AppUsage2Vec

This is Pytorch implementation for the paper:

Zhao, S., Luo, Z., Jiang, Z., Wang, H., Xu, F., Li, S., ... & Pan, G. (2019, April). Appusage2vec: Modeling smartphone app usage for prediction. In 2019 IEEE 35th International Conference on Data Engineering (ICDE) (pp. 1322-1333). IEEE.

Introduction

AppUsage2Vec is a novel framework for app usage prediction. This model exploited 'App Attention', 'Dual DNN', 'Temporal Context', and 'Top k Based Optimization' structures.

Environment Requirement

The code has been tested running under Python 3.7.7. Look at the requirements.txt for more detail

• torch==1.5.1

• scikit-learn==0.23.2

• jupyter-lab==2.2.6

• pandas==1.1.3

• matplotlib==3.3.4

• CUDA 10.1

• CUDNN 7

Docker

If you want to run the code by using docker, follow instructions below.

• Docker build

docker build -t appusage2vec .
docker run --rm -it --gpus all -p 8888:8888 -v {your_path}/AppUsage2Vec:/AppUsage2Vec appusage2vec /bin/bash

• Jupyter lab in container

jupyter lab --allow-root --ip=0.0.0.0

And copy the url in the terminal and paste it on the web browser to use jupyterlab in the container.

Dataset

Tsinghua App Usage Dataset

@article{yu2018smartphone,
    title={Smartphone app usage prediction using points of interest},
    author={Yu, Donghan and Li, Yong and Xu, Fengli and Zhang, Pengyu and Kostakos, Vassilis},
    journal={Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies},
    volume={1},
    number={4},
    pages={174},
    year={2018},
    publisher={ACM}}

Extract App_usage_trace.txt and put in in the data directory.

Preprocessing

Run preprocessing.ipynb. If you want to change sequence length, change seq_length in 4th cell.

How to run

After extracting dataset and preprocessing,

python main.py

Result

Please note that the result is not fine-trained well. (Just used default arguments in main.py)