Touch Gesture Emotion Modeling for Website Return Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional web analytics struggle to accurately capture the nuanced emotional experiences of users interacting with websites, leading to biased results and limited understanding of user behavior, particularly in influencing future visits.
Innovation Solution
A non-intrusive system measures touch-gesture based emotions using machine learning models to predict user return to a website by analyzing time-evolving movements, incorporating a Light Gradient Boosting Machine for enhanced prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional web analytics methods (eye tracking, speech tracking, surveys) are used to capture user emotional experience, then insight into user behavior is improved, but intrusiveness increases leading to biased results
Solution Approach 1:
The patent replaces traditional mechanical/intrusive measurement methods (eye tracking, speech tracking, surveys) with a computational approach that analyzes naturally occurring touchscreen gestures. Machine learning models process gesture data (touch duration, pressure, movement patterns) to infer emotional states without directly observing or interfering with the user, thereby eliminating the bias introduced by intrusive measurement while maintaining measurement precision.
2Loss of information
If intrusive methods like eye tracking and surveys are used, then understanding of user behavior is improved, but user interaction authenticity deteriorates
Solution Approach 1:
The system allows users to naturally interact with the touchscreen without any direct observation or intervention. Users perform gestures to navigate and engage with content, and the system passively collects gesture data that reflects their authentic emotional responses. This self-service approach maintains interaction authenticity while gathering behavioral insights, as users are unaware they are being measured and thus behave naturally.
3Measurement precision
If touch gesture analysis with machine learning is used to predict user return, then prediction accuracy is improved to over 90%, but system complexity increases
Solution Approach 1:
The patent leverages the touchscreen device's existing multi-functionality by repurposing its gesture recognition capabilities (already present for navigation and interaction) to also perform emotional state detection and user return prediction. The same hardware and basic software infrastructure that enables users to scroll, tap, and swipe are utilized to collect emotional data, eliminating the need for separate dedicated measurement devices and reducing overall system complexity despite the advanced analytical capabilities.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for predicting a return to a site. In some implementations, a system obtains data indicative of a time evolving movement of a user interacting with a website shown on the client device. The system determines, using a first trained machine learning model and based on the data indicative of the time evolving movement, a metric associated with an emotion of the user corresponding to the user's interaction with the website. The system obtains, from a metric database, metrics associated with an identifier of the user. The system provides, to a second trained machine learning model, (i) the metric associated with the emotion and (ii) data representing the obtained metrics associated with the identifier. The system generates, using the second trained machine learning model, a prediction indicating whether the user is likely to return to the website.