Sentiment analysis (SA) is an important research area in cognitive computation—thus, in-depth studies of patterns of sentiment analysis are necessary. At present, rich-resource data-based SA has been well-developed, while the more challenging and practical multi-source unsupervised SA (i.e., a target-domain SA by transferring from multiple source domains) is seldom studied. The challenges behind this problem mainly locate in the lack of supervision information, the semantic gaps among domains (i.e., domain shifts), and the loss of knowledge. However, existing methods either lack the distinguishable capacity of the semantic gaps among domains or lose private knowledge. To alleviate these problems, we propose a two-stage domain adaptation framework. In the first stage, a multi-task methodology-based shared-private architecture is employed to explicitly model the domain-common features and the domain-specific features for the labeled source domains. In the second stage, two elaborate mechanisms are embedded in the shared-private architecture to transfer knowledge from multiple source domains. The first mechanism is a selective domain adaptation (SDA) method, which transfers knowledge from the closest source domain. And the second mechanism is a target-oriented ensemble (TOE) method, in which knowledge is transferred through a well-designed ensemble method. Extensive experiment evaluations verify that the performance of the proposed framework outperforms unsupervised state-of-the-art competitors. What can be concluded from the experiments is that transferring from very different distributed source domains may degrade the target-domain performance, and it is crucial to choose proper source domains to transfer from.
Brain Imaging and Behavior - Postmortem studies on patients with Alzheimer’s disease (AD) have confirmed that the dorsal raphe nucleus (DRN) in the brainstem is the first brain structure... 相似文献
Four conjugated D‐π‐A copolymers, P1 – P4 , based on 11,12‐difluorodibenzo[a,c]phenazine as an acceptor unit, benzodithiophene (BDT) with an alkoxy side chain ( P1 and P2 ) or thiophene‐conjugated side chain ( P3 and P4 ) as a donor unit, and thiophene π‐bridges, are designed and synthesized to investigate the effect of the side‐chain nature and position on the photovoltaic performance of the conjugated polymer donor materials in polymer solar cells (PSCs). It is found that the copolymers of P3 and P4 with thiophene conjugated side chains on the BDT unit demonstrate broader absorption and better photovoltaic performance than that of P1 and P2 with alkoxy side chains on the BDT unit. The photovoltaic performance of P1 and P3 with an alkyl side chain on the thiophene π‐bridge close to the acceptor unit is improved over that of P2 and P4 with an alkyl side chain on the thiophene π‐bridge close to the donor unit, due to the lower steric hindrance of P1 and P3 . The results indicate that side‐chain engineering is very important for improving the photovoltaic performance of conjugated polymer donor materials.
Endothelial progenitor cells (EPCs) are defined as a special type of stem cell that have been found to directly incorporate into injured vessels and that participate in angiogenesis and reconstruction by differentiation into endothelial cells. EPCs are widely used to therapeutically treat cardiovascular disease, limb ischemia and vascular repair. However, the role of EPCs in inflammatory diseases, especially in lung injury, is less studied.
Objective
To investigate the application of EPCs to vascular repair, and the role of EPCs in acute lung injury (ALI) and acute respiratory distress syndrome (ARDS).
Methods
A computer-based online search was performed in the PubMed database and Web of Science database for articles published, concerning EPCs, angiogenesis, ALI/ARDS and stem cell transplantation
Conclusion
EPCs have a therapeutic potential for vascular regeneration and may emerge as novel strategy for the diseases that are associated with ALI/ARDS. 相似文献
Due to the shortage of freshwater, seawater has been widely considered for mineral flotation. However, the presence of Mg2+ in seawater plays an apparently negative role. In this work, two dispersants (i.e., sodium silicate (SS) and sodium hexametaphosphate (SH)) were applied to reduce the detrimental effects of Mg2+ on the flotation of molybdenite (MoS2). Various measurements including contact angle, zeta potential, FTIR and XPS were carried out to understand the impacts of these two dispersants on MoS2 flotation. Results indicate that both dispersants prevented the adsorption of colloidal Mg(OH)2 onto MoS2 surface under alkaline conditions, thereby improving MoS2 floatability. In addition, both dispersants are physically adsorbed on MoS2 surface, but chemically adsorbed on Mg(OH)2 surface. In addition, the extended Derjaguin–Landau–Verwey–Overbeek (DLVO) calculation suggests that both SS and SH reverse the total interaction energies between MoS2 and colloidal Mg(OH)2 from negative (attraction force) to positive (repulsive force), with the impact of SH being more significant.Due to the shortage of freshwater, seawater has been widely considered for mineral flotation.相似文献