Search

With the support of advanced hardware and software technology, Artificial Intelligence (AI) techniques, especially the increasing number of deep learning algorithms, have spawned the popularization of online intelligent services and accelerated the contemporary development and applications of chatbot systems. The promise of providing 24/7 uninterrupted business services and minimizing workforce costs has made business chatbots a hot topic due to the impact of the pandemic. It has attracted considerable attention from academic researchers and business practitioners. However, a thorough technical review of advanced chatbot technologies and their relevance and applications to various business domains is rare in the literature. The main contribution of this review article is the critical analysis of various chatbot development approaches and the underlying deep learning computational methods in the context of some business applications. We first conceptualize current business chatbot architectures and illustrate the technical characteristics of two common structures. Next, we explore the mainstream deep learning technologies in chatbot design from the perspective of computational methods and usages. Then, we propose a new framework to classify chatbot construction architectures and differentiate the traditional retrieval-based and generation-based chatbots in terms of the modern pipeline and end-to-end structures. Finally, we highlight future research directions for business chatbots to enable researchers to devote their efforts to the most promising research topics and commercial scenarios and for practitioners to benefit from realizing the trend in business chatbot development and applications. [ABSTRACT FROM AUTHOR]

The single cleavage drilled compression specimens of sandstone were impacted by the large-diameter split Hopkinson pressure bar, during which the whole model-I dynamical fracture process was successfully observed. A crack propagation gauge is used to monitor the key time moment of dynamic initiation, propagation, arrest and re-initiation, respectively. The fractal crack extension model is used to analysis the propagation speed of the tortuous crack, and with further combination of the experimental–numerical-analytical method, to determine the dynamic initiation toughness, dynamic propagation toughness, dynamic arrest toughness, and dynamic re-initiation toughness of sandstone. The results show that in the process of crack propagation, the crack propagation path is torturous; and for this curved path, the value of the universal function, which is characterized by the crack's velocity, is smaller than that with a straight path. The dynamic propagation toughness thus obtained is closer to its real value by using the fractal model. Sandstone's dynamic initiation toughness is greater than the dynamic arrest toughness, and the dynamic initiation toughness is slightly bigger than the dynamic re-initiation toughness. [ABSTRACT FROM AUTHOR]

This study proposes a bi-level framework for real-time crash risk forecasting (RTCF) for signalised intersections, leveraging the temporal dependency among crash risks of contiguous time slices. At the first level of RTCF, a non-stationary generalised extreme value (GEV) model is developed to estimate the rear-end crash risk in real time (i.e., at a signal cycle level). Artificial intelligence techniques, like YOLO and DeepSort were used to extract traffic conflicts and time-varying covariates from traffic movement videos at three signalised intersections in Queensland, Australia. The estimated crash frequency from the non-stationary GEV model is compared against the historical crashes for the study locations (serving as ground truth), and the results indicate a close match between the estimated and observed crashes. Notably, the estimated mean crashes lie within the confidence intervals of observed crashes, further demonstrating the accuracy of the extreme value model. At the second level of RTCF, the estimated signal cycle crash risk is fed to a recurrent neural network to predict the crash risk of the subsequent signal cycles. Results reveal that the model can reasonably estimate crash risk for the next 20–25 min. The RTCF framework provides new pathways for proactive safety management at signalised intersections. [ABSTRACT FROM AUTHOR]

Autoencoders have become a hot researched topic in unsupervised learning due to their ability to learn data features and act as a dimensionality reduction method. With rapid evolution of autoencoder methods, there has yet to be a complete study that provides a full autoencoders roadmap for both stimulating technical improvements and orienting research newbies to autoencoders. In this paper, we present a comprehensive survey of autoencoders, starting with an explanation of the principle of conventional autoencoder and their primary development process. We then provide a taxonomy of autoencoders based on their structures and principles and thoroughly analyze and discuss the related models. Furthermore, we review the applications of autoencoders in various fields, including machine vision, natural language processing, complex network, recommender system, speech process, anomaly detection, and others. Lastly, we summarize the limitations of current autoencoder algorithms and discuss the future directions of the field. [ABSTRACT FROM AUTHOR]

