A Multi-task Learning Model for Daily Activity Forecast in Smart Home

Abstract

Daily activity forecasts play an important role in the daily lives of residents in smart homes. Category forecasts and occurrence time forecasts of daily activity are two key tasks. Category forecasts of daily activity are correlated with occurrence time forecasts, however, existing research has only focused on one of the two tasks. Moreover, the performance of daily activity forecasts is low when the two tasks are performed in series. In this paper, a forecast model based on multi-task learning is proposed to forecast category and occurrence time of daily activity mutually and iteratively. Firstly, raw sensor events are pre-processed to form a feature space of daily activity. Secondly, a parallel multi-task learning model which combines a convolutional neural network (CNN) with bidirectional long short-term memory (Bi-LSTM) units are developed as the forecast model. Finally, five distinct datasets are used to evaluate the proposed model. The experimental results show that compared with the state-of-the-art single-task learning models, this model improves accuracy by at least 2.22%, and the metrics of NMAE, NRMSE and R2 are improved by at least 1.542%, 7.79% and 1.69%, respectively.

Using Knowledge Graphs for Machine Learning in Smart Home Forecasters

“…An example of a prediction algorithm in a smart home is a forecaster that predicts the future temperature of a room based on the current temperature of the room, the temperature of different rooms in the same house and the brightness of the sun on specific walls among other factors [16]. Other possible predictions include category and occurrence time forecasting of daily activity of occupants [21] and energy usage [12].…”

“…Multiple forecasters have been created before using machine learning models such as: forward stepwise linear resolution [16], multi-layer perceptron [12], convolutional neural network [21] among other methods.…”

Fault Diagnosis for Rolling Bearings Using Optimized Variational Mode Decomposition and Resonance Demodulation

“…The short-time Fourier transform is proposed to effectively identify the frequency and phase of sine wave in the local area of non-stationary signal, and its optimization processing [ 12 , 13 , 14 , 15 ]. Wavelet transform is a new transform analysis method, which inherits and develops the localization idea of short-time Fourier transform, and overcomes the shortcomings that the window size does not change with frequency [ 16 , 17 , 18 ].…”

Innovation Mode and Optimization Strategy of B2C E-Commerce Logistics Distribution under Big Data

“…Zhou et al [39] designed a B2C e-commerce management module. In addition, for logistics distribution, many researchers have proposed a lot of optimizations method to obtain the optimal logistics distribution schemes [40][41][42][43][44][45][46][47][48][49][50][51][52][53].…”