IV International Conference on Data Science and Applications (ICONDATA’20), Muğla, Turkey, 4 - 06 June 2021, pp.495-499
With the widespread use of the Internet of Things (IoT), the instantly generated data size and data production speed are constantly increasing. These data continuously produced by IoT are called streaming data. Streaming data is data that is generally sent simultaneously and in small sizes (in kilobyte size) and continuously generated by thousands of data sources. Flowing data is up-to-date because it is constantly produced, and useful information is obtained when analyzed with analysis methods such as data mining methods frequently used today. In order to analyze the flowing data, the data must be recorded in database systems or file systems in an appropriate format. However, saving the flowing data in the systems is a very difficult problem due to the speed and continuity of the data and also because there may be situations where the data must be processed before the data is saved. In this study, a buffer-based system is proposed for pre-processing and recording of data in the data class that is continuously sent and flowing by the validator devices in Kayseri Metropolitan Municipality (KBB) transportation vehicles. The performance of the proposed method was compared with the classical method on real data. Experimental results have shown that the proposed buffer-based method responds faster than the classical method.