Attributed graph clustering, the task of grouping nodes into communities using both graph structure and node attributes, is a fundamental problem in graph analysis. Recent approaches have utilized deep learning for node embedding followed by conventional clustering methods. However, these methods often suffer from the limitations of relying on the original network structure, which may be inadequate for clustering due to sparsity and noise, and using separate approaches that yield suboptimal embeddings for clustering. To address these limitations, we propose a novel method called Deep Attributed Clustering with High-order Proximity Preserve (DAC-HPP) for attributed graph clustering. DAC-HPP leverages an end-to-end deep clustering framework that integrates high-order proximities and fosters structural cohesiveness and attribute homogeneity. We introduce a modified Random Walk with Restart that captures k-order structural and attribute information, enabling the modelling of interactions between network structure and high-order proximities. A consensus matrix representation is constructed by combining diverse proximity measures, and a deep joint clustering approach is employed to leverage the complementary strengths of embedding and clustering. In summary, DAC-HPP offers a unique solution for attributed graph clustering by incorporating high-order proximities and employing an end-to-end deep clustering framework. Extensive experiments demonstrate its effectiveness, showcasing its superiority over existing methods. Evaluation on synthetic and real networks demonstrates that DAC-HPP outperforms seven state-of-the-art approaches, confirming its potential for advancing attributed graph clustering research. [ABSTRACT FROM AUTHOR]

Some thiazolamide derivatives containing diphenyl ether moiety were designed, synthesized, and studied for their antifungal activity and cell cytotoxicity. Of these, several compounds II possessed excellent activity against Rhizoctonia solani, which was better than that of boscalid and comparable to thifluzamide. Besides, the cytotoxicity test showed that the optimal compound IIf had lower cytotoxicity than thifluzamide in general. Further, molecular docking could give a reasonable explanation for the high activity of the target compounds. The above results indicated that thiazolamide derivatives containing diphenyl ether moiety are valuable for in-depth study to obtain high-efficiency and low-toxicity green pesticides. A series of new thiazolamide derivatives containing diphenyl ether moiety were designed and synthesized. The biological evaluation indicated optimal compound IIf displayed not only high fungicidal activity but also lower cytotoxicity. Besides, molecular docking gave a reasonable explanation for the high activity of the target compounds. [ABSTRACT FROM AUTHOR]

Background: A few studies have tested febuxostat for its usefulness in delaying chronic kidney disease (CKD) progression by treating hyperuricemia and results were controversial. Thus, we attempted to conduct a randomized controlled study using the Chinese population with advanced grade of CKD. Methods: One hundred CKD patients in stages 3 and 4 with asymptomatic hyperuricemia from seven medical centers were included in this prospective randomized controlled study and assigned to the control and febuxostat group, the latter of which received febuxostat to titrate to achieve serum uric acid (SUA) < 6 mg/dL. The observation period was 12 months. The primary outcomes included the event of estimated glomerular filtration rate (eGFR) decline ≥ 30% or 50% from baseline at 12 months, dialysis and death from CKD; secondary outcome was the change in eGFR. Safety analysis was also performed. Results: Forty-seven patients and 45 patients in the febuxostat and control groups, respectively completed the study. Seven of 47 (14.9%) participants reached 30% decline in eGFR in the febuxostat group, while 1 (2.1%) and 2 (4.3%) patients reached 50% decline in eGFR or dialysis. Thirteen (28.9%), 10 (22.2%) and 3 (6.7%) of 45 patients reached primary kidney outcomes separately in the control group. The change in eGFR after 12 months from baseline in the febuxostat group was 0.50 mL/min/1.73 m2, which was significantly higher than that in the control group − 4.46 mL/min/1.73 m2 (p = 0.006). Adverse events did not differ between two groups. Conclusions: Febuxostat effectively slowed eGFR decline in patients with CKD stages 3 and 4 and asymptomatic hyperuricemia. [ABSTRACT FROM AUTHOR]

