Search

Abstract The cold-start problem in recommender systems has been facing a great challenge. Cross-domain recommendation can improve the performance of cold-start user recommendations in the target domain by using the rich information of users in the source domain. In cross-domain cold-start recommendation, users in target domain lack sufficient historical behaviors. Existing meta-learning-based methods depend on the feature distribution of training data and limit the adaptability in new tasks. To address these issues, we propose a meta-adversarial framework for cross-domain cold-start recommendation (MAFCDR) . Specifically, we employ a multi-level feature attention mechanism for independently learning the weights of long-term and short-term features to construct preferences of users in source domain. To migrate user representations, we train a meta-adversarial network that utilizes feature embeddings in the source domain as input and enhances the robustness and stability of the model. Then, the personalized bridge function transfers the user preferences in the source domain to the target domain. We build three cross-domain tasks using Amazon dataset and conduct extensive experiments, which demonstrate the effectiveness of the proposed model in cold-start user recommendation. more...

Fiber metal laminates (FMLs) are a novel lightweight composite material, predominantly utilized in the aerospace sector for large-scale components like skin panels and fuselages. However, research on FMLs in the microsystem domain remains limited. Additionally, they are influenced by scale effects, rendering macroscopic forming theories inadequate for microforming applications. The application of ultra-thin fiber metal laminates in the microsystem field is hindered by this constraint. This paper investigates the friction performance of ultra-thin TA1 titanium alloy/carbon fiber-reinforced plastic (CFRP) laminates at the microscale. The content of the epoxy resin used is 38.0 ± 3.0%. Friction tests on ultra-thin TA1/CFRP laminates were conducted based on the Striebeck friction theory model. The effects of factors such as the weaving method, ply angle, normal force, tensile speed, and temperature on friction performance are explored in the study. Furthermore, the influences of geometric scale and grain scale on friction performance are examined. Geometric scale effects indicate that an increase in laminate width leads to an increase in the friction coefficient. Grain-scale effects demonstrate that as grain size increases, the friction coefficient also increases, attributed to reduced grain boundaries, increased twinning, and increased surface roughness of the metal. Finally, surface morphology analysis of the metal and fiber after friction tests further confirms the influence of grain size on the friction coefficient. Through detailed experimental design, result analysis and graphical representation, this paper provides a scientific basis for understanding and predicting the friction behavior of ultra-thin TA1/CFRP laminates. more...

Abstract Objective To determine the impact of preoperative stent placement on postradiotherapy stricture rate in patients with cervical cancer after radical resection. Methods This study was a retrospective analysis of data collected from 55 cervical cancer patients treated with radiotherapy between June 2016 and June 2020. Patients were divided into the stent and control groups. After 3 months, the stricture rate and the complications related to stent placement between the two groups were compared. Results There were 12 (46.2%) and 10 (34.5%) cases of ureteral stricture in the stent (n = 26) and control (n = 29) groups, respectively, three months after the end of radiotherapy. The incidence rates of ureter stricture in the two groups were not significantly different (P = 0.378). Moreover, there were 20 units (38.5%) and 15 units (25.9%) ureteral strictures in the stent and control groups, respectively. No significant difference in the incidence rates of ureteral strictures was found between the two groups (P = 0.157). There were 13 (50.0%) and 10 (34.5%) cases of ureteral stricture in the stent (n = 26) and control (n = 29) groups, respectively, six months after the end of the radiotherapy. The incidence rates of ureter stricture in the two groups were not significantly different (P = 0.244). Moreover, there were 21 units (40.4%) and 15 units (25.9%) ureteral strictures in the stent and control groups, respectively. No significant difference in the incidence rates of ureteral strictures was found between the two groups (P = 0.105). Complications related to stent placement such as urinary tract infections and bladder irritation were statistically significant (P = 0.006 and P = 0.036) between the two groups; while the other complications were not significantly different (P = 0.070, P = 0.092 and P = 0.586). Conclusions Ureteral stents may not reduce the incidence of ureteral stricture after radiotherapy in patients with cervical cancer. The stent needs to be replaced regularly, and the complications related to stent placement may occur at any time. Thus, preoperative stent placement should be cautious for the clinical management of cervical cancer patients treated with postoperative radiotherapy. more...

