Internet of Moving Things are connected to a variety of different types of sensors to form a world of moving things, including people, animals, vehicles, drones, and boats, etc. As the data of collectible moving things continue to increase, anomaly detection of moving things has become an increasingly popular data mining task. Traditional trajectory outlier detection algorithms can detect common anomalies effectively, but it is hard to detect generalized anomalies, such as viewable direction anomalies, gravity anomalies, and magnetic field anomalies which can be collected by the accelerometer, gyroscope, magnetometer, and RPM sensor, etc. For this, we proposed a generalized approach for anomaly detection from the Internet of Moving Things, called the moving things outlier detection algorithm (MTOD). We propose the distance of moving things, which is equal to the weighted sum of the location distance and the multi-sensor distance, and then use the multi-sensor data generalization and moving things partitioning and anomaly detection threestep framework to detect the generalized anomaly. The experimental results show that our MTOD algorithm can detect moving things anomaly efficiency and accurately.

In this article, we propose a novel method that can measure the similarity of FoV-tagged videos in two dimensions. Recently many researchers have focused on measuring the similarity of FoVtagged videos. The similarity measurement of FoV-tagged videos plays an important role in various societal applications, including urban road networks, traffic, and geographic information systems. Our preliminary work introduced the Largest Common View Subsequences (LCVS) algorithm for computing the similarity of FoV-tagged videos. However, LCVS requires a high computational cost for calculating common viewable regions between two FoV-tagged videos. To handle this limitation, we propose the largest View Vector Subsequence (VVS) algorithm for reducing the computational cost of FoV-tagged videos. VVS uses the movement distances and the viewable direction distances to support the simplified vector-based similarity measurement. We demonstrate the superiority of our approach by comparing it with the Longest Common Subsequences (LCSS) and our prior work (LCVS). Our experimental results show that VVS outperforms the prior work in terms of the computational cost and enhances the versatility and stability of the similarity measurement.INDEX TERMS FoV-tagged video, similarity measurement, recommendation system.