the classification model on historic labeled data and pre-computing baseline device interaction features for work and personal trips. This preliminary processing allows the model to make accurate classifications on new unlabeled trip data without requiring complex computation at classification time, thus improving accuracy while managing system complexity.
2Reliability
If device interaction features are collected and compared to classify trips, then classification reliability is improved, but loss of information increases
Solution Approach 1:
The system extracts and compares specific device interaction features from unlabeled trip data against pre-established baseline features for work and personal trips. By focusing on and extracting only the relevant feature comparisons rather than processing all available data, the system maintains high classification reliability while minimizing information loss through targeted feature analysis.
Solution Approach 2:
The system creates a copy of the classification process by comparing unlabeled trip features against replicated baseline feature sets for work and personal trips. This copying approach allows the system to evaluate trip classification reliability through feature similarity matching without requiring complex data transformation or processing that could lead to information loss.
Data Source
AI summary
Method, system, device, and non-transitory computer-readable medium for classifying a vehicle trip. In one aspect, a computer-implemented method includes: obtaining a user classification associated with an unlabeled vehicular trip; obtaining a first set of historic device interaction data associated with work; obtaining a second set of historic device interaction data associated with personal use; training a classification model based at least in part upon the first set of historic device interaction data and the second set of historic device interaction data; obtaining a set of unlabeled device interaction data associated with the unlabeled vehicular trip; identifying and comparing a first set of baseline device interaction features, a second set of baseline device interaction features, a set of representative device interaction features; classifying the unlabeled vehicular trip; and validating the user classification based at least in part upon the user classification and the classification made using the classification model.