Informatics paradigms for brain and mental health research have seen significant advances in recent years. These developments can largely be attributed to the emergence of new technologies such as machine learning, deep learning, and artificial intelligence. Data-driven methods have the potential to support mental health care by providing more precise and personalised approaches to detection, diagnosis, and treatment of depression. In particular, precision psychiatry is an emerging field that utilises advanced computational techniques to achieve a more individualised approach to mental health care. This survey provides an overview of the ways in which artificial intelligence is currently being used to support precision psychiatry. Advanced algorithms are being used to support all phases of the treatment cycle. These systems have the potential to identify individuals suffering from mental health conditions, allowing them to receive the care they need and tailor treatments to individual patients who are mostly to benefit. Additionally, unsupervised learning techniques are breaking down existing discrete diagnostic categories and highlighting the vast disease heterogeneity observed within depression diagnoses. Artificial intelligence also provides the opportunity to shift towards evidence-based treatment prescription, moving away from existing methods based on group averages. However, our analysis suggests there are several limitations currently inhibiting the progress of data-driven paradigms in care. Significantly, none of the surveyed articles demonstrate empirically improved patient outcomes over existing methods. Furthermore, greater consideration needs to be given to uncertainty quantification, model validation, constructing interdisciplinary teams of researchers, improved access to diverse data and standardised definitions within the field. Empirical validation of computer algorithms via randomised control trials which demonstrate measurable improvement to patient outcomes are the next step in progressing models to clinical implementation. [ABSTRACT FROM AUTHOR]

Autism spectrum disorder (ASD) is a complex neurodevelopmental condition that affects social interaction and behavior. Mutations in the gene encoding chromodomain helicase DNA-binding protein 8 (CHD8) lead to autism symptoms and macrocephaly by a haploinsufficiency mechanism. However, studies of small animal models showed inconsistent findings about the mechanisms for CHD8 deficiency-mediated autism symptoms and macrocephaly. Using the nonhuman primate as a model system, we found that CRISPR/Cas9-mediated CHD8 mutations in the embryos of cynomolgus monkeys led to increased gliogenesis to cause macrocephaly in cynomolgus monkeys. Disrupting CHD8 in the fetal monkey brain prior to gliogenesis increased the number of glial cells in newborn monkeys. Moreover, knocking down CHD8 via CRISPR/Cas9 in organotypic monkey brain slices from newborn monkeys also enhanced the proliferation of glial cells. Our findings suggest that gliogenesis is critical for brain size in primates and that abnormal gliogenesis may contribute to ASD. [ABSTRACT FROM AUTHOR]

Link prediction is one of the most widely studied problems in the area of complex network analysis, in which machine learning techniques can be applied to deal with it. The biggest drawback of the existing methods, however, is that in most cases they only consider the topological structure of the network, and therefore completely miss out on the great potential that stems from the nodal attributes. Both topological structure and nodes' attributes are essential in predicting the evolution of attributed networks and can act as complements to each other. To bring out their full potential in solving the link prediction problem, a novel Robust Graph Regularization Nonnegative Matrix Factorization for Attributed Networks (RGNMF-AN) was proposed, which models not only the topology structure of networks but also their node attributes for direct link prediction. This model, in particular, combines two types of information, namely network topology, and nodal attributes information, and calculates high-order proximities between nodes using the Structure-Attribute Random Walk Similarity (SARWS) method. The SARWS score matrix is an indicator structural and attributed matrix that collects more useful attributed information in high-order proximities, whereas graph regularization technology combines the SARWS score matrix with topological and attribute information to collect more valuable attributed information in high-order proximities. Furthermore, the RGNMF-AN employs the ℓ2,1-norm to constrain the loss function and regularization terms, effectively removing random noise and spurious links. According to empirical findings on nine real-world complex network datasets, the use of a combination of attributed and topological information in tandem enhances the prediction performance significantly compared to the baseline and other NMF-based algorithms. [ABSTRACT FROM AUTHOR]

