In this paper, we introduce a system of integrating activity recognition and collecting nursing care records at nursing care facilities as well as activity labels and sensors through smartphones, and describe experiments at a nursing care facility for 4 months. A system designed to be used even by staff not familiar with smartphones could collected enough number of data without losing but improving their workload for recording. For collected data, we revealed the nature of the collected data as for activities, care details, and timestamps, and considering them, we show a reference accuracy of recognition of nursing activity which is durable to time skewness, overlaps, and class imbalances. Moreover, we demonstrate the near future prediction to predict the next day's activities from the previous day's records which could be useful for proactive care management. The dataset collected is to be opened to the research community, and can be the utilized for activity recognition and data mining in care facilities.

In this paper, we address Zero-shot learning for sensor activity recognition using word embeddings. The goal of Zero-shot learning is to estimate an unknown activity class (i.e., an activity that does not exist in a given training dataset) by learning to recognize components of activities expressed in semantic vectors. The existing zero-shot methods use mainly 2 kinds of representation as semantic vectors, attribute vector and embedding word vector. However, few zero-shot activity recognition methods based on embedding vector have been studied; especially for sensor-based activity recognition, no such studies exist, to the best of our knowledge. In this paper, we compare and thoroughly evaluate the Zero-shot method with different semantic vectors: (1) attribute vector, (2) embedding vector, and (3) expanded embedding vector and analyze their correlation to performance. Our results indicate that the performance of the three spaces is similar but the use of word embedding leads to a more efficient method, since this type of semantic vector can be generated automatically. Moreover, our suggested method achieved higher accuracy than attribute-vector methods, in cases when there exist similar information in both the given sensor data and in the semantic vector; the results of this study help select suitable classes and sensor data to build a training dataset.