Search

Recent advances in natural language processing (NLP) have led to the development of revolutionized pretrained language models (PLMs) impacting various NLP tasks, including sentiment analysis in software engineering. Choosing the right PLMs is crucial to effectively leverage these advanced PLMs. This paper presents the largest comparative evaluation of the PLMs for sentiment analysis in software engineering. Specifically, the study initially quantifies the performances of four traditionally fine-tuned PLMs, five zero-shot PLMs including GPT-4 and GPT-3 models, and three few-shot PLMs on six domain-specific datasets. The performances of the selected PLMs are also compared against two software engineering domain-specific traditionally fine-tuned PLMs and two state-of-the-art tools. The quantitative analysis reveals varying strengths across the different PLM types. The traditionally fine-tuned domain-specific PLM seBERT achieves the best results in the larger datasets, whereas the few-shot PLMs, such as All-DistillRoBERTa, show the best performances in the smaller datasets. A qualitative error analysis with the help of an Explainable AI technique uncovers existing challenges faced by PLMs in sentiment analysis for software engineering. The comprehensive quantitative and qualitative experiments significantly enrich knowledge in sentiment analysis in software engineering through reproducible insights.

Background When caring for COVID-19 patients, using personal protective equipment (PPE) may significantly lower the risk of infection of health care workers (HCWs). However, adverse responses due to PPE use have been observed during the 2003 SARS pandemic. This study will highlight the different adverse reactions caused by face mask use, one of the essential components of PPE in the HCWs, and identify the factors associated with these problems. Methods This cross-sectional survey was conducted between September and October 2021. 404 HCWs were selected by snowball sampling from four randomly selected healthcare facilities of Bangladesh. Trained volunteers collected data by face-to-face interview using a pretested structured questionnaire. Data were analyzed using STATA (v.16) and summarized using frequency and relative frequency. Later, the chi-square test was used to explore bivariate relationships, and the binary logistic regression model was fit to identify the predictors. Results The majority of the respondents were 26–36 years (70.30%), male (69.80%), and doctors (74.50%). 48.76% of the respondents had unfavorable skin responses beneath the face masks; female gender, physicians, professionals working more than 32 hours a week, wearing N95, and more than one mask were predictors of skin problem. 28.47% and 60.15% of all participants suffered from some form of oral and neurological problems, respectively. Conclusion Face mask use sequelae, especially skin, oral and neurological problems, are prevalent among health care workers. Therefore, necessary precautionary measures should be taken to safeguard our frontlines.

Infectious bursal disease (IBD) is a highly contagious disease that causes significant economic loss in chickens. A cross-sectional study was conducted in the Mymensingh district of Bangladesh to determine the seroprevalence of IBD virus (IBDV) antibodies in backyard chickens and their association with different epidemiological risk factors. A total of 460 serum samples were randomly collected from backyard chickens that had not been previously vaccinated against IBDV. The collected sera were examined using an enzyme-linked immunosorbent assay (ELISA). Data on epidemiological risk factors were collected through face-to-face interviews with owners and subjected to both uni- and multivariable risk analyses to determine their association with IBDV infection. Using ELISA, the overall seroprevalence of IBDV antibodies in backyard chickens was 83.4% (95% confidence interval: 79.8%–86.6%), among which, a significantly higher seroprevalence was recorded in females (83.4%, 345/350), 4–6 weeks age group (95.3%, 244/256), and unhealthy (95.0%, 57/60) backyard chickens than those of males, other age groups, and healthy chickens, respectively. Furthermore, chickens reared in free-ranging housing systems (93.3%, 280/300) and poor-conditioned houses (98.0%, 147/150) showed a significantly higher seropositivity of IBDV antibodies than those reared in separated housing systems and other hygienic-conditioned houses, respectively. Moreover, compared with their counterparts, a higher but nonsignificant seroprevalence of IBDV antibodies was observed in backyard chickens that were selected from Fulbaria Upazila (88.8%; 80/90) and which were brought from the marketplace (85.7%, 60/70). A higher seropositivity of IBDV antibodies was shown to be statistically associated with various critical epidemiological risk factors, indicating that field strains of IBDV were exposed in backyard chickens and could be readily transferred horizontally. Proper prevention and control methods, villagers’ awareness of IBD, and the rapid and widespread use of seroepidemiological investigations could help to reduce the spread of IBDV infection in backyard chickens.