AAV-delivered CRISPR/Cas9 (AAV-CRISPR) has shown promising potentials in preclinical models to efficiently insert therapeutic gene sequences in somatic tissues. However, the AAV input doses required were prohibitively high and posed serious risk of toxicity. Here, we performed AAV-CRISPR mediated homology-independent knock-in at a new target site in mAlb 3'UTR and demonstrated that single dose of AAVs enabled long-term integration and expression of hF9 transgene in both adult and neonatal hemophilia B mice (mF9 −/−), yielding high levels of circulating human Factor IX (hFIX) and stable hemostasis restoration during entire 48-week observation period. Furthermore, we achieved hemostasis correction with a significantly lower AAV dose (2 × 109 vg/neonate and 1 × 1010 vg/adult mouse) through liver-specific gene knock-in using hyperactive hF9R338L variant. The plasma antibodies against Cas9 and AAV in the neonatal mice receiving low-dose AAV-CRISPR were negligible, which lent support to the development of AAV-CRISPR mediated somatic knock-in for treating inherited diseases. Here, the authors develop an AAV-CRISPR mediated somatic knock-in. They apply this system to restore hemostasis in neonatal hemophilia B mice and show liver-specificity of the knock-in and low serum antibody production. [ABSTRACT FROM AUTHOR]

Objective: The performance and stability of radiomics model caused by dimension reduction remain being confronted with major challenges. In this study, we aimed to propose a new scheme of global feature management independent of dimension reduction to improve it. Methods: The non-contrast computed tomography (NCCT) images of acute brainstem infarction (ABI) from two medical centers were used as test and validation sets. A new scheme was constructed based on global feature management, and the traditional scheme dependent on dimension reduction was used as control. The radiomic features of NCCT images were extracted in Matlab R2013a. The performance of prediction model was evaluated by the generalized linear model (GLM) and multivariate logistic regression. And, the stability of radiomics model was evaluated with the difference of area under curve (AUC) between the test and validation sets. Results: Compared with the traditional scheme, the new scheme presented a similar detection performance (AUC: 0.875 vs. 0.883), yet a better performance in predicting prognosis (AUC: 0.864, OR = 0.917, p = 0.021 vs. AUC:0.806, OR = 0.972, p = 0.007). All these results were well verified in an independent validation set. Moreover, the new scheme showed stronger stability in both the detection model (ΔAUC: 0.013 vs. 0.039) and prediction model (ΔAUC = 0.004 vs. 0.044). Conclusion: Although there might be several limitations, this study proved that the scheme of global feature management independent of dimension reduction could be a powerful supplement to the radiomics methodology. Key Points: • The new scheme (Swavelet) presented similar detection performances for ABI with the traditional scheme. • A better predictive performance for END was found in the new scheme (Swavelet) compared with the traditional scheme. • Stronger model stability was found in both the detection and prediction models based on the new scheme. [ABSTRACT FROM AUTHOR]

Objective: Although the guideline indicates that total kidney volume (TKV) is an important detection indicator in patients with autosomal dominant polycystic kidney disease (ADPKD), this study attempted to demonstrate that renal parenchymal information, combined with parenchymal volume and radiomics features, may have more valuable clinical guiding significance. Methods: A totals of 340 ADPKD patients with normal renal function were prospectively collected and followed-up for five years, with renal function tests and non-contrast computed tomography (CT) performed every six months. The relationship between renal function impairment and renal parenchymal volume (RPV) as along with radiomics features was explored using a multiple linear regression model and multiple logistic regression. Then, a combined model of RPV with radiomics features was constructed to comprehensively evaluate its predictive value. Results: Compared with TKV, decreased RPV presented a closer relationship with renal function impairment, namely, with the impairment rate (RPV: 82.3% vs. TVK: 67.1%) and eGFR (RPV: r = 0.614, p < 0.001 vs. TKV: r = -0.452, p < 0.001), and showed higher predictive power (RPV: AUC = 0.752 [95%CI 0.692–0.805], p < 0.001 vs. TKV: AUC = 0.711 [95%CI 0.649–0.768], p < 0.001). Correspondingly, the radiomics analysis that was derived from the renal parenchyma also showed a satisfactory result (AUC = 0.849 [95%Cl 0.797–0.892], p < 0.001). Importantly, the predictive power for renal function impairment was further improved in the combined model of RPV and radiomics features (AUC = 0.902 [95%Cl 0.857–0.937], p < 0.001). Conclusion: Renal parenchyma information may be a sensitive biomarker of renal function impairment in ADPKD, which can provide a new approach for clinically monitoring renal function, and may greatly improve the pre-hospital prevention and treatment effects. [ABSTRACT FROM AUTHOR]

