Multi-level semantic feature extraction method for behavior data of intelligent wearable equipment

Abstract

The invention discloses a multi-level semantic feature extraction method for behavior data of intelligent wearable equipment, and the method comprises the following steps: S101, constructing a single-level behavior space, and finding a group of bases in the behavior space; S102, constructing a multi-level behavior space, and analyzing the behavior data from different granularities; S103, extracting the multi-level semantic features of behavior data in the multi-level behavior space. The method does not need to manually mark the data, and can be suitable for any behavior while greatly reducing the manual cost. Meanwhile, the extracted semantic features can be used for analyzing the behaviors from different granularities, thereby guaranteeing the high recognition precision. Compared with a conventional method based on a pre-defined semantic feature space, the method greatly improves the accuracy. Compared with a conventional method based on a supervised depth neural network, the method can provide higher recognition precision.

Description

technical field [0001] The invention relates to the field of mobile perception, in particular to a multi-level semantic feature extraction method for behavior data of smart wearable devices. Background technique [0002] In recent years, due to the increasing maturity of communication technology, sensor technology and embedded computing technology, the market for smart wearable devices such as smart watches and bracelets has developed rapidly. Compared with smartphones, smart wearable devices are closer to the daily life of users and are easier to be worn by users for a long time. Therefore, the analysis and understanding of user behavior data through smart wearable devices will greatly promote the development of smart medical care, smart economy, insurance and other fields. However, the existing applications are still at a relatively low level: most of the applications in the application market, such as WeChat Sports, Gudong Running, etc., can only record the user's steps ...

AI Technical Summary

Problems solved by technology

However, the process of labeling behaviors requires a lot of manpower and material resources, and the trained network cannot understand unlabeled behaviors. This method is difficult to be applied to real-world scenarios.

[0004] Existing methods that can provide high recognition accuracy largely require researchers to have prior knowledge of usage scenarios or specific behavior types of smart wearable devices, which limits their practicality in real life

In fact, the use environment, equipment, and behavior of wearable device users are highly arbitrary, and manual labeling cannot enumerate all possible situations

With the penetration of wearable devices into people's daily life, the existing methods of understanding and analyzing user behavior data are far from meeting the needs

Images