BROWSE
Detail
Conference Paper
Outlier Elimination and Reliability Assessment for Peak and Declining Time Series Datasets
Cited 0 time in 
Share 
Share
- Authors
- Jungeun Yoon, Aekyeung Moon, Seung Woo Son
- Issue Date
- 2023-12
- Citation
- International Conference on Data Mining Workshops (ICDMW) 2023, pp.593-600
- Language
- English
- Type
- Conference Paper
- Abstract
- Environmental information such as climate and soil has been influencing every aspect of time-sensitive agriculture. With substantial advances in emerging and enabling technologies of IoT sensors, a vast amount of IoT-based environmental data allows preparation for the adverse impacts by providing helpful information for predictive and precise services. However, data acquired by IoT sensors can be corrupted by external environmental factors, which can negatively affect the integrity of data interpretation. To address this issue, our previous study proposed outlier detection techniques using transform-based sparse profiles. However, it would lose its worth without an evaluation methodology for data reliability after probing datasets by outlier detection. Therefore, this study proposes a process for preprocessing defective soil moisture sensor data through outlier detection methods and evaluating the fidelity of the data. To do this, we use the results of statistical evaluation experiments to determine whether the data outliers have been well eliminated, that is, the fidelity of the data. Experiments using real-world soil moisture datasets show that the transform-based OD (outlier detection) method outperforms other statistical outlier detection methods, suggesting that the preprocessed data can improve the integrity of IoT datasets.
- KSP Keywords
- Data Interpretation, Data reliability, Detection Method, Enabling technologies, Environmental data, Environmental information, IoT sensors, IoT-based, Outlier elimination, Real-world, Soil Moisture Sensor