Sensor Data Complement Method Based on Collaborative Filtering
LI Fe · Journal of Northeastern University · 2014
Completing the lack data monitored by the sensors is a key problem of the information sensing process in the Internet of things. In order to solve this problem,a method w as proposed, w hich can complete sensor data by using collaborative filtering. Considering that there are a lot of similarities among monitor data from sensors in the same area or from the one w ith different periods,properties of space-time correlation betw een sensors w ere adopted in the proposed method. Different similarity evaluation w as used to select similar sensors by classifying sensors w ith missing data in order to ensure the accuracy of the estimate. The results show ed that using this method to estimate the missing data w as better than other methods w hen there are large changes in the environment.