We reported an 85-year-old patient with malignant glomus tumor (GT) of the prostate. He presented with urinary frequency for more than 2 years and gross hematuria for 7 days. Computed tomography scan showed that the prostate was markedly irregularly enlarged, and the boundary between the prostate and the posterior wall of the bladder was unclear. Bilateral kidneys and ureters were dilated. Biochemical examinations showed that the serum potassium was 7.24 mmol/L and the serum creatinine was 974.6 μmol/L. Transurethral diagnostic resection was performed after restoring homeostasis through several times of bedside blood filtration. The pathological diagnosis was malignant GT. The patient’s renal function recovered after bilateral nephrostomy, and he refused further treatment and was out of contact after 9 months. We summarize the clinical and histopathological features of malignant GT of the prostate in order to improve the early recognition of the disease by clinicians. more...

ABSTRACT Microsporidia, as unicellular eukaryotes, also have an endomembrane system for transporting proteins, which is essentially similar to those of other eukaryotes. In eukaryotes, coat protein complex II (COPII) consists of Sar1, Sec23, Sec24, Sec13, and Sec31 and mediates protein transport from the endoplasmic reticulum (ER) to the Golgi apparatus. Sar1 is the central player in the regulation of coat protein complex II vesicle formation in the endoplasmic reticulum. In this study, we successfully cloned the NbSar1, NbSec23-1, NbSec23-2, NbSec24-1, NbSec24-2, NbSec13, NbSec31-1, and NbSec31-2 genes and prepared NbSar1 polyclonal antibody. We found that NbSar1 was localized mainly in the perinuclear cytoplasm of Nosema bombycis by immunofluorescence analysis (IFA). Yeast two-hybrid assays demonstrated that NbSar1 interacts with NbSec23-2, NbSec23-2 interacts with NbSec24-1 or NbSec24-2, NbSec23-1 interacts with NbSec31, and NbSec31 interacts with NbSec13. Moreover, the silencing of NbSar1 by RNA interference resulted in the aberrant expression of NbSar1, NbSec23-1, NbSec24-1, NbSec24-2, NbSec13, NbSec31-1, and NbSec31-2 and significantly inhibited the proliferation of N. bombycis. Altogether, these findings indicated that the subunits of coat protein complex II work together to perform functions in the proliferation of N. bombycis and that NbSar1 may play a crucial role in coat protein complex II vesicle formation. IMPORTANCE As eukaryotes, microsporidia have retained the endomembrane system for transporting and sorting proteins throughout their evolution. Whether the microsporidia form coat protein complex II (COPII) vesicles to transport cargo proteins and whether they play other roles besides cargo transport are not fully explained at present. Our results showed that NbSar1, NbSec23-1/NbSec23-2, NbSec24-1/NbSec24-2, NbSec13, and NbSec31 might be assembled to form COPII in the ER of N. bombycis, and the functions of COPII are also closely related to the proliferation of N. bombycis, this may be a new target for the prevention of pébrine disease of the silkworm. more...

