With the explosive growth of short texts on users' preferences, learning discriminative and coherent latent topics from short texts is a critical work, since many practical applications require semantic understandings that short texts convey explicitly and implicitly. However, existing short text topic learning methods face the challenge of fully capturing semantically related co-occurrence phrases. Therefore, this paper proposes a novel Heterogeneous Information Network-based Short Text Topic learning approach (HIN-ShoTT) in terms of parts of speech, without depending on any auxiliary information. Specifically, HIN-ShoTT can be decomposed into three phases: i) seeking semantic relations among words, where HIN-ShoTT models multiple semantic relations among words based on a Heterogeneous Information Network (HIN) in terms of parts of speech; ii) extracting co-occurrence phrases and filtering noises, where HIN-ShoTT defines parts-of-speech meta structures to guide co-occurrence phrase extraction and a self-adapting threshold filtering module is proposed for discarding noises; and iii) inferring topics, where HIN-ShoTT models the generative process of co-occurrence phrases to make topic learning effective with the abundant corpus-level information. Our experimental results on three real-world datasets not only show that HIN-ShoTT performs well, but also demonstrate that it is feasible to incorporate HIN into short text topic learning for accuracy improvement.
Combining various sources to create a complementary system plays a key role in utilizing clean energy sources economically and mitigating air pollution during the heating season in Northern China. However, there is a lack of unified and reasonable design methods for such systems, resulting in the excessive capacity of equipment and the waste of energy. In this work, a unified design method is proposed to solve this problem. A generalized structure and its mathematical model are firstly established, enabling transient simulations on the TRNSYS platform. Then, a preliminary screening criterion for the system composition a general operation strategy is proposed. Finally, the system configuration is optimized by using the genetic algorithm. The method is successfully applied in a demonstration project in China. The results show that the coupling system consisting of a biomass boiler (384 kW), an air-source heat pump (430 kW) and a ground-source heat pump (369 kW) is the most economical, and the annual cost is 26.7% lower than that of a single-equipment system. Additionally, the sensitive factors that strongly affect the optimization results are explored. The establishment of the generalized structure and its mathematical model enables the quick calculation and convenient comparison of various schemes, and simplifies the complicated optimization problem of the capacity optimization of each piece of equipment. The proposed design method can reduce the annual cost to a minimum value, and thus it provides a theoretical basis for the large-scale application of clean energy sources for heating.
As one kind of remote sensing image (RSI), Directional Polarimetric Camera (DPC) data are of great significance in atmospheric radiation transfer and climate feedback. The availability of DPC images is often hindered by clouds, and effective cloud detection is the premise of many applications. Conventional threshold-based cloud detection methods are limited in performance and generalization capability. In this paper, we propose an effective learning-based 3D multimodal fusion cloud detection network (MFCD-Net) model. The network is a three-input stream architecture with a 3D-Unet-like encoder-decoder structure to fuse the multiple modalities of reflectance image, polarization image Q, and polarization image U in DPC imagery, with consideration of the angle and spectral information. Furthermore, cross attention is utilized in fusing the polarization features into the spatial-angle-spectral features in the reflectance image to enhance the expression of the fused features. The dataset used in this paper is obtained from the DPC cloud product and the cloud mask product. The proposed MFCD-Net achieved excellent cloud detection performance, with a recognition accuracy of 95.74%, according to the results of the experiments.