Sensor Data Personal Characteristic Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for determining personal characteristics, such as gender, height, and age, require user compliance with wearing dedicated sensors and suffer from poor accuracy, making them unsuitable for real-life applications and leading to sub-optimal content targeting.
Innovation Solution
The use of machine learning models trained on sensor data from client devices to identify personal characteristics without the need for constant sensor wear, employing data preprocessing, reference sorting, feature extraction, and classification techniques to accurately determine characteristics like gender, height, and age.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dedicated sensors are worn on anatomical locations to gather data, then personal characteristics can be determined, but user compliance is required and the system becomes inapplicable to real-life situations
Solution Approach 1:
The patent applies universality by using a smartphone camera, which is a multi-functional device already present in most people's pockets, instead of dedicated single-purpose sensors. The camera serves both its original photography function and the new function of gathering data for personal characteristic identification, eliminating the need for specialized wearable sensors and improving ease of operation while maintaining measurement precision
Solution Approach 2:
The system applies self-service by utilizing the smartphone that individuals already carry and use daily. The camera and sensors in the smartphone automatically gather the necessary data without requiring the user to wear additional devices or actively participate in the measurement process, making the system compliant with real-life situations
2Ease of operation
If conventional methods are used to determine personal characteristics, then some identification can be achieved, but accuracy is poor and results in sub-optimal applications
Solution Approach 1:
The patent merges multiple data sources and processing techniques including camera images, sensor data (accelerometer, gyroscope, magnetometer), and machine learning models into a unified system. This combination of multiple measurement modalities and analytical approaches significantly improves measurement precision while maintaining ease of operation, as the smartphone already integrates these components
Data Source
AI summary
An online system trains machine learning models that, when applied to gathered sensor data, determines personal characteristics (e.g., age, gender, height) of an individual in a non-intrusive manner. Specifically, the online system trains a first machine learning model that analyzes sensor data gathered from a client device associated with the individual. The first machine learning model determines whether a trigger event, such as whether the individual is walking, is currently occurring. A second machine learning model trained by the online system analyzes sensor data corresponding to the trigger event to identify the personal characteristics of the walking individual.