GRU Neural Network for Application Preloading Prediction

Solution Overview

Problem

Existing application preloading mechanisms on terminals face challenges in optimizing resource allocation and reducing power consumption, leading to inefficient prediction of user behavior and increased memory occupation, which affects the fluency and efficiency of terminal operations.

Innovation Solution

A method utilizing a gated recurrent unit (GRU) neural network to predict application usage by analyzing user behavior samples, sorting applications by frequency, and grouping usage timing records to optimize preloading, thereby reducing unnecessary resource loading and improving prediction accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If applications are preloaded in advance to improve fluency, then application startup speed is improved, but memory occupation increases and power consumption increases

Engineering Contradiction:

Improveapplication startup speedVSAvoidpower consumption

Core Design Contradiction:

SpeedVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary actions by preloading applications before they are actually needed. The GRU neural network predicts which applications the user is likely to use next and preloads them into memory in advance, so that when the user actually launches an application, it starts immediately without loading delays, thus improving startup speed while avoiding the penalty of preloading unnecessary applications.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system changes the parameter of application selection from random or static criteria to dynamic prediction-based selection. The GRU neural network analyzes historical usage patterns and generates probability scores for different applications, dynamically determining which applications to preload based on predicted user behavior, thereby optimizing the balance between startup speed and power consumption.

Inventive Principle:

Principle #35Parameter changes

2Speed

If applications are preloaded in advance to improve fluency, then application startup speed is improved, but memory occupation increases

Engineering Contradiction:

Improveapplication startup speedVSAvoidmemory occupation

Core Design Contradiction:

SpeedVSQuantity of substance

Solution Approach 1:

The system performs preliminary actions by preloading applications before they are actually needed. The GRU neural network predicts which applications the user is likely to use next and preloads them into memory in advance, so that when the user actually launches an application, it starts immediately without loading delays, thus improving startup speed while avoiding the penalty of preloading unnecessary applications.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system changes the parameter of application selection from random or static criteria to dynamic prediction-based selection. The GRU neural network analyzes historical usage patterns and generates probability scores for different applications, dynamically determining which applications to preload based on predicted user behavior, thereby optimizing the balance between startup speed and power consumption.

Inventive Principle:

Principle #35Parameter changes

3Ease of operation

If random application preloading is performed, then application fluency is improved, but prediction accuracy decreases and resource waste increases

Engineering Contradiction:

Improveapplication fluencyVSAvoidprediction accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The system uses the terminal's own historical usage data to train the GRU neural network, allowing it to learn and predict the specific user's behavior patterns. The model processes sequences of application usage and generates predictions tailored to that user's habits, achieving high prediction accuracy without requiring external data or random preloading strategies.

Inventive Principle:

Principle #25Self-service

Data Source

PatentEP3486771B1Prediction of applications to be preloaded based on observed user behaviour and the order of starting the applications

Publication Date: 2021.10.06 GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP LTD

Figure 1~2
Figure 3~4
Figure 5

AI summary

Embodiments of the present application disclose an application prediction method, an application preloading method and an application preloading apparatus. The application prediction method includes: obtaining a user behavior sample in a preset time period, where the user behavior sample includes an association record of usage timing of at least two applications; grouping the association record of usage timing to obtain a plurality of association record groups of usage timing; and training a preset GRU neural network model according to the plurality of association record groups of usage timing to generate an application prediction model. Embodiments of the present application, by adopting the above solution, may take full advantage of the association record of usage timing of the applications which may truly reflect the user behavior, optimize the application preloading mechanism, improve the precision of the application prediction model training.