In the forming process of fiber-reinforced metal laminates (FMLs) product, the exploratory compound forming technology, including hot stamping of aluminum alloy and laying process of fiber prepreg which is named HFQ-FMLs was proposed to solve the puzzles such as weak rigidity, low strength and integration deformation with large difficulty, and the feasibility and mechanical properties of the innovative forming process were studied. Firstly, based on the modified metal volume fraction formula, the theoretical values of the mechanical properties of the HFQ-FMLs plates were calculated. Compared with the experimental results, the minimum error is 1%, proving that the HFQ-FMLs technology scheme is feasible. Secondly, three kinds of metal sheets with different heat treatments and specimens by HFQ-FMLs were carried out for the tensile tests, the mechanical properties distributions were demonstrated, and the influence regularity of the strain rate and rolling direction on the stress analysis was considered at the same time. As can be seen from the distribution of yield strength and tensile strength, the yield stress of metal sheets obtained by HFQ-FMLs technology along the 45° is superior to the raw material and can increase by 46% under strain rate = 0.01 s−1. While, because the vacuum thermal curing treatment makes the aluminum alloy happen double aging, the metal sheet strength dropped, and the jointing strength between the metal and fiber prepreg became weak too, which made the strength limit of the new material improve weakly. Thirdly, the fractured style of the FMLs under different conditions was studied qualitatively. It is helpful to achieve the development rule of defects, optimize the craft route, and avoid deformation failure. more...

Structures with negative Poisson’s ratios have wide applications in automotive design, owing to their excellent mechanical properties and good energy absorption capability. This type of structure can be used to design lightweight components. We used a double-arrow negative Poisson ratio structure and embedded lock assembly process to design a 3D lattice structure with a negative Poisson ratio. Moreover, we selected different design parameters for a simulation analysis and experimental research. The performances of the lattice structures with different parameters were compared and analyzed under different test conditions, including out-of-plane compression, three-point bending, and low-speed impact. The findings reported here pave the way for the development of new 3D auxetic materials with significantly expanded design space and potential applications by reducing the difficulty of the manufacturing process. more...

Along with the growth of graph neural networks (GNNs), many researchers have adopted metapath-based GNNs to handle complex heterogeneous graph embedding. The conventional definition of a metapath only distinguishes whether there is a connection between nodes in the network schema, where the type of edge is ignored. This leads to inaccurate node representation and subsequently results in suboptimal prediction performance. In heterogeneous graphs, a node can be connected by multiple types of edges. In fact, each type of edge represents one kind of scene. The intuition is that if the embedding of nodes is trained under different scenes, the complete representation of nodes can be obtained by organically combining them. In this paper, we propose a novel definition of a metapath whereby the edge type, i.e., the relation between nodes, is integrated into it. A heterogeneous graph can be considered as the compound of multiple relation subgraphs from the view of a novel metapath. In different subgraphs, the embeddings of a node are separately trained by encoding and aggregating the neighbors of the intrapaths, which are the instance levels of a novel metapath. Then, the final embedding of the node is obtained by the use of the attention mechanism which aggregates nodes from the interpaths, which is the semantic level of the novel metapaths. Link prediction is a downstream task by which to evaluate the effectiveness of the learned embeddings. We conduct extensive experiments on three real-world heterogeneous graph datasets for link prediction. The empirical results show that the proposed model outperforms the state-of-the-art baselines; in particular, when comparing it to the best baseline, the F1 metric is increased by 10.35% over an Alibaba dataset. more...

Abstract The binary coding technique has been widely used in approximate nearest neighbors (ANN) search tasks. Traditional hashing algorithms treat binary bits equally, which usually causes an ambiguous ranking. To solve this issue, we propose an innovative bitwise weight method dubbed minimal residual ordinal loss hashing (MROLH). Different from a two-step mechanism, MROLH simultaneously learns binary codes and bitwise weights by a feedback mechanism. When the algorithm converges, the binary codes and bitwise weights can be well adaptive to each other. Furthermore, we establish the ordinal relation preserving constraint based on quartic samples to enhance the power of preserving relative similarity. To decrease the training complexity, we utilize a tensor ordinal graph to represent quartic ordinal relation, and the original objective function is approximated by the one based on triplet samples. In this paper, we also assign different weight values to training samples. During the training procedure, the weight of each data is initialized to the same value, and we iteratively boost the weight of the data whose relative similarity is not well preserved. As a result, we can minimize the residual ordinal loss. Experimental results on three large-scale ANN search benchmark datasets, i.e., SIFT1M, GIST1M, and Cifar10, show that the proposed method MROLH achieves a superior ANN search performance in both the Hamming space and the weighted Hamming space over the sate-of-the-art approaches. more...

