Image
Image
Image
Image
Image
Image
Image
Image
Image
Image



Search
»

RFID Data for Feature Evolvable Streaming Learning

Summary

This package contains the RFID dataset collected by Mr. Bo-Jian Hou (houbj@lamda.nju.edu.cn) for feature evolvable streaming learning, which has been first used in:

[1] B.-J. Hou, L. Zhang, and Z.-H. Zhou. Learning with Feature Evolvable Streams. In: Advances in Neural Information Processing Systems 30 (NIPS'17) (Long Beach, CA), 2017.

The RFID dataset has also been used in:

[2] C. Hou and Z.-H. Zhou. One-pass learning with incremental and decremental features. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(11): 2776-2792.

[3] Y.-L. Zhang and Z.-H. Zhou. Multi-instance learning with key instance shift. In: Proceedings of the 26th International Joint Conference on Artificial Intelligence (IJCAI'17), Melbourne, Australia, 2017, pp.3441-3447.

[4] B.-J. Hou, L. Zhang, and Z.-H. Zhou. Prediction with Unpredictable Feature Evolution. Preprinted. [Arxiv]

[5] B.-J. Hou, L. Zhang, and Z.-H. Zhou. Learning with feature evolvable streams. IEEE Transactions on Knowledge and Data Engineering, in press.

ATTN: You can feel free to use the package (for academic purpose only) at your own risk. For other purposes, please contact Prof. Zhi-Hua Zhou (zhouzh@nju.edu.cn).

ATTN2: Please cite at least [5] if you use the data in any way.

Details

We use the RFID technique to collect the real data. RFID technique is widely used to do moving goods detection [6]. In our case, we want to utilize the RFID technique to predict the location of the moving goods attached by RFID tag. Concretely, we arranged several RFID aerials which are used to receive the tag signals around the indoor area. In each round, each RFID aerial received the tag signals, then the goods with tag moved~(only on the horizontal direction), at the same time, we recorded the goods' coordinate. Before the aerials expired, we arranged new aerials beside the old ones to avoid the situation without aerials. Therefore, in this overlapping period, we have data from both old and new feature spaces. After the old aerials expired, we continue to use the new ones to receive signals. Then we only have data from new feature space.

Number of examples: 940

Number of features: 78 (old feature space), 72 (new feature space)

[6] C. Wang, L. Xie, W. Wang, T. Xue, and S. Lu. Moving tag detection via physical layer analysis for large-scale RFID systems. In: Proceedings of the 35th Annual IEEE International Conference on Computer Communications, 2016 (INFOCOM), San Francisco, CA, USA, pp.1–9.

ATTN3: This package was developed by Mr. Bo-Jian Hou (houbj@lamda.nju.edu.cn). For any problem concerning the package, please feel free to contact Mr. Hou.

Download: [RFID] (958KB)
  Name Size

Image
PoweredBy © LAMDA, 2022