Recent advances in Internet applications have facilitated information spreading and, thanks to a wide variety of mobile devices and the burgeoning 5G networks, users easily and quickly gain access to information. Great amounts of digital information moreover have contributed to the emergence of recommender systems that help to filter information. When the rise of mobile networks has pushed forward the growth of social media networks and users get used to posting whatever they do and wherever they visit on the Web, such quick social media updates already make it difficult for users to find historical data. For this reason, this paper presents a social network-based recommender system. Our purpose is to build a user-centered recommender system to exclude the products that users are disinterested in according to user preferences and their friends' shopping experiences so as to make recommendations effective. Since there might be no corresponding reference value for new products or services, we use indirect relations between friends and “friends’ friends” as well as sentinel friends to improve the recommendation accuracy. The simulation result has proven that our proposed mechanism is efficient in enhancing recommendation accuracy.

To predict rainfall, our proposed model architecture combines the Convolutional Neural Network (CNN), which uses the ResNet-152 pre-training model, with the Recurrent Neural Network (RNN), which uses the Long Short-term Memory Network (LSTM) layer, for model training. By encoding the cloud images through CNN, we extract the image feature vectors in the training process and train the vectors and meteorological data as the input of RNN. After training, the accuracy of the prediction model can reach up to 82%. The result has proven not only the outperformance of our proposed rainfall prediction method in terms of cost and prediction time, but also its accuracy and feasibility compared with general prediction methods.

No abstract

No abstract

This study develops an automated crop growth detection APP, with the functionality to access the cadastral data for the target field, that was to be used for a satellite-imagery-based field survey. A total of 735 ground-truth records of the cabbage cultivation areas in Yunlin were collected via the implemented APP in order to train a deep learning model to make accurate predictions of the growth stages of the cabbage from 0 to 70 days. A regression analysis was performed by the gradient boosting decision tree (GBDT) technique. The model was trained on multitemporal multispectral satellite images, which were retrieved from the ground-truth data. The experimental results show that the mean average error of the predictions is 8.17 days, and that 75% of the predictions have errors less than 11 days. Moreover, the GBDT algorithm was also adopted for the classification analysis. After planting, the cabbage growth stages can be divided into the cupping, early heading, and mature stages. For each stage, the prediction capture rate is 0.73, 0.51, and 0.74, respectively. If the days of growth of the cabbages are partitioned into two groups, the prediction capture rate for 0–40 days is 0.83, and that for 40–70 days is 0.76. Therefore, by applying appropriate data mining techniques, together with multitemporal multispectral satellite images, the proposed method can predict the growth stages of the cabbage automatically, which can assist the governmental agriculture department to make cabbage yield predictions when creating precautionary measures to deal with the imbalance between production and sales when needed.