The cross-modal retrieval task can return different modal nearest neighbors, such as image or text. However, inconsistent distribution and diverse representation make it hard to directly measure the similarity relationship between different modal samples, which causes a heterogeneity gap. To bridge the above-mentioned gap, we propose the deep adversarial learning triplet similarity preserving cross-modal retrieval algorithm to map different modal samples into the common space, allowing their feature representation to preserve both the original inter- and intra-modal semantic similarity relationship. During the training process, we employ GANs, which has advantages in modeling data distribution and learning discriminative representation, in order to learn different modal features. As a result, it can align different modal feature distributions. Generally, many cross-modal retrieval algorithms only preserve the inter-modal similarity relationship, which makes the nearest neighbor retrieval results vulnerable to noise. In contrast, we establish the triplet similarity preserving function to simultaneously preserve the inter- and intra-modal similarity relationship in the common space and in each modal space, respectively. Thus, the proposed algorithm has a strong robustness to noise. In each modal space, to ensure that the generated features have the same semantic information as the sample labels, we establish a linear classifier and require that the generated features’ classification results be consistent with the sample labels. We conducted cross-modal retrieval comparative experiments on two widely used benchmark datasets—Pascal Sentence and Wikipedia. For the image to text task, our proposed method improved the mAP values by 1% and 0.7% on the Pascal sentence and Wikipedia datasets, respectively. Correspondingly, the proposed method separately improved the mAP values of the text to image performance by 0.6% and 0.8% on the Pascal sentence and Wikipedia datasets, respectively. The experimental results show that the proposed algorithm is better than the other state-of-the-art methods. more...

The glass fiber-reinforced metal laminates (GLARE) cannot be used to form complex laminate structures in the aerospace industry, because there is substantial variation in the plasticity of the heterogeneous materials. Hence, a compound process for composite materials based on the thermoforming technology for aluminum alloy and fiber-reinforced metal laminates (FMLs)-forming technology was proposed; it contains solution heat treatment, thermoforming, quick cold die quenching, artificial aging integrated process (HFQ), and the thermal consolidation of fiber-reinforced metal laminates, and it is named the HFQ-FMLs forming process. In order to test and judge the effect of the heat treatment on the properties of the materials obtained by the new technology, the pure metal sheet and the three kinds of HFQ-FMLs composite laminates fabricated with the different layup method were assessed with the Vickers hardness test and the Charpy impact test at the same time, and they were labeled #1, #2, #3, #4, respectively. In the Charpy impact test, in order to obtain accurate data, the shape and fixing position of the specimen was optimized so that the gap direction was parallel to the loading direction. After the heat treatment, the properties of the aluminum alloy were improved, the hardness will affect the energy absorption of the laminates, and the relationship between the thickness, hardness and impact properties will be analyzed. The hardness test results are 39.9 HV, 37.5 HV, 37.4 HV, 37.1 HV which indicates the pure metal sheet had the greatest hardness, and the greater the thickness of fiber layer, the lower was the hardness of the HFQ-FMLs composite laminate. The impact resistance of the HFQ-FMLs composite laminates was about two times of the pure metal sheet for the same thickness, and the values are 2.3 J, 4.8 J, 4.8 J, 4.8 J, respectively. In addition, the method of laying the fiber layer had no effect on the impact resistance of the composite laminates. For the novel composite laminates subjected to different cutting processes, the scanning electron microscope (SEM) results for the incision morphology suggest that the water cutting process ensures the structural integrity of the composite laminates after the edges’ and holes’ cutting procedures, and the performance maintains continuity. more...