Activated carbon (AC) was treated with nitric acid, hydrogen peroxide, and urea. The obtained products were then used as supports for the preparation of palladium/activated carbon (Pd/AC) catalysts. The characterization results revealed that the AC textural structure and surface properties were selectively modified via different treatments, and the surface functional groups of the supports could affect the particle size, dispersion, and chemical states of the Pd nanoparticles. The surface of the urea-treated AC (AC-U) contained more carbonylic groups and nitrogenous species but fewer hydroxylic and carboxylic groups than the surfaces of the ACs subjected to other treatments. Consequently, in the hydrogenation synthesis of meropenem, the Pd/AC-U catalyst provided a higher meropenem yield and lower amounts of impurities in the meropenem product than the other catalysts. The nitrogenous groups not only provided efficient anchoring sites for Pd nanoparticles, improving the reaction activity, but also reduced the side reactions, decreasing the impurity content in the product. [ABSTRACT FROM AUTHOR]

This paper investigates the influence of the welding process on the microstructure and mechanical properties of Au/Sn–3.0Ag–0.5Cu/Cu solder joints. In this study, the flip-chip LED chip and the substrate were connected by a heating platform, hot air reflow, and vacuum reflow soldering methods to form Au/Sn–Ag–Cu/Cu composite solder joints. Sn–3.0Ag–0.5Cu is selected as the solder. The microstructure of the solder joint interface and the inferred surface is studied to characterize the influence of different welding methods on the intermetallic compound (IMC) interface. The void ratio of the flip-chip solder layer was tested to characterize the effect of the effective connection area between the chip and the solder joint on the shear strength. The experimental results show that the void ratio of the solder joints welded by the heating platform is as high as 26%, the IMC interface thickness is higher, and the diffusion coefficient of the solder is the lowest. The void distribution of hot air reflow soldering joints is relatively uniform, and there is less solder residue on the fractured surface. Vacuum reflow soldering has the lowest void rate in the solder joints, and the solder joints have flat cross sections with almost no solder residues. The fracture occurs at the lower layer of solder. Considering comprehensively, when the soldering method is vacuum reflow, the IMC layer of the flip-chip solder joint is the thinnest, and the diffusion coefficient of the solder is the highest. There are almost no voids in the solder layer, and the mechanical properties of the solder joint are the best. At this time, the photo-thermoelectric parameters of the filament are the best, and the performance is the most stable and reliable. [ABSTRACT FROM AUTHOR]

We develop a model of free gift card promotions in a two-product supply chain. We analyze three cases with different gift card strategies based on whether the wholesale and retail prices are changed or unchanged. The results indicate that, in the strategy of unchanged wholesale and retail prices, the retailer and two manufacturers are better off. The gift card promotions can effectively stimulate the demands of two products. We find that the two manufacturers benefit more from the gift card promotion than the retailer does. In the strategy of unchanged wholesale prices and changed retail prices, if the gift card value is not very large, the retailer's profit margins and average consumer surplus from the gift card promotion are increasing in the gift card value. In the strategy of changed wholesale and retail prices, the retailer and two manufacturers have an incentive to raise the retail and wholesale prices under certain conditions. Furthermore, we show that, keeping the prices at the level of no gift cards is the dominant strategy for the supply chain and consumers, and the difference between retail prices and wholesale prices of two products is the key factor in determining the gift card value. [ABSTRACT FROM AUTHOR]

