On the Study of Transformers for Query Suggestion

When conducting a search task, users may find it difficult to articulate their need, even more so when the task is complex. To help them complete their search, search engine usually provide query suggestions. A good query suggestion system requires to model user behavior during the search session. In this article, we study multiple Transformer architectures applied to the query suggestion task and compare them with recurrent neural network (RNN)-based models. We experiment Transformer models with different tokenizers, with different Encoders (large pretrained models or fully trained ones), and with two kinds of architectures (flat or hierarchic). We study the performance and the behaviors of these various models, and observe that Transformer-based models outperform RNN-based ones. We show that while the hierarchical architectures exhibit very good performances for query suggestion, the flat models are more suitable for complex and long search tasks. Finally, we investigate the flat models behavior and demonstrate that they indeed learn to recover the hierarchy of a search session.

Structure-engineering of core-shell ZnCo2O4@NiO composites for high-performance asymmetric supercapacitors

The implementation of a structure-designed strategy to construct hierarchical architectures of multicomponent metal oxide-based electrode materials for the energy storage device is in limelight. Herein, we report NiO nanoflakes impregnated...

Simple synthesis of self-assembled nacre-like materials with 3D periodic layers from nanochitin via hydrogelation and mineralization

The superb mechanical properties of some natural materials usually result from highly ordered, multiscale and hierarchical architectures such as bone, nacre, exoskeleton etc. Nonetheless, the gene regulated process can not...

UBNet: Deep learning-based approach for automatic X-ray image detection of pneumonia and COVID-19 patients

BACKGROUND: Analysis of chest X-ray images is one of the primary standards in diagnosing patients with COVID-19 and pneumonia, which is faster than using PCR Swab method. However, accuracy of using X-ray images needs to be improved. OBJECTIVE: To develop a new deep learning system of chest X-ray images and evaluate whether it can quickly and accurately detect pneumonia and COVID-19 patients. METHODS: The developed deep learning system (UBNet v3) uses three architectural hierarchies, namely first, to build an architecture containing 7 convolution layers and 3 ANN layers (UBNet v1) to classify between normal images and pneumonia images. Second, using 4 layers of convolution and 3 layers of ANN (UBNet v2) to classify between bacterial and viral pneumonia images. Third, using UBNet v1 to classify between pneumonia virus images and COVID-19 virus infected images. An open-source database with 9,250 chest X-ray images including 3,592 COVID-19 images were used in this study to train and test the developed deep learning models. RESULTS: CNN architecture with a hierarchical scheme developed in UBNet v3 using a simple architecture yielded following performance indices to detect chest X-ray images of COVID-19 patients namely, 99.6%accuracy, 99.7%precision, 99.7%sensitivity, 99.1%specificity, and F1 score of 99.74%. A desktop GUI-based monitoring and classification system supported by a simple CNN architecture can process each chest X-ray image to detect and classify COVID-19 image with an average time of 1.21 seconds. CONCLUSION: Using three hierarchical architectures in UBNet v3 improves system performance in classifying chest X-ray images of pneumonia and COVID-19 patients. A simple architecture also speeds up image processing time.

Methods to Generate Structurally Hierarchical Architectures in Nanoporous Coinage Metals

The fundamental essence of material design towards creating functional materials lies in bringing together the competing aspects of a large specific surface area and rapid transport pathways. The generation of structural hierarchy on distinct and well-defined length scales has successfully solved many problems in porous materials. Important applications of these hierarchical materials in the fields of catalysis and electrochemistry are briefly discussed. This review summarizes the recent advances in the strategies to create a hierarchical bicontinuous morphology in porous metals, focusing mainly on the hierarchical architectures in nanoporous gold. Starting from the traditional dealloying method and subsequently moving to other non-traditional top-down and bottom-up manufacturing processes including templating, 3D printing, and electrodeposition, this review will thoroughly examine the chemistry of creating hierarchical nanoporous gold and other coinage metals. Finally, we conclude with a discussion about the future opportunities for the advancement in the methodologies to create bimodal structures with enhanced sensitivity.

3D Ultralight Hollow NiCo Compound@MXene Composites for Tunable and High-Efficient Microwave Absorption

The 3D hollow hierarchical architectures tend to be designed for inhibiting stack of MXene flakes to obtain satisfactory lightweight, high-efficient and broadband absorbers. Herein, the hollow NiCo compound@MXene networks were prepared by etching the ZIF 67 template and subsequently anchoring the Ti3C2Tx nanosheets through electrostatic self-assembly. The electromagnetic parameters and microwave absorption property can be distinctly or slightly regulated by adjusting the filler loading and decoration of Ti3C2Tx nanoflakes. Based on the synergistic effects of multi-components and special well-constructed structure, NiCo layered double hydroxides@Ti3C2Tx (LDHT-9) absorber remarkably achieves unexpected effective absorption bandwidth (EAB) of 6.72 GHz with a thickness of 2.10 mm, covering the entire Ku-band. After calcination, transition metal oxide@Ti3C2Tx (TMOT-21) absorber near the percolation threshold possesses minimum reflection loss (RLmin) value of − 67.22 dB at 1.70 mm within a filler loading of only 5 wt%. This work enlightens a simple strategy for constructing MXene-based composites to achieve high-efficient microwave absorbents with lightweight and tunable EAB."Image missing"