In recent years, binary coding methods have become increasingly popular for tasks of searching approximate nearest neighbors (ANNs). High-dimensional data can be quantized into binary codes to give an efficient similarity approximation via a Hamming distance. However, most of existing schemes consider the importance of each binary bit as the same and treat training samples at different positions equally, which causes many data pairs to share the same Hamming distance and a larger retrieval loss at the top position. To handle these problems, we propose a novel method dubbed by the top-position-sensitive ordinal-relation-preserving bitwise weight (TORBW) method. The core idea is to penalize data points without preserving an ordinal relation at the top position of a ranking list more than those at the bottom and assign different weight values to their binary bits according to the distribution of query data. Specifically, we design an iterative optimization mechanism to simultaneously learn binary codes and bitwise weights, which makes their learning processes related to each other. When the iterative procedure converges, the binary codes and bitwise weights are effectively adapted to each other. To reduce the training complexity, we relax the discrete constraints of both the binary codes and the indicator function. Furthermore, we pretrain a tensor ordinal graph to decrease the time consumption of computing a relative similarity relationship among data points. Experimental results on three large-scale ANN search benchmark datasets, i.e., SIFT1M, GIST1M, and Cifar10, show that the proposed TORBW method can achieve superior performance over state-of-the-art approaches. more...

Silkworm's fungal infection is a kind of communicable diseases caused by parasitic fungi. In this study, a rare fungal Bm1 strain was isolated from infected silkworms in cocoon production in Zhenjiang city, China. The result from morphological investigation showed that the conidiogenous cells of the isolated fungal strain had flask-like phialides with slightly swollen or columnar bases. Most of the conidia were oval and arranged in a broadly 'V'-like manner that resulted in irregularly curved chains. The color of the fungal colony was white at the initial growing stage and gradually became pink. The average hyphal growth rate of the strain was 3.42±1.24 mm/d on the PDA solid medium. Phylogenetic analysis showed that the Bm1 strain was clustered in the same clade with Isaria cateniobliqua, Cordyceps cateniobliqua, and Evlachovaea kintrischica. Combining the morphological characteristics with rRNA-ITS sequence, this newly isolated fungal strain was identified as a C. cateniobliqua strain, named as C. cateniobliqua Bm1. The C. cateniobliqua Bm1 could infect the silkworm through contacting transmission, and the course of this disease was about 5 to 19 d. The mortality rates of the silkworm caused by infection of C. cateniobliqua Bm1 positively increased with the concentration of inoculated conidial suspension. The estimated LC50 value was 1.18×109 conidia/ml, a low pathogenicity to the silkworm compared with the Beauveria bassiana. [ABSTRACT FROM AUTHOR] more...

Purpose: To compare the influence of three operative approaches [transurethral en bloc resection of bladder tumor by pin-shaped electrode (pin-ERBT), transurethral resection of bladder tumor (TURBT) and transurethral holmium laser resection of bladder tumor (HoLRBT)] on the recurrence rate of non-muscle-invasive bladder cancer (NMIBC) at low dimension (i.e. diameter below 3 cm).Materials and Methods: A retrospective analysis was conducted for a total of 115 patients affected by solitary NMIBC, with a diameter <3 cm, who were submitted to operation between March 2013 to May 2017. The patients were divided according to the operative method applied (pin-ERBT, TURBT and HoLRBT groups, respectively). The 2-year recurrence rate was compared among the three groups, and multivariat Cox hazard model analysis was applied to analyze the influencing factor(s) for postoperative recurrence.Results: The 2-year recurrence rate was 10.0% in ERBT, 38.5% in TURBT and 40.0% in HoLRBT group, with a significant difference (P =0.014). According to the Cox hazard model analysis, age(HR=1.058, 95% CI: 1.019~1.098，P=0.003), operative method(HR=2.974,6.508, 95% CI: 0.862~10.255,1.657~25.566, P=0.023), smoking(HR=2.399, 95% CI: 1.147~5.017, P=0.020) and pathological grade(HR=2.012,95% CI: 1.279~3.165, P=0.002) were risk factors for postoperative recurrence of bladder cancer.Conclusion: Pin-ERBT can prominently decrease the postoperative recurrence rate of solitary NMIBC with a diameter <3 cm. [ABSTRACT FROM AUTHOR] more...