Jiasong Han, Yufei Feng, Xiaofeng Zhong
arXiv 8 Feb 2026 · Econometrics
arXiv:2602.07772 · PDF · DOI · OpenAlex · Extracted main text
Communication scene recognition has been widely applied in practice, but using deep learning to address this problem faces challenges such as insufficient data and imbalanced data distribution. To address this, we designed a weighted loss function structure, named FilterLoss, which assigns different loss function weights to different sample points. This allows the deep learning model to focus primarily on high-value samples while appropriately accounting for noisy, boundary-level data points. Additionally, we developed a matching weight filtering algorithm that evaluates the quality of sample points in the input dataset and assigns different weight values to samples based on their quality. By applying this method, when using transfer learning on a highly imbalanced new dataset, the accuracy of the transferred model was restored to 92.34% of the original model's performance. Our experiments also revealed that using this loss function structure allowed the model to maintain good stability despite insufficient and imbalanced data.
appendix boundary found by none_found · 100% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Feng, Y. and Zhong, X. and Zhou, S. and Chen, X (2023) Communication Scene Recognition Method Based on Multi Phone Sensors and Deep Learning self | 1.000 | 5 | 3 | 100% |
| 2 | Chawla, N. V. and Bowyer, K. W. and Hall, L. O. and Kegelmeyer, W. P (2002) SMOTE: synthetic minority over-sampling technique | 0.928 | 4 | 3 | 100% |
| 3 | He, Haibo and Bai, Yang and Garcia, E. A. and Li, Shutao (2008) ADASYN: Adaptive synthetic sampling approach for imbalanced learning | 0.928 | 4 | 3 | 100% |
| 4 | Tomek, I (1976) Two Modifications of CNN | 0.737 | 3 | 2 | 100% |
| 5 | Tang, B. and He, H (2015) ENN: Extended Nearest Neighbor Method for Pattern Recognition [Research Frontier] | 0.737 | 3 | 2 | 100% |
| 6 | Kubat, M. and Matwin, S (1997) Addressing the curse of imbalanced training sets: one-sided selection | 0.737 | 3 | 2 | 100% |
| 7 | Krichen, M (2021) Anomalies detection through smartphone sensors: A review | 0.737 | 3 | 2 | 100% |
| 8 | Lin, T.-Y. and Goyal, P. and Girshick, R. and He, K. and Dollár, P (2017) Focal loss for dense object detection | 0.585 | 3 | 1 | 100% |
| 9 | Bosch Sensortec (2024) BMM150: Geomagnetic sensor | 0.585 | 3 | 1 | 100% |
| 10 | Bosch Sensortec (2024) BMA253: Digital, triaxial acceleration sensor | 0.511 | 2 | 1 | 100% |
Showing the top 10 of 25 scored citations.