A UVC-LED is a deep ultraviolet LED with a wavelength of 200–280 nm, which has a wide range of applications in disinfection, water purification and other fields. However, UVC-LED chips have limitations of high heat and poor heat dissipation, so it is necessary to test and analyze the thermal and optical properties of UVC-LEDs. This paper studies the photothermal characteristics of a UVC-LED under different driving currents. By measuring and analyzing the radiation flux, radiation intensity, junction temperature and thermal resistance of the UVC-LED under different driving currents, it is found that the junction temperature of the UVC-LED under the transient and steady state increases with the increase in the driving current, while the thermal resistance decreases. The radiation flux and intensity of the UVC-LED increase with the increase in driving current. The peak wavelength appears as a redshift with the increase in current, and the half-wave width also increases with the increase in current. [ABSTRACT FROM AUTHOR]

This article focuses on the study of voids in the solder joint interface of flip-chip LEDs. Additional flux is added to the solder paste to reduce the frequency of voids, and the mechanical properties of the solder joints are compared and analyzed. Two kinds of fluxes (NC559-TPF, TSF-6516) are selected. They are used in two kinds of solder pastes (lead and lead-free solder paste). The amount of flux added is 0 mL, 0.4 mL, and 0.8 mL. The other preparation conditions for all combinations are the same and are within the recommended scope of the manufacturer's documentation. X-rays are used to detect and analyze the distribution and quantity of voids in the solder layer. A push–pull tester is used to analyze and evaluate the mechanical strength of the solder joints. A scanning electron microscope was used to observe the microscopic morphology of the cross-section of the solder joint and the fracture morphology of the interface after the shear test. It was found that adding a proper amount of high-activity flux to the solder paste can reduce the frequency of voids and improve the mechanical reliability of the solder joints. Finally, we tested the thermal performance of the flip-chip LED filament for verification. [ABSTRACT FROM AUTHOR]

With the increasing number of customer reviews on the Web, there is a growing need for effective methods to retrieve valuable information hidden in these reviews, as sellers need to gain a deep understanding of customers' preferences in a timely manner. With the continuous enhancement of opinion mining or sentiment analysis research, researchers have proposed many automatic mining and classification methods. However, how to choose a trusted method is a difficult problem for companies, because customer reviews (or opinions) contain a lot of uncertain information and noise. This article reports on a detailed survey of recent opinion mining literature. It also reviews how to extract text features in opinions that may contain noise or uncertainties, how to express knowledge in opinions, and how to classify them. Through this extensive study, this paper discusses open questions and recommends future research directions for building the next generation of opinion mining systems. [ABSTRACT FROM AUTHOR]

In the actual production process, soldering is a key step in the flip-chip packaging process. The reliability of the interconnection interface of flip-chip LED chips is considered to be one of the most important reliability issues. This paper studies the influence of different soldering temperatures on the microstructure and shear behavior of the Au/Sn-3.0Ag-0.5Cu/Cu solder joints of flip-chip LED chips. The lead-free solder Sn-3.0Ag-0.5Cu was selected as the solder. The microstructure of the intermetallic compound (IMC) interface and the inferred surface of the solder joint is observed, and the microstructure evolution of the solder joint is analyzed. The void ratio of the solder joints at 250°C, 260°C, 270°C, 280°C, and 290°C soldering temperature was tested to characterize the influence of the contact area between the chip and the solder joint on the shear stress. In addition, the solder joints were aged for 1000 hours in an environment with a relative humidity of 85°C/85%, and a shear test was performed to evaluate the influence of the interface reaction on the mechanical reliability of the solder joints during the isothermal aging process. The research show that when the soldering temperature is 270°C, the Au on the bottom of the chip, and the solder and Cu on the substrate have fully reacted, and the solder joints have high shear resistance. The shear strength of the aging solder joints increases first and then decreases, because the high temperature repairs the defects in the solder layer caused by the low soldering temperature to a certain extent. With the extension of aging, cracks and voids gradually appear at the fracture interface, and the effective contact area decreases. The shear strength of the solder joint decreases, and the fracture mode becomes brittle. [ABSTRACT FROM AUTHOR]