ABSTRACT
The danger of cell phone addiction, a widespread phenomenon among young people, is that it can unconsciously take up a lot of time for study and work, causing worries about school and work. The existing methods to effectively improve this problem are usually pre-planning and post-analysis, relying on the user's self-control. We propose an LSTM-based neural network predictive classification method from the perspective of preventing problems before they occur. By mining the weeks-long application usage data information collected from users' cell phones, the historical data is classified as whether the timeout is over, and the LSTM network model is trained using these data sequences for learning, and after several rounds of training iterations to form a stable network, and a predictive judgment classification is performed in the test set for short-term future time. The performance of the network model is evaluated using the F1-score, a common judgment indicator in classification problems, and it is verified that the network can effectively learn the correlation between data sequences in a short period scenario and has a good prediction efficiency, thus providing a preventive solution to the problem of cell phone addiction that does not emphasize reliance on subjective self-control.
- Z. -X. Liao, P. -R. Lei, T. -J. Shen, S. -C. Li and W. -C. Peng, "AppNow: Predicting Usages of Mobile Applications on Smart Phones," 2012 Conference on Technologies and Applications of Artificial Intelligence, Tainan, Taiwan, 2012, pp. 300-303, doi: 10.1109/TAAI.2012.18Google ScholarDigital Library
- S. Acharya, A. Shenoy, M. Lewis and N. Desai, "Analysis and Prediction of Application Usage in Android Phones," 2016 2nd International Conference on Advances in Electrical, Electronics, Information, Communication and Bio-Informatics (AEEICB), Chennai, India, 2016, pp. 530-534, doi: 10.1109/AEEICB.2016.7538346Google ScholarCross Ref
- Zachary Chase Lipton. A Critical Review of Recurrent Neural Networks for Sequence Learning. [J]. CoRR,2015, abs/1506.00019.Google Scholar
- A. Joshi, P. K. Deshmukh and J. Lohokare, "Comparative analysis of Vanilla LSTM and Peephole LSTM for stock market price prediction," 2022 International Conference on Computing, Communication, Security and Intelligent Systems (IC3SIS), Kochi, India, 2022, pp. 1-6, doi: 10.1109/IC3SIS54991.2022.9885528Google ScholarCross Ref
- Xin Wang, Ji Wu, Chao Liu, Haiyan Yang, Yanli Du, Wensheng Niu. Fault time series prediction based on LSTM cyclic neural network[J].Journal of Beijing university of aeronautics and astronautics,2018,44(04):772-784.DOI:10.13700/j.bh.1001-5965.2017.0285Google ScholarCross Ref
- Gaoping Li, Zhibang Qiu, Jiaqing Miao, Jing Wang, Xiaojie Ren, Rixin Cheng. Air quality prediction model based on LSTM [J]. Journal of Southwest University for Nationalities (Science & Technology Edition),2023,49(01):67-73Google Scholar
- Rui Ma, Xuefeng Zhang, Ming Hao, Wenjie Fan, Yixian Liu. Research on Anomaly Detection Method of Teaching Management Cloud Platform System based on LSTM [J/OL]. Electronic design engineering: 1-5[2023-03-28]. http://kns.cnki.net/kcms/detail/61.1477.tn.20230316.1431.004.html.Google Scholar
Index Terms
- Mobile Application Usage Forecast Based on LSTM
Recommendations
Cerebral LSTM: A Better Alternative for Single- and Multi-Stacked LSTM Cell-Based RNNs
AbstractDeep learning has rapidly transformed the natural language processing domain with its recurrent neural networks. LSTM is one such popular repeating cell unit used for building these recurrent neural network-based deep learning architectures. In ...
Fast Training of Deep LSTM Networks
Advances in Neural Networks – ISNN 2019AbstractDeep recurrent neural networks (RNN), such as LSTM, have many advantages over forward networks. However, the LSTM training method, such as backward propagation through time (BPTT), is really slow.
In this paper, by separating the LSTM cell into ...
Sales Demand Forecast based on Recurrent Neural Network∗
EITCE '22: Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer EngineeringIn this paper, we compare the performance of RNN, LSTM and GRU, which are the most popular cyclic neural networks, in predicting the total sales of products in each store next month. A dropout layer is added to the model to reduce over-fitting. The ...
Comments