Trip Classification via Hot Spot Predictive Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing location-based services struggle to accurately classify user activities, such as trips, due to insufficient data from location information alone, leading to difficulties in distinguishing between business and personal activities.
Innovation Solution
A computer-implemented method that uses a predictive model trained on historical trip records to classify trips by determining hot spots based on these records and associating them with classification information, allowing for more accurate predictions based on new trip records.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If location information alone is used to classify user activities, then the system is simple to operate, but the classification accuracy is insufficient
Solution Approach 1:
The patent combines multiple data sources including location information, trip attributes (duration, distance, time), and historical classification data to classify user activities. This merging of multiple information sources improves classification accuracy beyond what location data alone could provide
Solution Approach 2:
The system performs preliminary classification by determining hot spots from historical trip records before classifying new trips. This pre-processing step creates a knowledge base that accelerates and improves subsequent classification operations
2Measurement precision
If hot spots are determined from historical trip records, then classification accuracy improves, but data processing complexity increases
Solution Approach 1:
The system determines hot spots from historical trip records in advance, creating a pre-processed knowledge base. This preliminary action separates the complex data analysis phase from the real-time classification phase, improving accuracy while managing processing complexity through time-separated operations
Solution Approach 2:
The system creates simplified representations of complex location data by generating hot spot regions that encompass multiple locations. These copied/abstracted representations are easier to process while retaining the essential classification information
3Measurement precision
If parking lots are registered as separate locations, then location tracking precision is high, but activity classification becomes difficult
Solution Approach 1:
The patent merges nearby locations such as parking lots with their associated buildings into unified hot spot regions. This allows the system to maintain precise location tracking while treating related locations as a single classification unit, resolving the difficulty of determining whether a trip to a parking lot is business or personal
Solution Approach 2:
The hot spot concept acts as an intermediary that connects precise location data with classification categories. Instead of directly classifying individual locations like parking lots, the system uses hot spots as intermediate entities that aggregate multiple locations and their associated classification information
4Adaptability or versatility
If self-employed users travel to varied locations, then service coverage is improved, but determining business vs personal travel becomes difficult
Solution Approach 1:
The system changes the classification parameters from individual location-based decisions to patterns across multiple trips. By analyzing trip attributes such as duration, distance, time of day, and frequency to and from various locations, the system can distinguish business from personal travel even when destinations vary widely
Solution Approach 2:
The system uses historical trip records to establish baseline patterns for each user before classifying new trips. This preliminary analysis of user behavior patterns enables accurate classification of trips to varied locations by comparing them against established business and personal travel patterns
Data Source
AI summary
Aspects of the present disclosure provide techniques for classifying a trip. Embodiments include receiving, from a plurality of users, a plurality of historical trip records. Each of the plurality of historical trip records may comprise one or more historical trip attributes and historical classification information. Embodiments include training a predictive model, using the plurality of historical trip records, to classify trips based on trip records. Training the predictive model may comprise determining a plurality of hot spots based on the historical trip records, each of the plurality of hot spots comprising a region encompassing one or more locations, and associating, in the predictive model, the plurality of hot spots with historical classification information. Embodiments include receiving, from a user, a new trip record comprising a plurality of trip attributes related to a trip and using the predictive model to predict a classification for the trip based on the trip record.