Building User Profiling via Transport Mode Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in accurately determining transport modes and user profiles in buildings using sensor data from user equipment, leading to inaccuracies in mobility patterns and service innovations.
Innovation Solution
A system utilizing a trained machine learning model to analyze mobility features from user equipment, determining transport modes and user profiles by classifying users as health aware, with a stroller, recurrent visitor, or new visitor based on indoor mobility data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If sensor data from user equipment is used to determine transport modes and user profiles, then service innovation and mobility pattern insights are enhanced, but measurement precision and accuracy deteriorate
Solution Approach 1:
The system segments the determination process into distinct stages: first determining transport modes (elevator, escalator, stairs) from sensor data, then using those transport mode determinations to infer user profiles (health aware, with stroller, recurrent visitor, new visitor). This segmentation allows each stage to be optimized independently, improving overall measurement precision while maintaining service innovation capability.
Solution Approach 2:
The patent introduces transport mode determination as an intermediary step between raw sensor data collection and final user profile inference. This intermediary layer acts as a mediator that translates ambiguous sensor patterns into specific transport mode classifications, which then serve as more reliable inputs for user profile determination, thereby improving accuracy throughout the chain.
2Measurement precision
If machine learning models are used to classify user profiles, then determination accuracy is improved, but device complexity increases
Solution Approach 1:
The system performs preliminary classification by determining transport modes before proceeding to user profile classification. By pre-processing the data into intermediate transport mode categories (elevator, escalator, stairs), the subsequent user profile classification becomes simpler and more accurate, reducing the overall system complexity while maintaining high determination accuracy.
Solution Approach 2:
The patent implements a dynamic, multi-stage classification system where the complexity is distributed across different levels. The first stage (transport mode determination) handles basic movement pattern classification, while the second stage (user profile determination) builds upon those results. This dynamic approach allows the system to manage complexity efficiently by breaking down the classification task into manageable stages.
Data Source
AI summary
A system, a method and a computer program product are provided to determine user profile of one or more users in one or more buildings, using a machine learning model. The system may include at least one memory configured to store computer executable instructions and at least one processor configured to execute the computer executable instructions to obtain mobility features associated with the one or more buildings in a geographic region. The processor may be configured to determine using a trained machine learning model, one or more transport modes for the one or more buildings. The processor may be further configured to obtain indoor mobility data of the one or more users based on the one or more transport modes for the one or more buildings. The processor may be further configured determine the user profile of the one or more users in the one or more buildings.