BackgroundWhen caring for COVID-19 patients, using personal protective equipment (PPE) may significantly lower the risk of infection of health care workers (HCWs). However, adverse responses due to PPE use have been observed during the 2003 SARS pandemic. This study will highlight the different adverse reactions caused by face mask use, one of the essential components of PPE in the HCWs, and identify the factors associated with these problems.MethodsThis cross-sectional survey was conducted between September and October 2021. 404 HCWs were selected by snowball sampling from four randomly selected healthcare facilities of Bangladesh. Trained volunteers collected data by face-to-face interview using a pretested structured questionnaire. Data were analyzed using STATA (v.16) and summarized using frequency and relative frequency. Later, the chi-square test was used to explore bivariate relationships, and the binary logistic regression model was fit to identify the predictors.ResultsThe majority of the respondents were 26-36 years (70.30%), male (69.80%), and doctors (74.50%). 48.76% of the respondents had unfavorable skin responses beneath the face masks; female gender, physicians, professionals working more than 32 hours a week, wearing N95, and more than one mask were predictors of skin problem. 28.47% and 60.15% of all participants suffered from some form of oral and neurological problems, respectively.ConclusionFace mask use sequelae, especially skin, oral and neurological problems, are prevalent among health care workers. Therefore, necessary precautionary measures should be taken to safeguard our frontlines.

Abstract Background Alzheimer’s disease (AD), one of the major causes of dementia, is an overwhelming neurodegenerative disease that particularly affects the brain, leading to memory loss and impairment of language and judgment capacity. The aim of the present study was to investigate the antioxidant and anticholinesterase properties of the leaves of Elatostema papillosum (EPL) and correlate with their phytochemical profiles, which are relevant to the treatment of AD. Methods The dried coarse powder of EPL was extracted with 80% methanol (EPL-M80) by cold extraction method. The resultant EPL-M80 was assessed for acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) inhibitory activity by the Ellman method. The antioxidant activity was determined by DPPH (1, 1-diphenyl-2-picrylhydrazyl) and hydroxyl radical scavenging assays. Quantitative phytochemical (phenolic and flavonoid contents) analysis of endogenous substances in EPL-M80 was performed by standard spectrophotometric methods. Results EPL-M80 significantly (p < 0.05) inhibited AChE and BChE activity with IC50 of 165.40 ± 4.01 and 213.81 ± 3.57 μg/mL, respectively in a dose-dependent manner. Additionally, EPL-M80 exhibited strong radical scavenging activity against DPPH (IC50 = 32.35 ± 0.68 μg/mL) and hydroxyl radical (IC50 = 19.67 ± 1.42 μg/mL) when compared to that of standards. EPL-M80 was found to be rich in phenolic (23.74 mg gallic acid equivalent/g of dry extract) and flavonoid (31.18 mg quercetin equivalent/g of dry extract) content. Furthermore, a positive correlation (p < 0.001) was observed between the total phenolics and antioxidant as well as the anticholinesterase potential. Conclusions The marked inhibition of AChE and BChE, and potent antioxidant activity of the leaves of Elatostema papillosum highlight its potential to provide an effective treatment for AD.

Anisomeles indica (L.) kuntze is widely used in folk medicine against various disorders including allergy, sores, inflammation, and fever. This research investigated the antinociceptive, anxiolytic and sedative effects of A. indica methanol extract. The antinociceptive activity was assessed with the acetic acid-induced writhing test and formalin-induced flicking test while sedative effects with open field and hole cross tests and anxiolytic effects with elevated plus maze (EPM) and thiopental-induced sleeping time tests were assayed. Computer aided (pass prediction, docking) analyses were undertaken to find out the best-fit phytoconstituent of total 14 isolated compounds of this plant for aforesaid effects. Acetic acid treated mice taking different concentrations of extract (50, 100, and 200 mg/kg, intraperitoneal) displayed reduced the writhing number. In the formalin-induced test, extract minimized the paw licking time of mice during the first phase and the second phase significantly. The open field and hole-cross tests were noticed with a dose-dependent reduction of locomotor activity. The EPM test demonstrated an increase of time spent percentage in open arms. Methanol extract potentiated the effect of thiopental-induced hypnosis in lesser extent comparing with Diazepam. The results may account for the use of A. indica as an alternative treatment of antinociception and neuropharmacological abnormalities with further intensive studies. The compound, 3,4-dihydroxybenzoic acid was found to be most effective in computer aided models.