Quality Management of Crowd Sensing Data Based on Machine Learning
Zhongxian Bai, Rongqing Zhuo · 2020 International Conference on Computer Information and Big Data Applications (CIBDA) · 2020
Recently, research on crowd sensing data quality management is a new subject area developed based on wireless sensor network related concepts. Crowd sensing data has also experienced many links in the process of network propagation, so it is inevitable that there are abnormal data in the database. Therefore, how to filter these unreliable data to get more real data is particularly important. This paper takes somatosensory temperature as an example, solves the problem of calculating the similarity of unequal long-term sequences by using DTW technology, and then clusters and compares the data in the database to find out the abnormal data. Thereby, the accuracy of the somatosensory temperature database is improved, and relevant users can obtain more accurate information. The experimental results show that when the minimum DTW value exceeds the threshold t = 0.7, the more the number of simulation sequences, the more stable the accuracy rate, and the faster the growth rate of the running time.