Synthetic-Augmented-5G-features of the CICIDS2017 dataset

Initial public release of the Synthetic Augmented 5G Features Dataset. The synthetic 5G features were obtained from the Wednesday data of the CICIDS2017 dataset. After preprocessing by removing entire rows with infinity or not a number (NaN), we added 5G-specific features: network latency, signal strength, and user equipment (UE) throughput to each record. Following the prepressing approach highlighted above, the dataset is free of overfitting, data leakage, or strong feature-target correlations and hence ready for immediate use. We provide both the CSV file and the RAR file for convenient download and usage. Please cite the dataset and the supporting paper whenever you make use of it. Please cite the dataset with the following citations: 1. BibTeX format@dataset{ajibade_2026_20363569, author = {Ajibade, Ganiyu Ayodele}, title = {Version 1.0: Synthetic-Augmented-5G-features of the CICIDS2017 dataset }, month = may, year = 2026, publisher = {Zenodo}, doi = {10.5281/zenodo.20363569}, url = {https://doi.org/10.5281/zenodo.20363569},} 2. APA formatAjibade, G. A. (2026). Version 1.0: Synthetic-Augmented-5G-features of the CICIDS2017 dataset [Data set].In Explainable AI for Anomaly Detection in 5G Networks: Bridging Accuracy and Interpretability. Zenodo.https://doi.org/10.5281/zenodo.20363569